JP2000323401A - Charged particle beam aligner and exposing method - Google Patents

Abstract

PROBLEM TO BE SOLVED: To suppress the maximum total current amount of a charged particle beam by thinning out a part of the number of times of exposure in pixels in an exposing region, making the number different about a part of pixels, and setting the exposure value of pixels to be exposed in a specified range. SOLUTION: The upper limit of on-bit number, and the upper limit of changing amount of on-bit number are decided (201). Data of cell stripe size are read from a disk to read a work area of a unit, and a bit map is formed (202). Contents of cell stripe size to which previously formed change candidate pattern data correspond are read (203). On-bit number at the respective positions of the cell stripe is calculated, and its changing amount is calculated (204). It is judged whether the calculated on-bit number and its changing amount are equal to or lower than the upper limit value (205). By the result, data of the bit map are changed (206). Data are so stored that the exposure value of pixels to be exposed is in a specified range (207, 208).

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 apparatus such as an electron beam exposure apparatus and an exposure method.
In particular, a blanking aperture array (BAA) in which apertures having a large number of control electrodes are arranged and a beam applied to the control electrodes can be turned on / off by a signal applied to the control electrodes.
The present invention relates to a BAA type charged particle beam exposure apparatus and an exposure method for controlling a BAA in synchronization with scanning of a charged particle beam.

Semiconductor integrated circuits tend to be more highly integrated with advances in microfabrication technology, and the performance required for microfabrication technology is becoming increasingly severe. In particular, in the case of the exposure technology, the limit of the light exposure technology used for the steppers and the like conventionally used is expected.
Electron beam exposure technology is a technology that is likely to be the next generation of fine processing in place of light exposure technology. Hereinafter, description will be made by taking electron beam exposure as an example.

[0003]

2. Description of the Related Art An electron beam exposure system is one in which a sample is irradiated with an accelerated and shaped electron beam to obtain an exposure pattern, and is characterized by having a higher resolution than a light exposure technology such as a stepper. For example, in electron beam exposure, fine processing of 0.05 μm or
It can be realized with the following alignment accuracy. However, electron beam exposure has been considered to be unusable for mass production of LSIs because of its low throughput. However, this method uses a single-stroke writing method, scans a sub-deflection range or a desired range with a fine electron beam, and responds to patterns. The discussion was about a raster scan method that controls the electron beam on and off.

In recent years, a block exposure method has been proposed to improve the throughput in electron beam exposure. The block exposure method is a method of exposing a pattern by dividing an exposure pattern into basic blocks and exposing an electron beam formed by passing through apertures (apertures) corresponding to each block. There is a problem that cannot be drawn. In order to solve such a problem, a blanking aperture array (BA
A) A method has been proposed. A detailed configuration of a BAA type electron beam exposure apparatus is disclosed in Japanese Patent Application Laid-Open No. Hei 5-166707.

FIG. 1 is a diagram showing an example of a blanking aperture array (BAA) used in the BAA system. The BAA 10 has a large number of apertures 11 having change electrodes 12 and electrodes 13 for ground potential on both sides arranged as shown in the figure, and a uniform electron beam is incident thereon. When a voltage is not applied to the electrode 12, the electron beam passes through the opening 11 and passes as it is to irradiate the sample with the electron beam. However, when a voltage is applied to the electrode 12, the direction of the beam is bent, so that the electron beam is shielded. If not, the sample is not irradiated. That is, depending on whether or not a voltage is applied to the electrode 12, the opening 1
1 can be turned on and off. In addition,
In this example, the electrode 13 is always at the ground potential, but a different potential (for example, a positive or negative potential) is applied between the electrodes 12 and 13 when the deflection is performed, and the same potential (for example, the ground potential) is applied when the deflection is not performed. You may.

FIG. 2 is a diagram showing a basic configuration of exposure by the BAA method. In the exposure in the BAA method, the deflection of the electron beam corresponding to the scan and the on / off pattern of the BAA 10 are changed in synchronization. As shown in FIG. 2A, charged particles (electrons) emitted from the electron gun 51
The beam passes through a group of apertures in the BAA 11 and is shaped into a fine beam bundle 56. The beam is scanned over the sample 100 in the direction of arrow F by a deflecting unit 53 composed of an electromagnetic deflector or an electrostatic deflector. . B1 to B3 indicate electron beam fluxes that move on the sample 100. As shown in FIG. 1, a plurality of apertures 31 are two-dimensionally arranged in the BAA 10, and the electron beam is split into a plurality of fine beams by passing through the BAA 10. As described above, depending on whether or not a voltage is applied between the electrodes 12 and 13 provided on both sides of each opening 11, the electron beam passing through each opening is turned on.
Can be turned off. Here, the case where the fine beam that has passed through the opening without applying a voltage between the electrodes 12 and 13 is irradiated on the sample 100 is turned on, and a voltage is applied between the electrodes 12 and 13 to open the sample. The fine beam that has passed through the hole
It is turned off when the light is not irradiated upward.

