JPH0653129A - Electron beam exposure method - Google Patents

Abstract

(57) [Summary] [Object] The present invention relates to an electron beam exposure method for exposing a sample by irradiating it with an electron beam, and an object thereof is to perform proximity effect correction by minimum scanning and perform high-speed exposure. [Structure] On the same field area (SF ₁ ) of the sample,
The first exposure pattern is exposed. Then, a second exposure pattern for correcting the first exposure pattern is overlapped and exposed, and is exposed twice on the same field region.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an electron beam exposure method for exposing a sample by irradiating it with an electron beam.

In recent years, ICs, which are expected to play a role as nuclear technology for technological progress in all industries, will take about 4 years in a few years.
Achieved double high integration. For example, in DRAM,
1M, 4M, 16M, 64M, 256M, 1G and their integration are progressing.

Such high integration is solely due to the progress of fine processing technology. Optical technology continues to advance to enable 0.5 μm microfabrication. However, the limit of optical technology is about 0.3 μm. In addition, it is becoming more and more difficult to ensure accuracy in opening a contact hole window and aligning with a lower layer pattern.

In electron beam exposure, fine processing of 0.1 μm or less can be realized with alignment accuracy of 0.05 μm or less. In recent years, the block exposure by the present inventors and the exposure of the blanking aperture array method have been applied to 1 cm ²
Throughput of about 1 sec can now be expected. And, it is becoming possible to expect faster exposure.

[0005]

2. Description of the Related Art Conventionally, there are a raster method and a vector method for exposing a sample such as a wafer or a mask with an electron beam. In either case, the sample is exposed by irradiating a predetermined amount of the electron beam. .

Although not shown in the drawings, the electron beam exposure apparatus conventionally used is generally an electron gun for emitting electrons, an aperture, a predetermined lens, a deflection electrode, an XY stage on which a sample is mounted, and a control circuit. , Laser interferometer, etc.

In such an electron beam exposure apparatus,
When exposing a predetermined pattern with a predetermined exposure amount, it is generally known that the exposure amount is corrected by changing the number of times of electron beam irradiation in order to correct the proximity effect.

Here, the proximity effect means that, for example, when a resist layer is formed on a silicon substrate and electron beam exposure is performed, the electrons incident on the resist layer undergo forward scattering while traveling in the resist layer, and Proceed from the layer into the silicon substrate. The electrons returning from the substrate by multiple scattering and re-incident on the resist layer are further scattered in the resist layer to form backscattering.

Due to such forward scattering and back scattering, an incidental pattern due to scattering is formed around the desired exposure pattern. If the patterns to be exposed are dense, incidental patterns due to scattering from the respective patterns overlap and the exposure intensity exceeds the threshold value for development.

As a result, a phenomenon that a pattern larger than a desired pattern size is drawn occurs. Therefore, it is necessary to perform proximity effect correction.

In the case of the above-mentioned method, since the pattern data is described by a simple ON / OFF signal, the control circuit becomes very complicated to describe this pattern data in multivalue.

Therefore, it is known to scan the electron beam a plurality of times for exposure by the raster method. This is to correct the exposure amount by exposing the stripe on the sample by moving the stage (XY stage) at least once and then returning the XY stage to overlap the stripe exposure on the same portion.

[0013]

However, in the multi-valued exposure by the raster method, the wafer is exposed once and then re-exposed, or the multi-valued exposure is repeated for each stripe. Therefore, the return operation of the XY stage is performed. The time used for is wasted time.

For example, in the case of performing N times (N passes) of exposure by the raster method, the time (about 0.3 seconds) required for one turn of the XY stage is wasted, and N times the time is required for N times of scanning. Becomes

In this case, it is possible to increase the speed by N times in N-pass exposure because it moves at a constant speed during the time other than the turn-around of each stripe, but there is a limit to the response speed of the laser interferometer, and the accuracy is high. The upper limit of the product of the velocity and the velocity, for example, 0.005 μm (14 MHz at λ / 120 minutes), the upper limit is about 70 mm / sec.

Further, when the moving speed of the XY stage becomes high, the leakage magnetic field of the lens magnetic field causes an eddy current induced in the sample or the like, which causes a displacement of the electron beam, which deteriorates the accuracy.

