JPH08213301A - Charged particle beam exposure method and apparatus - Google Patents

Abstract

(57) [Summary] [Purpose] A blanking aperture array is used to gently change the transfer pattern of the connecting portion between adjacent regions. Configuration: Bit map data for one row of a blanking aperture array 30 is output from a buffer memory 41 and supplied to one input terminal of AND gates 431 to 43n. Modulation data is supplied from the cyclic shift registers 461 to 46n to the other input ends of the AND gates 431 to 43n. In the non-modulation part of the band-shaped area, the modulation data are all "1" and the cyclic shift registers 461-
46n is not shifted. In the modulation unit in the band-shaped area, the cyclic shift registers 461 to 46n are shifted, and the output of the cyclic shift registers 461 to 46n causes the output of the buffer memory 41 to be AND gate 431.
Modulated at ~ 43n. In the non-modulation parts on one end side and the other end side of the strip-shaped region, the cyclic shift registers 461 to 46n
Shift directions are opposite to each other.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a charged particle beam exposure method and apparatus for performing exposure by irradiating a resist on an object to be exposed such as a semiconductor wafer or a mask with a charged particle beam.

[0002]

2. Description of the Related Art With the miniaturization of semiconductor integrated circuit devices, charged particle beam exposure apparatuses have been used. With this apparatus, it is possible to perform fine processing of 0.05 μm or less with alignment accuracy of 0.02 μm or less. In the conventional method of exposing one point with one shot, as shown in FIG. 12A, the stage is moved stepwise by exposing the field 1 which is the electron beam deflection range, and then the stage 2 is exposed. Due to an error, a deflection error, a temperature drift of an analog circuit, a drift of an electron beam, and the like, a deviation occurs in a connection portion between the wiring pattern W1 in the field 1 and the wiring pattern W2 in the field 2.

As shown in FIG. 13 (A1), when the field 1 and the field 2 are displaced in the direction of the boundary line between them, the transfer pattern is as shown in FIG. 13 (A2). There is a risk of short circuits between them. FIG.
As shown in (B1), when the field 1 and the field 2 are displaced in a direction in which they approach each other at a right angle to the boundary line between them, the transfer pattern becomes as shown in FIG. 13 (B2) due to the proximity effect. May short circuit. FIG.
As shown in (C1), when the field 1 and the field 2 are displaced in the direction perpendicular to the boundary line between them, the transfer pattern becomes as shown in FIG. 13 (C2), and the wiring is broken at the connecting portion. This problem becomes more serious as the transfer pattern becomes finer.

Therefore, the range of fields 1 and 2 is expanded to the outside by the length L, and the dose amount of this expanded part is shown in FIG.
As shown in FIG. 12B, a method has been proposed in which the transfer pattern is gradually changed outward so that the transfer pattern is gradually changed at the connecting portion (Japanese Patent Application Laid-Open No. HEI 2). -265236). According to this method, transfer patterns as shown in FIGS. 12 (A3), (B3), and (C3) are obtained instead of the transfer patterns shown in FIGS. 12 (A2), (B2), and (C2). The transfer pattern of No. 1 changes gently to solve the problems of short circuit and disconnection.

[0005]

However, in the method of exposing one point with one shot, although the throughput of the exposure apparatus is low without using the above method, as shown in FIG. In the enlarged area of, the throughput must be further reduced because exposure must be performed step by step while decreasing the dose amount stepwise as shown by the bar graph.

On the other hand, a blanking aperture array in which openings are formed in a grid pattern on the substrate and a pair of electrodes are formed on the edges of each opening of the substrate is arranged in the optical path of the electron beam, and based on the bit map data of the pattern. A method has been proposed in which a voltage is applied to an electrode array and whether or not an electron beam that has passed through an opening is irradiated onto an exposure object depending on whether or not a voltage is applied between a pair of electrodes. According to this method, it is expected to expose a fine pattern at a high speed of about 1 cm t2 t / sec.

This exposure method is described in the above-mentioned JP-A-2-265.
Since the method of Japanese Patent No. 236 cannot be applied, in an exposure method and apparatus using a blanking aperture array,
It is required to develop a method and apparatus for gradually changing the transfer pattern of the connecting portion. In view of such a point, an object of the present invention is to provide a charged particle beam exposure method and apparatus using a blanking aperture array capable of gently changing a transfer pattern of a connecting portion of adjacent regions. It is in.