Now, a pattern as shown in (2) of FIG. 2 is formed by three columns (R1, R2, R3) as shown in FIG.
A case where exposure is performed using nine BAAs 10 in three rows (L1, L2, L3) will be described. As shown in (1) of FIG. 3, the beam passing through the BAA 10 is deflected so as to be irradiated to the position B1. At this time, assuming that only the apertures (L1, R2) are on and the other apertures are off, the aperture (L
1, R2), the beam on P1 (L
(1, R2). Next, the scanning proceeds a little, and as shown in (2) of FIG. 3, the beam that has passed through the BAA 10 is deflected so as to irradiate the position B2. Here, the openings (L2, R2), (L1, R1),
Assuming that only (L1, R3) is on and the other apertures are off, the apertures (L2, R2), (L1, R1), (L1,
Only the beam that has passed through R3) is P1 (L
2, R2), P2 (L1, R1), P3 (L1, R3)
Is irradiated to the position. Here, it is assumed that the deflection amount in (1) and (2) changes so that P1 (L2, R2) is at the same position as P1 (L1, R2). The scanning further proceeds, and as shown in (3) of FIG. 3, the beam that has passed through the BAA 10 is deflected so as to irradiate the position B3. Here, the openings (L3, R2), (L2, R1),
(L2, R3), (L1, R1), (L1, R2),
Assuming that (L1, R3) is on and the other apertures are off, only the beams passing through these apertures are P1 (L3, R2), P2 (L2, R1), P3 (L
2, R3), P4 (L1, R1), P5 (L1, R1)
2), irradiation is performed at the position of P4 (L1, R3). Here, the deflection amount in (2) and (3) is P1 (L3, R2)
Is P1 (L2, R2) and P2 (L2, R1) is P2
(L1, R1) and P3 (L2, R3) are P3 (L3,
It is assumed that the position has changed so as to be at the same position as R3).

Next, although not shown, the pattern of the column L2 in FIG. 3 (3) is set to L1 and the pattern of the column L1 is set to L2.
Then, a new pattern is applied to L1, and at the same time, beam scanning is advanced. By performing such an operation,
An arbitrary pattern can be exposed with the length of the scanning range and the width of the row of apertures of the BAA 30. At this time, exposure for the number of rows of apertures is performed for each point. As shown in FIG. 1, the apertures 31 are arranged in a staggered manner, and by scanning in the vertical direction in FIG. 1, the space between spots in each row is filled so that uniform exposure can be performed. I have.

As described above, in the BAA method, since the scan signal only needs to be changed continuously, there is no beam settling time required in the case of vector scanning, high-speed scanning is possible, and exposure is performed with high efficiency. With the advantage of being able to,
There is an advantage over the block exposure method that any pattern can be freely exposed. Further, since the same spot is exposed a plurality of times, the exposure energy is high, and the drawing speed can be increased by the number of columns compared to the one-stroke writing method. In the case of a single-row BAA, the drawing speed can be similarly increased by the number of rows. However, in the case of a single-beam writing or a single-row BAA, the full aperture changes from the on state to the off state. However, such a change causes a defocus phenomenon in which the focus position of the electron beam is shifted. When such a phenomenon occurs, the quality of the exposure pattern is degraded, so it is necessary to correct the focus position. , The focus position could be adjusted. However, the change is so rapid that the correction delay is inevitable,
This limits the scanning speed and hinders the improvement of the drawing speed. On the other hand, in the BAA exposure method, the ratio of the number of holes that change from the off state to the on state to the total number of holes is reduced to approximately 1 / the number of rows, so that the change in the correction amount of the focus position is small. And the scanning speed can be improved.

In an electron beam exposure apparatus, an electromagnetic deflector for generating a magnetic field using a coil and an electrostatic deflector for generating an electrostatic field are used to deflect an electron beam. The electromagnetic deflector can deflect in a relatively large range as compared with the electrostatic deflector, but has a slow response speed. Therefore, in order to perform scanning efficiently, a deflecting unit is realized by combining deflectors having different characteristics. More specifically, the electromagnetic deflector is used as a main deflector, and the electrostatic deflector is used as a sub-deflector. After fixing the deflection position of the main deflector, the sub-deflector deflects and scans within the deflection range. When the scanning in this range is completed, the same scanning is performed by changing the deflection position of the main deflector, and the scanning is performed so as to cover the entire exposure range.
Actually, scanning is performed efficiently by combining movement of the stage.

FIGS. 4 and 5 show examples of scanning in the BAA exposure method. In FIG. 5, reference numeral 105 denotes a width that can be exposed in one scan of the fine beam bundle. The stage 101 moves the wafer 101 in one direction (Y direction) by a stage, and deflects the fine beam bundle of width 105 in that direction (Y direction) by a sub-deflector to form a straight line of about 100 μm (actually, a slight curve due to correction). Scan the area indicated by reference numeral 106. During this scanning, an ON / OFF signal is applied to the electrodes of the BAA to draw a desired pattern. The area drawn by one scan is called a cell stripe. When the writing of one cell stripe is completed, the amount of deflection of the sub deflector in the X direction is changed, and the next adjacent cell stripe is similarly drawn. Such a cell stripe is formed by X in accordance with the deflection range of the sub deflector.
Draw about 10 in the direction. This is the area indicated by reference numeral 104. When the drawing of the area 104 is completed, the deflection position of the main deflector in the X direction is sequentially changed, and
Are drawn in the same manner as described above, and when the drawing of the width 103 is completed, the deflection position of the main deflector in the Y direction is changed, and the cell stripes are sequentially drawn in the reverse direction. Such an operation is repeated to draw the width 103 of the chip 102 in the same column. Thereafter, drawing of the width 103 of the corresponding portion of the adjacent chip row is performed while moving the stage in the opposite direction. After such an operation is performed for all the chip rows, drawing of the width 103 of the next adjacent portion is performed. Actually, the stage moves while drawing while deflecting by the sub deflector or the main deflector, and performs drawing while correcting the deflection amount of the sub deflector or the main deflector.