Therefore, the speed of the XY stage cannot be increased, and the proximity effect correction can be performed closely by a plurality of exposures, but there is a problem that the exposure takes a long time in the end.

On the other hand, increasing the number of exposures leads to an increase in the pattern data, and there is a problem in that the amount of data must be increased due to the data storage capacity of the computer and the data transfer speed.

Therefore, the present invention has been made in view of the above problems, and an object thereof is to provide an electron beam exposure method for performing high-speed exposure by performing proximity effect correction with minimum scanning.

[0020]

FIG. 1 shows an explanatory view of the principle of the method of the present invention. 1, a beam generating means for generating a beam, a blanking aperture array having a two-dimensionally arranged blanking aperture for shaping the generated beam, a stage for supporting a sample, and the blanking are provided. An electron beam exposure method using an apparatus comprising a deflecting means for deflecting the beam by a predetermined amount so as to irradiate a predetermined position of the sample with a pattern of the beam transmitted through an aperture array, In the first step, a first exposure pattern is exposed with a predetermined exposure amount on a predetermined field area of a predetermined number of field areas for exposing the sample. In the second step, the first exposure pattern is overlapped on the same predetermined field region, and a second exposure pattern that supplements the first exposure pattern is exposed with a predetermined exposure amount, Obtain a composite exposure pattern. Then, in the third step, the field regions after the predetermined field region are sequentially exposed twice by the first and second exposure patterns.

[0021]

As shown in FIG. 1, the same field region of the sample is sequentially exposed with the first and second exposure patterns.
Repeat exposure is performed in each field area. The second exposure pattern is a pattern that complements the first exposure pattern. For example, exposing the first exposure pattern with a desired pattern,
The exposure pattern of is corrected with the proximity effect correction pattern.

This makes it possible to perform proximity effect correction and perform high-speed exposure with a minimum of two scans.

[0023]

FIG. 2 shows a block diagram of an embodiment of the present invention.
In FIG. 2, the electron beam exposure apparatus for carrying out the method of the present invention comprises an exposure unit 1 and a control unit 5.

The exposure section 1 includes an electron gun 11 formed of a cathode, a Wehnelt, and an anode, focusing electron lenses 12 to 15, and a blanking aperture array (BA).
A) 16 (described later), a blanking electrode 17, a round aperture 18, a main deflector 19, a sub deflector 20, and an XY stage 21. A sample (for example, a wafer) is placed on the XY stage 21.
22 is placed. Reference numeral 23 is a gate, which allows the XY stage 21 to move in and out.

On the other hand, the control unit 5 includes a CPU 51, two data expansion circuit units 52 _A and 52 _B , a whole exposure controller 53, a data emission circuit 54, and a BAA register 5.
5, a blanking circuit 56, and a sub differential scanning circuit 57.
A main diff scan set ring circuit 58, a main diff stig focus 59, and a stage control circuit 60.
, Autoloader control circuit 61, and laser interferometer 62
And a clock setting circuit 63.

Two data expansion circuit sections 52 _A and 52
_B is a buffer memory 52 _A1 , 52 _B1 , a data expansion circuit 52 _A2 , 52 _B2, and a canvas memory 5, respectively.
2 _A3 , 52 _B3 and data rearrangement transfer circuits 52 _A4 , 52
Formed with _B4 .

Data such as an exposure pattern is stored in each of the buffer memories 52 _A1 and 52 _B1.
It is read by the signal from 51. For example, the buffer memory 52 _A1 stores the first exposure pattern and the density pattern, and the buffer memory 52 _B1 stores the second exposure pattern and the density pattern.

Buffer memory 52_A1, 52_B1Read more
The exposed exposure pattern and density pattern are
The data expansion circuit 52 is instructed by the controller 53._A2，
52 _B2Is expanded to bitmap data with
Mori 52_A3, 52_B3Memorized in. And canvas
Memory 52_A3, 52_B3Bitmap data of
Rearrangement transfer circuit 52_A4, 52_B4Sorted by bus
The exposure pattern data is temporarily stored in the BAA register 55.
It is delivered and sent to the blanking circuit 56. Also, the density
The turn data is sent to the data emitting circuit 54, and the data is emitted.
In response to an instruction from the overall exposure controller 53 from the output circuit 54
It is sent to the blanking circuit 56.