[0008]

In the first invention, a blanking aperture array in which a plurality of openings are formed in a grid in a substrate and a pair of electrodes are formed at the edge of each opening of the substrate is provided. Whether to place the charged particle beam in the optical path of the charged particle beam and irradiate the charged particle beam passing through the opening onto the object to be exposed depending on whether or not a voltage is applied between the pair of electrodes based on the bit map data of the pattern In the charged particle beam exposure method for controlling whether the charged particle beam is exposed, one end of a first band-shaped region formed by scanning the charged particle beam in a first linear direction by a deflector Formed by scanning in the first straight line direction so as to overlap with one end portion of the second strip-shaped region adjacent in the first straight line direction,
In addition, the bit map data corresponding to both ends are made to be the same, and the bit map data corresponding to the one end of the first band-shaped area is made valid / invalid for each bit of the bit map data. Modulate with the first modulation data to enable the bit map data corresponding to the one end of the second strip-shaped area, and enable / disable each bit of the bit map data.
Modulating with the second modulation data to be invalidated so that the corresponding bits of the first modulation data and the second modulation data become substantially complementary data, and the first modulation data and the second modulation data
The total value of the values of the bits corresponding to the line of the modulation data that is perpendicular to the linear direction is increased or decreased along the direction corresponding to the linear direction.

According to the first aspect of the present invention, in the exposure method using the blanking aperture array having a high throughput, it is possible to gradually change the transfer pattern of the connecting portion of the adjacent regions, and the wiring short circuit or disconnection occurs. To be prevented. Further, since the ratio of the length of the modulation region to the length of the strip region is relatively small and the scanning speeds of the non-modulation region and the modulation region in the strip region are substantially the same, the throughput of the exposure apparatus decreases. To be prevented.

Further, since it is only necessary to modulate the bit map data of the pattern in the modulation area, the processing of the connecting portion is simple. In a first aspect of the first aspect of the invention, a plurality of sets of the bitmap data are provided for the same pattern in order to change the number of times the charged particle beam is irradiated to substantially the same point on the exposure object. For each of the plurality of sets of the bitmap data, the first modulation data and the second modulation data
It has modulation data.

According to the first aspect, the pattern of the connecting portion can be changed more gently, which is preferable in preventing a short circuit or disconnection of the wiring. In the second aspect of the first invention, the first modulated data and the second modulated data are obtained by sequentially reading the same data in opposite directions. According to the second aspect, the first modulated data and the second modulated data can be obtained using the same data, so that the configuration is simplified.

In a third aspect of the first aspect of the invention, while the charged particle beam is continuously deflected by the electromagnetic main deflector in the first linear direction, the movable stage on which the exposure object is mounted is mounted on the first stage. By continuously moving in a second straight line direction perpendicular to the straight line direction, and deflecting the charged particle beam in the second straight line direction by following the movement of the moving stage by an electrostatic sub-deflector, The band-shaped area is formed.

According to the third aspect, since the length of the strip-shaped region is equal to the deflection range of the electromagnetic main deflector and is relatively long, for example, 2 mm, the ratio of the length of the modulation region to the length of the strip-shaped region is large. As a result, the throughput of the exposure apparatus is further prevented from decreasing. The following second invention and the first to third thereof
Aspects correspond to the above first invention and the first to third aspects thereof, respectively.

In the second aspect of the invention, a blanking aperture array in which a plurality of openings are formed in a grid in a substrate and a pair of electrodes are formed at the edge of each opening of the substrate is arranged in the optical path of the charged particle beam. , It is controlled based on the bit map data of the pattern whether to apply a voltage between the pair of electrodes or not to irradiate the charged particle beam that has passed through the opening onto the exposure target. In the charged particle beam exposure apparatus that scans the charged particle beam, one end of a first band-shaped region formed by scanning the charged particle beam in a first linear direction causes the deflector to move the charged particle beam to the first linear region. Deflector drive means for supplying a drive signal to the deflector so as to overlap one end of a second strip-shaped region formed by scanning in one linear direction and adjacent to the one direction, and the one end of the first strip-shaped region. Part and said Bit map data output means for outputting the bit map data so that the bit map data corresponding to the one end portion of the two strip-shaped areas are the same, and the bit map data output from the bit map data output means. With respect to the first strip area, the bitmap data corresponding to the one end portion of the first strip area is modulated with the first modulation data that enables / disables each bit of the bitmap data, and the one end portion of the second strip area is modulated. Modulation means for modulating the bitmap data corresponding to the second bitmap data for validating / invalidating each bit of the bitmap data,
Blanking aperture array driving means for applying a voltage to the electrode of the blanking aperture array based on the bitmap data that is modulated or not modulated by the modulating means, and the first modulation data and the The bit corresponding to the second modulation data is substantially complementary data, and the sum of the values of the bits corresponding to the first modulation data and the second modulation data on the line in the direction perpendicular to the linear direction. The value is substantially increased or decreased along the direction corresponding to the linear direction.