In the BAA exposure method, the area exposed by one aperture from the point of resolution is, for example, 0.128 μm square.
It is. Therefore, in comparison with the variable rectangular method of 3 μm square, for example, in the case of the pattern of the full filling, if there are no openings of 25 rows or more, the throughput is reduced. Therefore, in order to realize a practically high drawing speed, for example, the arrangement of the apertures of the BAA is 64 rows in the direction perpendicular to the scanning,
Eight rows are arranged in a zigzag pattern in the scanning direction. Arranging the holes in a zigzag pattern is not possible because they cannot be arranged in the same row without interruption, so it is necessary to apply the on / off data by shifting in the scanning direction and sending it temporally. Thus, it is possible to draw one pixel row without interruption.

In the case of an exposure apparatus using a charged particle beam (electron beam), the density of exposure charge is a value obtained by dividing the product of the beam current and the beam residence time by the beam area. Therefore, in order to improve the throughput, generally, it is sufficient to increase the beam current amount and shorten the beam stay time. Therefore, also in the BAA exposure method, the number of apertures is increased, the amount of current of the beam emitted from the electron gun is increased, and the total amount of current is increased in order to improve the throughput.

However, in an exposure apparatus using a charged particle beam (electron beam), it is known that a phenomenon occurs when a desired pattern size and an actual exposure pattern size differ due to Coulomb interaction of charged particles. I have. The aforementioned defocus phenomenon corresponds to this. This phenomenon causes (1) a problem that the minimum resolution size is deteriorated, and (2) a problem that the degree of blurring of the exposure pattern changes depending on the exposure pattern. The first problem is an unavoidable problem caused by Coulomb interaction, but the second problem is corrected by using the above-described refocus lens.

FIG. 6 is a diagram for explaining a change in the number of on-beams in the conventional BAA exposure method. .., 11B-1, and 11B-. As shown in FIG.
When scanning with 2,..., 11C-1, 11C-2,.
The fine beam in each row, for example, the fine beam 11A-
1, 11A-2,... Change to the ON state as shown in each of the exposure positions (a) to (f) as shown in (2). The arrow indicates the position of the edge of the pattern. Therefore, the total current changes approximately as shown in (3). That is, all fine beams are exposed to the pixels in the region T, and the number of exposures is zero in the region outside the region T. For example, 8
In the case of a row opening, exposure is performed eight times for pixels in the region T, and no exposure is performed for a region outside the region T.

FIG. 7 is a diagram showing an example of the exposure amount distribution in the conventional BAA exposure method as described above, wherein (1) shows the exposure pattern, and (2) shows the exposure amount distribution on the line AA '. In the drawing, a solid line indicates an ideal exposure amount distribution, and a dotted line indicates an actual exposure amount distribution. As shown in FIG. 7 (2), ideally, the exposure amount in the non-exposure area is zero, and the exposure amount of any pixel in the exposure area is the same. By performing such exposure, an exposure amount distribution with high contrast can be obtained. However, because of the backscattering of the beam,
The exposure amount distribution becomes like a dotted line with a blunt edge. However, in the case of a large exposure area, the maximum exposure is at a certain distance from the edge, and the exposure is zero in a large non-exposure area.

[0017]

As described above, the BAA exposure method in which the apertures are arranged two-dimensionally uses a single beam or one beam.
Since the change in the number of fine beams in the ON state, that is, the change in the total current amount is reduced as compared with the fine beams in the row, there is an advantage that the change in the correction amount by the refocus lens can be reduced. The smaller the change in the correction amount by the refocus lens, the smaller the change in the drive signal of the refocus lens and the lower the slew rate of the drive circuit for driving the refocus lens. Therefore, if the driving circuits have the same slew rate, the settling time of the refocus lens can be shortened, and higher-speed operation can be performed.

However, in order to further improve the throughput, it is necessary to further increase the total current amount. As the total current amount increases, the change width of the total current amount also increases. , The change in the correction amount also increases. Therefore, in order to further improve throughput,
It is necessary to improve the slew rate of the drive circuit of the refocus lens.

The scanning speed can be further increased by increasing the total current amount. However, when the scanning speed is increased, the exposure time at each scanning position of the fine beam is shortened. Therefore, it is necessary to further improve the slew rate of the drive circuit for driving the refocus lens. For the above reasons, the exposure time of the fine beam is extremely fast, several nanoseconds. Even in the case of the BAA method, the throughput is further improved due to the slew rate of the drive circuit of the refocusing lens. Is getting harder. Therefore, the throughput is not determined by the change rate of the exposure pattern shape determined by the ON / OFF signal of the BAA, but the throughput is regulated by the slew rate of the drive circuit for driving the refocus lens, and the maximum total current amount is suppressed and the refocus The above problem is avoided by setting the throughput within a range that the lens can follow.