For example, these are RAM (DRAM or S
RAM), shift register, etc., 64 bits read data in parallel, the shift register stores the data, and then serially outputs the data.

The blanking circuit 56 drives and controls the BAA 16, and performs 128 ways of control. That is, the blanking circuit 56 is composed of 128 circuits, each of which outputs 8 outputs (sub-output). The deflector 20 performs control with 8 poles. Therefore, the BAA register 55 is
The exposure pattern data is sequentially sent to each of the 128 blanking circuits 56. Further, the data emission circuit 54 is composed of 128 circuits corresponding to the blanking circuit 56, and the output of each one is eight.

On the other hand, the main differential scanning settling circuit 5
Reference numeral 8 sets the exposure position, and controls the main deflector 19 by the main diff stig focus 59 so that the distortion is corrected and the sample 22 is focused. Further, the sub-deflection scanning circuit 57 includes the sub-deflector 2
By controlling 0, the electron beam is scanned on the minute area.

In accordance with an instruction from the CPU 51, the stage control circuit 60 moves the XY stage 21 on which the sample 22 is mounted to move the sample 22 in the exposure area.

The auto loader control circuit 61 includes a gate 23
This is a control circuit for allowing the XY stage 21 to move in and out of the exposure unit 1.

The laser interferometer 62 detects the movement of the XY stage 21, and a detection signal is supplied to the main differential scanning settling circuit 58. That is, the electron beam is deflected by the amount of movement of the XY stage 21, and the sample 2
Irradiation is always performed on predetermined positions on the two surfaces.

The clock setting means 63 sets the on / off frequency of the electron beam on the sample 21, and is determined by the relationship between the exposure amount and the current density.

Here, the BA in FIG. 2 is shown in FIGS.
The figure for demonstrating A is shown. FIG. 3 is a diagram showing the arrangement of holes, FIG. 4 is a diagram showing the entire wiring, and FIG. 5 is an enlarged view of the holes.

In FIG. 3, the BAA 16 has a blanking aperture 31 in the opening having a quadrangular cross section with a length S of one side, and these are staggered in different stages in the column direction at each stage. It is arranged in a predetermined number.

That is, as shown in FIG. 3, only pitches 2S (50 μm as an example, 0.1 μm in terms of the dimension on the sample surface) mutually in the direction perpendicular to the scanning direction of the electron beam (vertical direction in the figure). The staggered first aperture stages are, for example, 1A (the part of the apertures is 1A ₁ , 1A ₂ , 1
Represented as _{A 3, 1A 4, 1A 5} ... 1A n, and the number thereof to the example and 64). In addition, the first opening step 1A is arranged so that the first opening step 1A and the opening openings do not come into contact with each other.
To a predetermined distance in the scanning direction of the electron beam (vertical direction in the figure) (50 μm as an example, converted to the dimension on the sample surface)
0.1 μm), and S (that is, 25 μm in the above case, 0.05 μm in terms of the dimension on the sample surface) in the direction perpendicular to the scanning direction of the electron beam and the first aperture step.
For example, the second opening stages arranged by being shifted by 1B
(Part of the holes are represented as 1B ₁ , 1B ₂ , 1B ₃ ... 1B _n , and the number thereof is the same as the number of the first hole stages, for example, 64). Then, a plurality of hole stage groups (for example, the above 1A and 1B) are arranged (for example, 8 groups like 1A, 1B; 2A, 2B; ... 8A, 8B).

Therefore, in the above case, the dimension a shown in FIG. 3 (that is, the deviation between the hole stage 1B and the hole stage 2B, for example)
Is 100 μm (0.2 μm converted to the size on the surface of the exposed object)
Further, the dimension b becomes 800 μm (1.6 μm in terms of the dimension on the exposed surface), and the dimension c becomes 3200 μm (6.4 μm in terms of the dimension on the exposed surface).

As described above, if the number of holes in one hole stage, for example 1A, is 64, the number of holes in one hole group (for example, 1A and 1B) is 128. The total number of open holes (8 in the above case) is 128.
× 8 = 1024.

That is, in the present embodiment, a plurality of blanking aperture stages having different array phases are set as one set, and the plurality of sets are repeated.