In the first aspect of the second aspect of the invention, the bit map data output means sets the number of times the charged particle beam is irradiated to substantially the same point on the object to be exposed, in order to change the number of times for the same pattern. A plurality of sets of the bitmap data are output, and the modulation means has the first modulation data and the second modulation data for each of the plurality of sets of the bitmap data.

In the second aspect of the second aspect of the invention, the modulating means obtains the first modulated data by shifting the data in one direction and sequentially reading the data, and shifts the data in the opposite direction to the one direction. It has a plurality of circular shift registers having the same bit length with each other, which sequentially read out to obtain the second modulated data. In a third aspect of the second invention, the deflector includes an electromagnetic main deflector and an electrostatic sub-deflector, and further, a moving stage on which the exposure target is mounted, and the electromagnetic main deflector. While continuously deflecting the charged particle beam in the first linear direction by a device, the moving stage is continuously moved in a second direction orthogonal to the first linear direction, and the electrostatic sub-deflection is performed. And a deflection control means for deflecting the charged particle beam in the second linear direction by following the movement of the moving stage.

[0017]

Embodiments of the present invention will be described below with reference to the drawings. [First Embodiment] FIG. 1 shows a main part of a charged particle beam exposure apparatus according to the first embodiment. 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 detected by the laser interferometer length measuring device 16 and fed back to the stage control circuit 14. A resist film is deposited on the semiconductor wafer 10, and the semiconductor wafer 10 is exposed by irradiating it with an electron beam EB2 that passes through the circular opening of 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. Scanning is performed as shown in FIG.

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. Further, the simultaneous exposure area 10
Electron beam EB2 so that band 102, which is a region obtained by one main scan of 1 in the direction X on semiconductor wafer 10.
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
High speed of m / s. The electron beam EB2 is scanned by the main deflector 20 and the sub-deflector 22 along the dotted line of the portion enlarged and illustrated on the outside of the semiconductor wafer 10, and as shown by the arrow on the dotted line, the main scanning direction is changed for each band. The opposite is true. The moving stage 12 is scanned in the Y direction along the dotted line shown on the semiconductor wafer 10, and the frame 103 is formed by the one scanning. As indicated by the dotted arrow,
The scanning direction of the moving stage 12 is reversed every frame. The band 102 has modulation regions 105 and 1 having the same length in the X direction at both ends of the non-modulation region 104.
06, and in the ideal case the modulation regions 105 and 10
Each 6 exactly overlaps the modulation area of the adjacent frame.

In FIG. 1, the main control circuit 24 supplies the target position to the stage control circuit 14, supplies a periodic sawtooth wave signal to the amplifier circuit 26, and receives the stage detection position from the laser interferometer length measuring device 16. , A signal proportional to the sub-deflection distance is supplied to the amplifier 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. As shown in FIG. 2, the blanking aperture array plate 30 has openings 33 formed in a zigzag pattern in a region 32 of a thin substrate 31. For each opening 33, the substrate 31
A pair of common electrodes 34 and a blanking electrode 35 are formed on the upper side, and the common electrodes 34 are connected to the ground line.
FIG. 2 shows the upper surface of the blanking aperture array plate 30, and the lower surface of the blanking aperture array plate 30 has a pair of deflection electrodes which are electrically connected to the common electrode 34 and the blanking electrode 35 as shown in FIG. It projects downward from the edge of 33.

An electron beam EB0 having a substantially uniform current density is projected onto a region 32 on the substrate 31. The electron beam EB2 passing through the opening 33 changes the potential of the blanking electrode 35 to 0.
When it is set to V, it passes through the opening of the aperture plate 18 as shown in FIG. 1, but when a constant potential Vs is applied to the blanking electrode 35, it is deflected and blocked by the aperture plate 18.
Therefore, a desired fine pattern can be exposed on the semiconductor wafer 10 by applying a potential of, for example, 0 / Vs to the blanking electrode 35 corresponding to the exposure dot data of which the bit is 1/0.