The present invention has been made in view of such a problem, and the exposure sequence in the BAA exposure method is devised so as to suppress the maximum total current amount of the charged particle beam and reduce the change in the correction amount by the refocus lens. It is an object of the present invention to realize a charged particle beam exposure apparatus and an exposure method capable of reducing the number of charged particles.

[0021]

In order to achieve the above object, a charged particle beam exposure apparatus and an exposure method of the BAA system according to the present invention reduce the number of exposures for pixels in an exposure area, and at least partially reduce the number of exposures. The number of exposures is made different for each pixel so that the exposure amount of the pixel to be exposed falls within a predetermined range. Further, at least a part of the non-exposed pixels may be exposed at least once with a fine beam.

The selection of the pixel and the exposure timing for thinning out the number of exposures in the exposure area or the exposure timing and the pixel to be exposed in the non-exposure area is performed so as to obtain an exposure pattern which does not cause a problem in post-processing such as development. In addition, it is determined so as to suppress the total current amount and reduce the change in the total current amount. FIG. 8 is a view for explaining the principle of a BAA type charged particle beam exposure apparatus and exposure method of the present invention, wherein (1) shows an exposure pattern, (2) shows an exposure amount distribution on the line AA ′, (3) shows the exposure amount range in the exposure area and the non-exposure area. In (2) of FIG. 7, the solid line shows the exposure amount distribution in an ideal case, and the dotted line shows the actual exposure amount distribution. As shown in FIG. 7, in the conventional example, each pixel in the exposure area is exposed to the fine beam a predetermined number of times, so that all the pixels in the exposure area have the same exposure amount. On the other hand, in the charged particle beam exposure apparatus and the exposure method of the BAA method of the present invention, the number of times of exposure of some of the pixels in the exposure area, in FIG. In this case, the exposure amount is reduced by P. If the width of the end portion at which the number of exposures is not reduced is shorter than the length of the row of the opening of the BAA, the total current amount can be reduced as compared with the conventional example, and the change amount of the total current amount can be reduced.

Further, in some pixels in the non-exposed area, in FIG. 8, pixels in the large non-exposed area are exposed a smaller number of times, while the number of exposures in the other non-exposed areas is zero. Thus, only Q may be exposed. As a result, the total amount of current cannot be reduced, but the amount of change in the total amount of current can be reduced. By performing the above exposure, as shown in (3) of FIG. 8, the exposure amount of at least some of the pixels in the exposure area is different, and the exposure amount of the pixel is equal to the maximum exposure amount (maximum exposure amount−P ). Even in the non-exposure area, the exposure amount differs for at least some of the pixels, and the exposure amount of the pixel exists between zero and Q. Therefore, although the contrast is reduced for some pixels, when a pattern is exposed on a resist or the like by charged particle beam exposure, a threshold value exists from its characteristics, and if the threshold value is set between the maximum exposure amount and zero, Even if the contrast is slightly reduced, no problem occurs. Further, when exposing an exposure pattern arranged on a wafer to a certain area with a fine beam of BAA, the ratio of the number of on-beams (hereinafter referred to as a fill ratio) has a distribution of 0 to 100%. This fill ratio corresponds to the total beam current in one exposure. The fill ratio becomes a value close to 100% in a large exposure area. In such a case, even if the beam for exposing the pattern is reduced to a small amount by a small amount, the pattern is formed due to the influence of back scattering of the beam. There is no problem because the pattern does not differ and the exposure distribution is actually as shown by the dotted line in FIG. 8 (2). Furthermore,
For a narrow line or the like, exposure is performed for the number of apertures in one row, so that the same exposure pattern as that of the related art is obtained.

In addition, for the large exposure area and the non-exposure area as shown in FIG. 8 (2), not only the number of exposures is thinned out or added, but also as shown in FIG. If no problem arises from the characteristics such as the above, it is also possible to reduce the change in the total current amount by thinning out or adding an extra number of exposures for a fine pattern portion.

There are various methods for changing the number of exposures of each pixel at which exposure timing. For example,
The upper limit of the number of on-beams that are simultaneously turned on is determined in advance, and the pixel to be changed and the exposure timing are determined so as not to exceed the upper limit and not to deteriorate the exposure pattern. Alternatively, an upper limit value of the change in the number of on-beams may be determined, and the pixel to be changed and the exposure timing may be determined so as not to exceed the upper limit value and not to deteriorate the exposure pattern.

Normally, the exposure control means of the BAA exposure system
ON / OFF indicating for each exposure timing whether to turn on or off the fine beam that has passed through the aperture corresponding to each pixel of the BAA.
Although the present invention has a bitmap for developing the OFF data corresponding to the pixels, the present invention further includes an on-beam counting means for counting the number of on-beams at each exposure timing, and an on-beam counting means for reducing the number of on-beams to a predetermined upper limit or less. And a correcting means for changing the on / off data and storing the corrected on / off data in a bit map. The corrected on / off data is produced and stored before the exposure operation. The process of changing the number of exposures of a pixel takes a considerably long time. However, with this configuration, the exposure is performed after the correction is completed, so that the exposure throughput is not reduced. Further, once the correction data is produced, the same exposure pattern can be used repeatedly.

It is necessary to change the number of exposures of a pixel according to the pattern. Pixels inside a large exposure area or a non-exposure area may be changed, but a pattern obtained by changing with a fine pattern pixel is obtained. Deteriorates, which is not preferable. Therefore, the exposure control means is provided with a change candidate bitmap storing pixels or regions to be changed in advance according to the exposure pattern, and correction is performed by turning on / off the pixels of the pixels or regions determined in the change candidate bitmap. If the correction is performed by changing the data, the correction process is easy. The end of the exposure pattern is excluded from the change candidate bitmap, and the portion other than the end is stored in the change candidate bitmap as a pixel or area to be changed. The width of the end is, for example, one pixel or two pixels. In addition, priorities for changing may be determined according to the pattern, and correction is performed according to the priorities. Further, when a large number of pixels in the same area are changed, the obtained exposure distribution is largely changed, which is not desirable. Therefore, the density of the pixel to be changed may be determined in advance, and the change may not be performed at the density higher than the density. It is preferable that data indicating a pixel or an area to be changed stored in the change candidate bitmap is combined with the pattern data.

[0028]

FIG. 9 is a diagram showing a configuration of an electron beam exposure apparatus according to an embodiment of the present invention. An electron beam exposure apparatus of the BAA exposure system is disclosed in
Since it is disclosed in, for example, Japanese Patent Application Laid-Open No. 7-1995, a detailed description is omitted here, and only a portion related to the present invention will be described. As shown, an electron gun 51 for generating an electron beam is provided in a column 50 of the BAA exposure type electron beam exposure apparatus.
, BAA 10, convergent optical system (electromagnetic lens) 52,
A deflecting unit 53, a stage 54, and an auto-loading unit 55 for automatically supplying and collecting a wafer are provided. The deflecting means 53 includes a main deflector and a sub deflector as described above. The converging optical system 52 has B
An electromagnetic lens that converges the fine beam bundle that has passed through the AA 10 onto a wafer mounted on the stage 54, a refocus lens that dynamically corrects the focus position, and the like are provided.

The optical control unit (EOS) 42 controls the electron gun 51 and the converging optical system 52. The analog control unit (AMP) 43 includes a deflection control unit (DDGEN) 40 and a BAAF / G control unit (S
HT) 41 to control the deflecting means 53, and perform the scanning as described with reference to FIGS. 4 and 5 during exposure. The stage drive unit (SMDAMP) 44 controls the stage 54 according to the data from the deflection control unit 40.
Is moved, and the movement as described with reference to FIGS. 4 and 5 is performed during the exposure. BAA drive unit (BAAD
RV) 45 receives the bit data from the BAAF / G control unit 41 and turns on each deflection electrode of the BAA 10.
Apply an off signal. The autoloader control unit 46 controls the autoloading unit 55. BAA
The F / G control unit 41 is provided with a shooting memory (high-through buffer) for storing bit data corresponding to an on / off signal applied to each deflection electrode of the BAA 10, and is synchronized with deflection by the deflection means 53. To output bit data.

As shown, the data operation computer 3
2, an exposure control computer 33 and an ALC control (automatic loading control) computer 34 are connected to each other via a communication cable 31 and cooperatively control the entire apparatus. The exposure control computer 33 has a deflection control unit 40 and a BAAF / G control unit 41 via a bus 39.
Control. The ALC control computer 34 controls an autoloader control unit 46. A bit map creation unit (BMGEN) 36 and a data transfer control unit 37 are connected to the data operation computer 32 via a bus 35. The disk 38 stores pattern data, and the data operation computer 32 develops the pattern data read via the data transfer control unit 37 into a bitmap of the bitmap creation unit 36. Further, by processing the bit data developed in the bit map, refocusing data for correcting a phenomenon in which the focal length of the electron beam changes is produced, and a process for changing a part of the bit data is performed. This will be described later. The expanded and changed bit data is stored in the disk 38 via the data transfer control unit 37 for each cell stripe or for a plurality of cell stripes. At the same time, the refocusing data is also stored. Since the above operation is performed before the exposure operation is started, the exposure speed is not limited to the speed required for developing the bit data. The bit data and the refocusing data once developed can be used as they are when the same pattern is exposed.

At the time of exposure, the corrected bit data and the refocusing data stored in the disk 38 are shot by the BAAF / G control unit 41 via the data transfer control unit 37 for each cell stripe or a plurality of cell stripes. Transfer to memory. The shooting memory has a capacity capable of storing two pieces of data of about 20 mm square chips. When the size of the chip to be exposed is smaller than 20 mm square, the shooting memory is divided into two parts and data is simultaneously input to one side and the other side is input. , The data of the chip to be exposed next can be transferred during the current chip exposure. If it is larger than 20 mm □, data of a capacity of 20 mm □ × 2 can be exposed.

The on / off data output from the shooting memory is converted to parallel / serial by the BAA drive unit 45, and the timing is controlled so that the actual
The voltage is amplified and output to the voltage applied to the deflection electrode of AA10. Scanning during exposure is performed as described with reference to FIGS. In the continuous movement of the stage 54, for example, one movement in the Y direction is defined as one unit, and the exposure control computer 33 controls start and stop of the deflection control unit 40, and the deflection control unit 40 A signal for continuous movement is output to the unit 44, and a deflection signal is output to the analog control unit 43 according to the detected movement position, so that scanning as shown in FIGS. 4 and 5 is performed. The BAAF / G control unit 41 receives scanning timing signals from the deflection control unit 40 and outputs on / off data from the shooting memory.