For example, one opening 33 has a side of 25 μm.
The area exposed on the semiconductor wafer 10 through the opening 33 is a substantially square with one side of 0.08 μm. The Y direction is referred to as the longitudinal direction of the rows of the openings 33, and the openings 33
The apparent two lines of is one line. Although the opening 33 has 3 rows and 20 columns for simplification in FIG. 2, it is actually 8 rows and 128 columns, for example. The opening 33 has m rows and n columns, and the opening 33 and the blanking electrode 35 on the i-th row and the j-th column are the opening 33 (i, j) and the blanking electrode 35, respectively.
Represented as (i, j). The pitch p of the openings 33 in the X direction is, for example, three times the length a of one side of the openings 33 in order to secure a region for electrodes and wiring.

The BAA control circuit 40 includes a buffer memory 41 to which the bit map data of the pattern is supplied and written from the main control circuit 24 of FIG. The buffer memory 41 is divided into, for example, two areas, and the bitmap data is written in one area by the direct memory access method, and at the same time, the bitmap data is read from the other area to read one frame and Each time writing is completed, both areas are switched between the reading area and the writing area. The buffer memory 41 synchronizes with the clock from the AND gate 42,
The n-bit parallel data corresponding to the first to nth columns of the blanking electrode 35 is output.

The AND gate 42 is supplied with the clock φ and the scanning area signal SA as shown in FIG. The scanning area signal SA becomes “1” during the period of scanning the band 102 in FIG. 5, and becomes “0” during the other periods. The n-bit output of the buffer memory 41 is the AND gate 4 respectively.
It is supplied to one input end of 31-432. The outputs of the AND gates 431 to 432 are respectively output to the shift register 4
It is supplied to the data input terminals of the least significant bits 41 to 44n. The shift registers 441 to 44n are shifted by 1 bit to the upper side with each pulse of the clock φ.

If j is an odd number, shift register 44j
Is {2 (p / a) (n-1) +1} bits, and its m bits are supplied to the driver 45, whereby the potentials of the blanking electrodes 35 (i, j), i = 1 to m. Is determined. In detail, the second {2 (p /
a) When the output of (i-1) +1} bits is "1", the potential of the blanking electrode 35 (i, j) becomes 0 V, and the electron beam passing through the opening 33 (i, j) is a semiconductor wafer. When the output is "0", the potential is Vs.
Then, the electron beam passing through the opening 33 (i, j) is blocked by the aperture plate 18 and is not irradiated onto the semiconductor wafer 10. However, the least significant bit of the shift register 44j is the first bit.

If j is an even number, shift register 44j
Is {(p / a) (2n-1) +1} bits, and the m bits thereof are supplied to the driver 45, whereby the potentials of the blanking electrodes 35 (i, j), i = 1 to m. Determined. Specifically, the shift register 44j has the {(p /
a) When the output of (2i-1) +1} bits is "1", the potential of the blanking electrode 35 (i, j) becomes 0V,
The electron beam passing through the opening 33 (i, j) is irradiated onto the semiconductor wafer 10, and when the output is "0", the potential is V
At s, the electron beam passing through the opening 33 (i, j) is blocked by the aperture plate 18 and is not irradiated onto the semiconductor wafer 10.

FIG. 3 shows a part of the above relationship when p / a = 3 and n = 3. The switch in the driver 45 is a non-contact changeover switch, and the changeover is controlled by the output of the shift register. FIG. 6 schematically shows scanning of the electron beam in the X direction. In reality, reduction projection exposure is performed, but in FIG. 6, the reduction ratio of projection is set to 1 for simplification of description.

As for the scanning speed of the electron beam, assuming that the period of the clock φ is T and the electron beam passing through the opening 33 (1, j) at the time t = 0 irradiates the point P on the semiconductor wafer 10, the time t is obtained. = 2 (p / a) T, 4 (p / a) T, ...
..Aperture 33 at 2 (m-1) (p / a) T
(2, j), opening 33 (3, j), ..., opening 33
The electron beam passing through (m, j) is adjusted so as to irradiate the same point P on the semiconductor wafer 10.