FIGS. 10 to 12 show B in this embodiment.
It is a figure showing the example of composition of AA10. BAA10 is shown in FIG.
It has a thin plate-like shape having an appearance as shown in FIG. 0. A large number of apertures and deflection electrodes are arranged in an aperture portion 14 at the center, and is held inside a column 50 to be irradiated with an electron beam. Reference numeral 15 indicates the ON / OFF voltage applied to the deflection electrode.
Reference numeral 16 denotes an electrode pad for an off signal, and reference numeral 16 denotes an electrode pad for a signal line to a ground electrode.

As shown in FIG. 11, the aperture section 14 has a large number of apertures 11 arranged in a staggered pattern. A deflection electrode 17 is provided on one side of each aperture, and a ground electrode 18 is provided on the remaining three sides. The ground electrodes 18 in the same row of apertures are connected to each other. Reference number 1
Reference numeral 9 denotes a signal line connected to the electrode pad 15 and the deflection electrode 1.
This is an electrode pad for connecting between the ground electrode 7 and the ground electrode 18. When the electrodes of the BAA 10 are formed as shown in FIG. 1, there is a problem that the electric potentials of the electrodes of the adjacent openings affect each other and a desired electric field cannot be generated in each opening. In order to solve such a problem, an electron beam exposure apparatus using an electrode to which a ground potential is applied as a guard electrode has been developed by the “Fast Electron Beam Lithography Syst.
em with 1024Beams Individually Controlled by Blank
ing Aperture Array "(Jpn. J. Appl. Phys. Vol. 32 (199
3) The aperture 14 shown in FIG. 11 has the same shape as that disclosed in this document.

For the sake of manufacturing, the BAA 30 is formed by thinning a silicon substrate by etching or the like and then forming an opening 11.
The deflection electrode 17 and the ground electrode 18 are formed on the substrate by plating or the like.
To form A ground potential is applied to the ground electrode 18 in common, and a ground potential is applied to each deflection electrode 17 when irradiating an electron beam passing through the opening (on), and an electric field is generated in the opening. When an electron beam passing through the opening is not irradiated (when off), a predetermined potential other than the ground potential is applied to form an electric field in the opening.
The passing electron beam is deflected so as not to reach the sample 100.

The apertures 11 are arranged in four groups of two rows in the scanning direction, and are arranged in the X and Y directions, each being shifted by half the size of the aperture. In the scanning direction in the same group, the subsequent stage performs repainting of the pattern exposed by the preceding stage with the same data with a time delay. When exposed by this BAA, four groups of fine beams in the same row are exposed to overlap with each other by half of the pitch, and as a whole, the interval between irradiated pixels is half that in the case of two groups of staggered lattices. become. In this way, by arranging the apertures on the BAA in a shifted manner and outputting the data, the fine adjustment of the dimension of the pattern edge portion and the proximity effect correction with a term accuracy can be performed with a value smaller than the pixel size. Becomes possible.

The configuration described above is the same as the basic configuration of the conventional example disclosed in Japanese Patent Laid-Open No. 5-166707 and the like, except that bit data is changed. Hereinafter, a configuration for changing bit data, which is a feature of the present invention, will be described. As described above, the bit data is changed by the data operation computer 32 in FIG. FIG. 12 is a diagram showing the configuration of a bitmap of the bitmap production unit 36. When the wafer is exposed eight times at the same position on the wafer, it has a data width of eight bits. On the other hand, in the conventional example, the same position is either exposed or not exposed, and is a bit map having a 1-bit data width. For positions where the bit data is not changed, all data bits are "0" when on and all data bits are "1" when off. The position where the bit data is changed is changed for some data bits. For example, when it is on, at least one bit is “0”
From “1” to “1”, and when off, changes at least one bit from “1” to “0”. At the time of exposure,
The first data bit is read and output when the first fine beam comes to the corresponding position, and the second data bit is read and output when the second fine beam comes to the corresponding position. Is turned on and off. Therefore, when the data bit at the same position is changed with the bit data in the same column, the number of on-beams changes rapidly. Therefore, it is necessary to change the position of the data bit to be changed at least for the bit data in the same column.

How the bit data is changed depends on the pattern. For example, as described with reference to FIG. 8, the total current amount is controlled to be equal to or less than a predetermined upper limit, The change is made so that the change is equal to or less than a predetermined upper limit value and the exposure pattern is not deteriorated.
FIG. 13 is a flowchart for explaining the bit data changing process, and FIG. 14 shows an example in which the change candidate pattern data is superimposed on the cell stripe size bitmap data. The change processing of the bit data will be described with reference to FIG.

In step 201, the upper limit of the number of ON bits (ie, the total amount of current) and the upper limit of the amount of change in the number of ON bits are determined. The upper limit is determined based on, for example, the response speed of the refocus lens. In step 202, the actual pattern data of the cell stripe size is read from the disk 38 to the work area of the bitmap production unit 36, and a bitmap of the cell stripe size as shown in FIG. 14 is produced. If the data at the end of an adjacent cell stripe is required to change the end of the cell stripe, a bitmap slightly larger than the cell stripe size is also created for the actual pattern data at that end.

In step 203, as shown in FIG. 14, the corresponding cell stripe size corresponding to the previously prepared change candidate pattern data is read. The change candidate pattern data is represented by (a:
Change processing permission density, b: change processing priority order). The change processing permission density a indicates the maximum density at which the bit data can be changed, and the change processing priority b indicates the priority of changing the bit data. For example, as shown in FIG. 14, both the change processing permission density a and the change processing priority b are high inside a large pattern, and the change processing permission density a has the same value in an elongated pattern having a certain width. , The change processing priority b becomes lower toward both ends of the line and becomes higher toward the center. Note that the correction can be performed without using the change candidate pattern data, and in this case, the change candidate pattern data and this step can be omitted.

In step 204, the number of on bits at each position of the cell stripe is calculated, and the amount of change is calculated. This process is, for example, a process of counting the number of on-bits at the position of the opening 11 in the bit map as shown in FIG. Is counted. In step 205, it is determined whether the number of on bits calculated in step 204 and the amount of change are equal to or less than the upper limit determined in step 201. If it is equal to or less than the upper limit value, the process proceeds to step 207,
If it is not less than the upper limit value, the process proceeds to step 206.

In step 206, the number of on-bits and the change amount of the on-bit number are reduced, while referring to the change processing permission density a and the change processing priority b of the change candidate pattern data, Change bit data. In addition, without using the change candidate pattern data, the exposure pattern is calculated from the spread of the fine beam in consideration of the backscattering and the like, and the exposure pattern is less deteriorated.
The change processing may be performed based on an algorithm for changing the maximum value of the number of ON bits and the amount of change in the number of ON bits to be small.

After the end of step 206, the process returns to step 204 again, and from step 204 to step 2 until the condition of step 205 is satisfied and the process proceeds to step 207.
Repeat step 06. In step 207, the bitmap data of the cell stripe size is stored in the disk 38.

In step 208, it is determined whether or not the above processing has been completed for all cell stripes, and the above processing is repeated until the change processing has been completed for all cell stripes. By performing the above-described processing, pattern data in the bitmap format in which the maximum value of the number of ON bits and the amount of change in the number of ON bits are small and the deterioration of the exposure pattern is small is stored on the disk 38. Thereafter, exposure may be performed as in the conventional case.

The embodiment of the present invention has been described above.
Various modifications can be made to the pattern data changing process in consideration of the characteristics of the apparatus and the characteristics of the resist.

[0046]

As described above, according to the charged particle beam exposure apparatus and method of the BAA system of the present invention, the maximum value of the number of ON bits and the number of ON bits of the BAA can be obtained without substantially deteriorating the exposure pattern. Since the change amount of the refocus lens becomes small because the change amount of the refocus lens becomes small, the scanning speed can be increased if the slew rate of the drive circuit of the refocus lens is the same as the conventional one, and the throughput of the apparatus is improved. I do. If the scanning speed is the same as the conventional one, the slew rate of the drive circuit for the refocus lens can be reduced, so that a cheaper drive circuit can be realized.

[Brief description of the drawings]

FIG. 1 Blanking Aperture Array (BAA)
It is a figure showing the example of.

FIG. 2 is a diagram illustrating exposure in a BAA method.

FIG. 3 is a diagram illustrating exposure in a BAA method.

FIG. 4 is a diagram illustrating scanning in the BAA method.

FIG. 5 is a diagram illustrating scanning in the BAA method.

FIG. 6 is a diagram showing a change in an on-beam and a change in a total current amount at an edge portion in the BAA method.

FIG. 7 is a diagram for explaining an exposure amount distribution in a conventional BAA method.

FIG. 8 is a diagram illustrating the exposure amount distribution and the principle in the BAA method of the present invention.

FIG. 9 is a diagram illustrating the overall configuration of an electron beam exposure apparatus according to an embodiment of the present invention.

FIG. 10 is an overall view of a BAA according to an embodiment.

FIG. 11 is a top view of an aperture of the BAA according to the embodiment.

FIG. 12 is a diagram illustrating a configuration of a bitmap for expanding on / off data according to the embodiment.

FIG. 13 is a flowchart illustrating a process of changing pattern data in the embodiment.

FIG. 14 is a diagram showing an example in which change candidate pattern data is superimposed on bitmap data of a cell stripe size in the embodiment.

[Explanation of symbols]

 DESCRIPTION OF SYMBOLS 10 ... Blanking aperture array (BAA) 11 ... Aperture 14 ... Aperture part 17 ... Change electrode 18 ... Ground electrode 32 ... Data operation computer 36 ... Bitmap production unit 38 ... Disk 40 ... Deflection control unit 41 ... BAAF / G control unit 45 BAA drive unit 51 electron gun 52 focusing optics 53 deflecting means 54 stage 100 sample 106 cell stripe

Claims (22)