As a result, on the semiconductor wafer 10,
The same point is exposed m times with the same data. Also, the opening 33
(I, j), j = 1, 3, 5, ..., N−1, and the openings 33 (i,
j), j = 2, 4, 6, ..., N through time t +
It is exposed at (p / a) T. In FIG. 2, the other input ends of the AND gates 431 to 43n are connected to the cyclic shift registers 461 to 46 of the cyclic shift register group 46, respectively.
The bit output at one end of n is provided. The number of bits of each of the cyclic shift registers 461 to 46n is equal to each other and is (the number of X-direction exposure dots of the modulation area 105) +1. The cyclic shift registers 461 to 46n are shifted by the clock from the AND gate 47, and the shift direction is as shown in FIG.
The shift direction signal DR as shown in FIG. The AND gate 47 is supplied with the clock φ and the shift period signal ST as shown in FIG. Shift period signal ST
After scanning the modulation area 105 or 106 in FIG.
It becomes "1" in the period from the end of scanning to the lapse of one cycle T of the clock φ, and becomes "0" in other periods. As the shift direction signal DR, the shift period signal ST is set to 1
It is a signal divided by two.

Circular shift registers 461 to 46 when n = 9
Circular shift register 4 when 9 is 9 bits each
An example of the modulation data initially set to 61 to 469 is shown in FIG.
It shows in (A). A white square indicates a bit "1", and a shaded square indicates a bit "0". In the initial setting state, (a) the number of '1's of the i-th bit of the cyclic shift registers 461 to 469, i = 0 to 8, is i,
(B) For i = 1 to 4, the i-th bit and the (9-i) -th bit
The bit and the value are complementary to each other.

In the non-modulation area 104 in FIG. 5, since the shift period signal ST is "0", the AND gate 4
7 is closed and the cyclic shift register group 46 is not shifted, and the outputs thereof are all "1". As a result, all the AND gates 431 to 43n are opened and the buffer memory 4
The output of 1 is not modulated. The shift period signal ST is “1” and the shift direction signal DR is at the left end of the modulation region 106 in FIG.
Becomes "0", the AND gate 47 is opened, the cyclic shift register group 46 is shifted in the direction FW in synchronization with the clock φ, and the output of the buffer memory 41 is modulated by the outputs of the cyclic shift registers 461 to 46n. . After the scanning area signal SA becomes "0" at the right end (end of M2 in FIG. 4) of the modulation area 106 in FIG. 5, the cyclic shift register group 46 is shifted once more and the cyclic shift register 461.
After all the outputs of ~ 46n become "1", the shift period signal ST becomes "0" and the shift of the cyclic shift register group 46 is stopped.

The scanning area signal SA, the shift period signal ST and the shift direction signal DR become "1" at the time when the scanning of the next band 102 is started (starting end of M1 in FIG. 4) by folding back.
The AND gate 47 is opened and the cyclic shift register group 46 is opened.
Is shifted in the opposite direction FB in synchronization with the clock φ, and the outputs of the cyclic shift registers 461 to 46n modulate the output of the buffer memory 41. Modulation area 106
The circular shift register group 46 is shifted once again by the clock immediately after the left end of the circular shift registers 461 to 46.
After all the outputs of n become "1", the shift period signal S
T becomes "0", the shift of the cyclic shift register group 46 is stopped, and the non-modulation area 104 is scanned.

The scanning of the modulation area 105 is the same as the scanning of the modulation area 106. Therefore, the modulation data output from the cyclic shift register group 46 corresponding to one band is as shown in FIG. 7 (B). FIG.
(A) shows bitmap data in the case where the wiring pattern indicated by diagonal lines extends in the band 102A and in the band 102B adjacent in the longitudinal direction thereof. A white square indicates a bit "0", and a shaded square indicates a bit "1". In this case, the modulated data at the right end of the band 102A and the left end of the band 102B are as shown in FIGS. 8B and 8C, respectively.