[Claims] 1. A charged particle beam generator for generating a charged particle beam, and a plurality of apertures having a first electrode and a second electrode around the charged particle beam generator are arranged. A plurality of fine beams corresponding to the above, a blanking aperture array capable of independently controlling whether or not each fine beam is irradiated onto the sample; and the charged particle beam passing through the blanking aperture array Deflecting means for deflecting, the converging means for converging the charged particle beam having passed through the blanking aperture array on a sample, scanning the sample with the plurality of fine beams, and according to an exposure pattern, By controlling the signal applied to the electrode of each aperture of the blanking aperture array, it is possible to perform multiple exposures with different fine beams for each exposed pixel. Ri charged particle beam exposure apparatus comprising an exposure control means for exposing a desired pattern, the different number of light exposure times due to the fine beams of pixels to be exposed, a different number of times for at least some of the pixels,
A charged particle beam exposure apparatus, wherein an exposure amount of a pixel to be exposed is within a predetermined range. 2. The charged particle beam exposure apparatus according to claim 1, wherein, among the pixels that are not exposed, there are pixels that are exposed at least once by the fine beam. 3. The charged particle beam exposure apparatus according to claim 1, wherein the exposure control means determines that a number of on-beams of the plurality of micro-beams simultaneously irradiated on the sample is a predetermined number. A charged particle beam exposure apparatus that controls the exposure to be equal to or less than the upper limit. 4. The charged particle beam exposure apparatus according to claim 3, wherein the exposure control means includes: a fine beam passing through an opening corresponding to each pixel of the blanking aperture array is formed on the sample; A bit map that expands on / off data indicating whether or not irradiation is performed on a pixel, corresponding to pixels, an on-beam counting unit that counts the number of the on-beams at each exposure timing, The on / off data is changed and corrected so that the
A charged particle beam exposure apparatus comprising: a correction unit that stores off data in the bit map. 5. The charged particle beam exposure apparatus according to claim 4, wherein said exposure control means includes a change candidate bitmap storing pixels or areas to be changed in advance according to an exposure pattern. A charged particle beam exposure apparatus for changing the on / off data of a pixel or a pixel of a region determined in the change candidate bitmap; 6. The charged particle beam exposure apparatus according to claim 5, wherein the change candidate bitmap stores a portion other than an end of the exposure pattern as a pixel or an area to be changed. . 7. The charged particle beam exposure apparatus according to claim 6, wherein an end of the exposure pattern has a width of one pixel or two pixels. 8. The charged particle beam exposure apparatus according to claim 5, wherein a pixel or an area to be changed stored in the change candidate bitmap has a change priority determined. . 9. The charged particle beam exposure apparatus according to claim 5, wherein, in the change pixel or area stored in the change candidate bitmap, the density of the change pixel is determined. Exposure equipment. 10. The charged particle beam exposure apparatus according to claim 5, wherein the data indicating a pixel or an area to be changed stored in the change candidate bitmap is included in pattern data. apparatus. 11. The charged particle beam exposure apparatus according to claim 1, wherein the exposure control unit controls an on-beam of the plurality of fine beams that is simultaneously irradiated on the sample. A charged particle beam exposure apparatus that controls a change in the number at each exposure timing to be equal to or less than a predetermined upper limit value. 12. A method in which each of the fine beams is scanned on the sample with a plurality of fine beams capable of independently controlling whether or not each of the fine beams is irradiated on the sample, and the plurality of fine beams are controlled to correspond to an exposure pattern. In a charged particle beam exposure method of exposing a desired pattern by exposing each pixel to be exposed to a plurality of different fine beams a plurality of times, the number of times of exposure of the pixels to be exposed by the different fine beams differs for at least some of the pixels Number of times,
A charged particle beam exposure method, wherein an exposure amount of a pixel to be exposed is within a predetermined range. 13. The charged particle beam exposure method according to claim 12, wherein, of the pixels that are not exposed, there are pixels that are exposed at least once by the fine beam. 14. The charged particle beam exposure method according to claim 12, wherein, among the plurality of fine beams, the number of on-beams simultaneously irradiated on the sample is equal to or less than a predetermined upper limit. Particle beam exposure method for exposing to light. 15. The charged particle beam exposure method according to claim 14, wherein on / off data indicating whether or not the plurality of fine beams corresponding to each pixel is irradiated on the sample is bit-mapped. A developing step of developing corresponding to the pixel above; an on-beam counting step of counting the number of the on-beams for each exposure timing; and the on / off so that the number of the on-beams is equal to or less than the predetermined upper limit. Changed data and corrected
A charged particle beam exposure method, comprising: a correction step of storing off data in the bit map; and performing exposure based on the corrected on / off data. 16. The charged particle beam exposure method according to claim 15, wherein a pixel or a region to be changed is determined in advance as a change candidate according to an exposure pattern, and in the correction step, the change candidate is changed. A charged particle beam exposure method for changing the on / off data of a pixel or a pixel of an area. 17. The charged particle beam exposure method according to claim 15, wherein the change candidate is a pixel or an area of a portion other than an end of the exposure pattern. 18. The charged particle beam exposure method according to claim 17, wherein an end of the exposure pattern has a width of one pixel or two pixels. 19. The charged particle beam exposure method according to claim 16, wherein the change candidate pixel or region has a change priority determined. 20. The charged particle beam exposure method according to claim 16, wherein a density of a pixel to be changed is determined in the change candidate pixel or area. 21. The charged particle beam exposure method according to claim 16, wherein the data indicating the change candidate pixel or area is included in pattern data. 22. The charged particle beam exposure method according to claim 12, wherein the number of on-beams of the plurality of fine beams that are simultaneously irradiated on the sample is different for each exposure timing. A charged particle beam exposure apparatus that controls a change to be equal to or less than a predetermined upper limit.