In an ideal case where the connecting portion between the band 102A and the band 102B is not displaced, the band 102
Since the right side modulation area of A and the left side modulation area of band 102B exactly overlap each other, the logical sum of the corresponding bits of both matches the bitmap data of FIG. 8 (A). However, in practice, the band 102A and the band 10A are affected by stage movement error, deflection error, temperature drift of the analog circuit, electron beam drift, and the like.
There is a gap in the connection with 2B. 9 (A) to (C)
Are the wiring pattern data when the connection deviations occur in the direction perpendicular to the longitudinal direction of the bands, in the longitudinal direction of the bands and between the bands, and in the longitudinal direction of the bands and the direction between the bands. Indicates. The transfer pattern in this case is shown in FIG.
3), (B3) and (C3),
Short circuit and disconnection of wiring are prevented.

Further, for example, the length of the band 102 is 2 mm
On the other hand, the sum of the lengths of the modulation area 105 and the modulation area 106 may be about 1.6 μm, and the scanning speeds of the non-modulation area 104 and the modulation areas 105 and 106 are the same.
A decrease in throughput of the exposure apparatus is prevented. Further, in the modulation areas 105 and 106, the cyclic shift register group 4
Since it is only necessary to modulate the data using 6, the processing of the connecting portion is simple.

[Second Embodiment] FIG. 8 shows the structure of a second embodiment corresponding to FIG. In the above-described first embodiment, the case where the same point on the semiconductor wafer 10 is repeatedly irradiated with the electron beam m times or not even once with the same bit data has been described, but by changing the number of times of repeated irradiation, Regardless of the connecting portion, it is possible to perform dimensional adjustment of the edge portion of an arbitrary pattern and highly accurate proximity effect correction.

Therefore, in the second embodiment, the BAA of FIG.
BAA control circuits 40A and 40 having the same configuration as the control circuit 40
B and 40C are used to supply these outputs to the blanking aperture array plate 30 via the selector 50. The cyclic shift register groups 46A, 46B and 46C of the BAA control circuits 40A, 40B and 40C respectively correspond to FIG. 7A, for example, as shown in FIGS.
Data shown in (C) and (C) are initialized.
All of these three initial setting data satisfy the above conditions (a) and (b).

For example, at time t = 0, the bit data modulated by the bit A1 of the cyclic shift register group 46A is selected by the selector 50 to determine the potential of the blanking electrode 35 (1,1) shown in FIG. And time t = 2
At (p / a) T, the cyclic shift register group 46B
, The bit data modulated by the bit B1 corresponding to the bit A1 is selected by the selector 50 to determine the potential of the blanking electrode 35 (2,1), and the time point 4 (p / a)
At T, bit A of the cyclic shift register group 46C
The bit data modulated by the bit C1 corresponding to 1 is selected by the selector 50, and the blanking electrode 35 is selected.
By determining the potential of (3, 1), the number of electron beam irradiations on the same point P on the semiconductor wafer 10 is set to any of 0 to 3 times. The same applies to other modulation data.

This is preferable because the distribution of the exposure amount in the modulation area becomes gentler than in the first embodiment, and the pattern of the connecting portion changes more gently than in the first embodiment. Further, due to the reason described in the first embodiment, the decrease in throughput of the exposure apparatus is prevented. In addition, the present invention includes various modifications.

For example, instead of the cyclic shift register group 46, a shift register group corresponding to each of the modulation regions at one end and the other end of the band may be provided and the condition (b) above may not be satisfied. Good. The data modulation may be performed in the previous stage of the buffer memory 41. In this case, the logical product of the same modulation data as the data held therein and the bitmap data of the pattern may be simultaneously calculated without using the cyclic shift register.

Further, FIG. 10 shows BAA control circuits 40A-4A.
In the case where the outputs of 0C are paralleled, one BAA control circuit 40 is used to triple the frequency of the clock φ,
It is also possible to output 3-bit data of the same exposure point in series and select this with a selector so that the result is the same as in the case of the second embodiment. Further, the present invention can be applied to the case where a charged particle beam is scanned in a linear direction by various scanning methods to form a band-shaped region, and this band-shaped region stops, for example, the operation of the main deflector and the sub-deflector. May be formed by scanning the charged particle beam in a linear direction.

In order to further increase the throughput, the modulation region is not formed in a portion where it is clear that there is no pattern extending between adjacent strip-shaped regions in the longitudinal direction, or the length of the strip-shaped region is changed. The length of the modulation area may be changed. In the claims, the first
It is limited to the preferable case where the total value of the values of the bits corresponding to the linear direction of the modulated data and the second modulated data in the direction perpendicular to the linear direction increases or decreases substantially along the direction corresponding to the linear direction. However, for example, a complicated change may be made along the direction such that it monotonically increases and then decreases, or increases and then increases again.

[0044]

As described above, according to the charged particle beam exposure method of the first invention and the charged particle beam exposure apparatus of the second invention, in the exposure method using the blanking aperture array with high throughput, It is possible to gently change the transfer pattern of the connection part of the matching area, prevent short circuit and disconnection of the wiring, and that the ratio of the length of the modulation area to the length of the strip area is relatively small, and Since the scanning speeds of the non-modulation area and the modulation area in the area are substantially the same, it is possible to prevent the throughput of the exposure apparatus from decreasing, and furthermore, in the modulation area, the bitmap data of the pattern may be simply modulated. It has an excellent effect that the processing of the part is simple.

According to the first aspect of the first and second inventions,
The pattern of the connecting portion can be changed more gently, and there is an effect that it is preferable in terms of prevention of short circuit and disconnection of wiring. According to the second aspect of the first and second aspects of the invention, the first modulated data and the second modulated data can be obtained by using the same data, so that the configuration is simplified.

According to the third aspect of the first and second inventions,
Since the length of the strip-shaped region becomes equal to the deflection range of the electromagnetic main deflector, the ratio of the length of the modulation region to the length of the strip-shaped region becomes smaller, and the decrease in throughput of the exposure apparatus can be prevented more effectively. Play.

[Brief description of drawings]

FIG. 1 is a block diagram of essential parts of a charged particle beam exposure apparatus according to a first embodiment of the present invention.

FIG. 2 is a diagram showing a configuration of a blanking aperture array and its control circuit in FIG.

FIG. 3 is a diagram showing a part of details of a blanking aperture array control circuit in FIG.

FIG. 4 is a waveform diagram of signals supplied to the blanking aperture array control circuit in FIG.

FIG. 5 is an explanatory diagram of an electron beam scanning method.

FIG. 6 is a schematic side view showing scanning with an electron beam.

7A is a diagram showing initial values of a modulation shift register, and FIG. 7B is a diagram showing modulation data of one band.

FIG. 8 is an explanatory diagram of modulation of pattern bitmap data.

FIG. 9 is a diagram showing a connection part between frames of wiring pattern data.

FIG. 10 is a block diagram showing a configuration corresponding to FIG. 2 of the second exemplary embodiment of the present invention.

11 is a diagram showing initial values of a modulation shift register used in the blanking aperture array control circuit in FIG.

FIG. 12 is an explanatory diagram of a conventional exposure method in a field connection portion.

FIG. 13 is a diagram showing an effect of the exposure method of FIG.

[Explanation of symbols]

 10 Semiconductor Wafer 20 Main Deflector 22 Sub Deflector 30 Blanking Aperture Array Plate 33 Opening 34 Common Electrode 35 Blanking Electrode 40, 40A-40C BAA Control Circuit 41 Buffer Memory 441-44n Shift Register 45 Driver 461-46n Circular Shift Register 50 selectors 102, 102A, 102B band 103 frame 104 non-modulation area 105, 106 modulation area

 ─────────────────────────────────────────────────── ─── Continuation of the front page (72) Inventor Soichiro Arai 1015 Kamiodanaka, Nakahara-ku, Kawasaki City, Kanagawa Prefecture, Fujitsu Limited (72) Inventor Kenichi Miyazawa, 1015, Kamiodanaka, Nakahara-ku, Kawasaki City, Kanagawa Prefecture, Fujitsu Limited ( 72) Inventor Hiroshi Yasuda 1015 Kamiodanaka, Nakahara-ku, Kawasaki City, Kanagawa Prefecture, Fujitsu Limited

Claims (8)

[Claims] 1. A blanking aperture array, in which a plurality of openings are formed in a grid on a substrate and a pair of electrodes is formed at the edge of each opening of the substrate, is arranged in the optical path of a charged particle beam to form a pattern. In a charged particle beam exposure method for controlling whether or not a charged particle beam that has passed through the opening is irradiated onto an exposure target object by applying or not applying a voltage between the pair of electrodes based on bitmap data, One end of a first band-shaped region formed by scanning the charged particle beam in the first linear direction by the deflector is formed by scanning the charged particle beam in the first linear direction by the deflector. In correspondence with the one end portion of the first strip-shaped area, the bitmap data corresponding to both end portions of the second strip-shaped area adjacent to each other in one straight line direction are made to be the same as each other. The bitmap data is modulated with the first modulation data for validating / invalidating each bit of the bitmap data, and the bitmap data corresponding to the one end of the second band-shaped region is converted into each of the bitmap data. Modulating the bit with the second modulation data for enabling / disabling, so that the corresponding bits of the first modulation data and the second modulation data are substantially complementary data, and the first modulation data and the Charging, characterized in that the total value of the values of the bits corresponding to the line of the second modulation data which is perpendicular to the linear direction is increased or decreased along the direction corresponding to the linear direction. Particle beam exposure method. 2. A plurality of sets of the bitmap data are provided for the same pattern in order to vary the number of times the charged particle beam is irradiated to substantially the same point on the exposure object, and the plurality of sets of the bitmap data are provided. The first for each of the bitmap data
The charged particle beam exposure method according to claim 1, further comprising modulation data and the second modulation data. 3. The charged particle beam exposure method according to claim 1, wherein the first modulation data and the second modulation data are obtained by sequentially reading the same data in opposite directions. . 4. An electromagnetic main deflector continuously deflects the charged particle beam in the first linear direction, and a movable stage on which the exposure object is mounted is moved to a second direction perpendicular to the first linear direction. The belt-shaped region is formed by continuously moving in a linear direction and deflecting the charged particle beam in the second linear direction by following the movement of the moving stage by an electrostatic sub-deflector. The charged particle beam exposure method according to any one of claims 1 to 3, wherein 5. A blanking aperture array, in which a plurality of openings are formed in a grid on a substrate and a pair of electrodes is formed at the edge of each opening of the substrate, is arranged in the optical path of a charged particle beam to form a pattern. Based on the bit map data, it is controlled whether or not the charged particle beam passing through the opening is irradiated onto the object to be exposed depending on whether or not a voltage is applied between the pair of electrodes, and the charged particle is deflected by a deflector. In a charged particle beam exposure apparatus for scanning a beam, one end of a first band-shaped region formed by scanning the charged particle beam in a first linear direction causes the deflector to direct the charged particle beam to the first linear direction. Deflector driving means for supplying a drive signal to the deflector so as to overlap with one end of a second strip-shaped region which is formed by scanning to and adjacent to the one direction, and the one end of the first strip-shaped region and Second strip area The bitmap data output means for outputting the bitmap data so that the bitmap data corresponding to the one end become the same as each other, and the bitmap data output from the bitmap data output means The bitmap data corresponding to the one end of the one strip area is modulated with the first modulation data that enables / disables each bit of the bitmap data, and the bitmap data corresponding to the one end of the second strip area is modulated. Bitmap data is valid / valid for each bit of the bitmap data.
Modulation means for modulating with the second modulation data to be invalidated, and blanking aperture array drive for applying a voltage to the electrode of the blanking aperture array based on the bitmap data that is or is not modulated by the modulation means. Means for making the corresponding bits of the first modulated data and the second modulated data substantially complementary data, and in the linear direction of the first modulated data and the second modulated data. A charged particle beam exposure apparatus, wherein a total value of bit values corresponding to a line in a direction orthogonal to is increased or decreased substantially along a direction corresponding to the linear direction. 6. The bitmap data output means sets a plurality of sets of bitmap data for the same pattern in order to vary the number of times the charged particle beam is irradiated to substantially the same point on the exposure object. 6. The charged particle beam exposure apparatus according to claim 5, wherein the modulating unit has the first modulation data and the second modulation data for each of the plurality of sets of the bitmap data. 7. The modulation means obtains the first modulated data by shifting data in one direction and sequentially reading the data, and shifts the data in a direction opposite to the one direction and sequentially reads the second data. 7. The charged particle beam exposure apparatus according to claim 5, further comprising a plurality of circular shift registers having the same bit length for obtaining modulated data. 8. The deflector has an electromagnetic main deflector and an electrostatic sub-deflector, and further comprises a moving stage on which the object to be exposed is mounted, and the electromagnetic main deflector for charging the charged object. The particle beam is the first
While continuously deflecting in a linear direction, the moving stage is continuously moved in a second direction perpendicular to the first linear direction,
9. A deflection control means for deflecting the charged particle beam in the second linear direction by following the movement of the moving stage by the electrostatic sub-deflector. Either one
Charged particle beam exposure apparatus.