Further, as shown in FIG. 4, electrodes (described in FIG. 5) are formed on the opposite sides of the blanking aperture 31 in the BAA 16, while the electrodes are formed in the middle portion of a set of blanking aperture stages. A main line portion 32 connected to the constant voltage power source is formed. And the main line 3
2, the constant voltage application wiring 33 connected to the above-mentioned electrode is formed.

Therefore, a variable voltage applying electrode 34 and a constant voltage electrode 35 are formed on the opposite sides of the blanking aperture 31 in FIGS. 5A and 5B. Note that FIG.
5A shows a portion belonging to the hole stages 1A to 4B in FIG. 3, and FIG. 5B shows a portion belonging to the hole stages 5A to 8B.

In this case, the variable voltage applying electrode 34 shown in FIG. 5A is provided on the end 36 side, and the variable voltage applying electrode 34 shown in FIG. 5B is provided on the end 37 side.

The main line portion 32 and the constant voltage electrode 35 are respectively extended and branched by the constant voltage applying wiring 33 and connected, and the variable voltage applying electrode 34 is variably extended from the outer periphery of the electron beam passage region. It is connected to the voltage wiring 38.

That is, as shown in FIGS. 5A and 5B, the wiring to the variable voltage applying electrode 34 is indicated by a broken line. In this case, l ₁ is a wiring to the variable voltage applying electrode 34 of the blanking aperture 1A ₁ in the opening step 1A, and similarly, l ₂ , l ₃ and l ₄ are the wirings in the opening step 2A, respectively. Ranking aperture 2A ₁ ,
The holes 3A ₁ in the hole stage 3A and the hole row 4
This is a wiring to the variable voltage applying electrode 34 of the opening 4A ₁ in A. The wirings l _{1 to} l ₄ are the blanking circuit 56.
The electron beam is turned on by connecting to (Fig. 2),
Off is done.

The main line portion 32 is supplied with a predetermined constant voltage, for example, ground (GN).

As described above, by wiring the BAA 16, one integrated constant voltage applying wiring 33 is branched near the blanking aperture 31 and one of them is connected to one electrode. The number of wirings passing through the space between the blanking apertures 31 is reduced, and the area of the wiring region is reduced. Moreover, in such a device, signals can be sent to the electrodes 34 and 35 provided in the plurality of blanking apertures 31 almost at the same time, so that the selection of the electrodes is instantaneously performed and the throughput is improved.

Further, since no element is formed in the electron beam irradiation area in the BAA 16, no malfunction occurs.

Further, the integrated constant voltage applying wiring 33 is formed.
And the variable voltage applying wiring 38 are prevented from crossing, and the portion for applying the constant voltage is sufficient at one place in the concentrated portion. Therefore, the connection point between the blanking circuit 56 for transmitting the pattern signal and the BAA 16 is Less.

Further, since the constant voltage electrode 35 of the blanking aperture 31 is arranged in the direction in which the main line portion 32 of the blanking aperture 31 is present, the constant voltage applying wiring 33 and the constant voltage electrode 35 are provided. The wiring route is shortened and the wiring area is prevented from being expanded.

Next, the operating principle of the electron beam exposure apparatus described above will be described.

In this case, the XY stage 21 in FIG.
Moves in the Y-axis direction at the time of exposure to perform exposure in the continuous movement mode. The main deflector 19 is an electromagnetic deflector using a coil and can deflect in both the XY directions, but it mainly deflects in the X direction by 1 to 2 mm. in this case,
The Y direction is an amount determined by the difference between the target value and the current value of the XY stage 21 at each time, and is controlled and moved by feeding back an error amount from the target schedule of the XY stage 21, and is usually 100 μm to 200 μm
Is the amount of.

That is, the main deflector 19 deflects the electron beam to the exposure position, and DA (not shown).
The converter performs digital / analog conversion. By this conversion, the electron beam is settled in 1 μs, but the current flowing through the coil is not settled. Therefore, by feeding back the main differential error to the sub deflector 20, the beam position is settled at high speed. The waiting time for the beam deflecting by the main deflector 19 is 1 μs for each cell stripe.

The sub-deflector 20 performs electrostatic deflection with eight poles, and scans the electron beam in a subfield of 100 μm □ in the XY directions. Normally, 100 μm is continuously scanned only in the Y direction.

On the other hand, the maximum frequency of the clock that is the basis of the exposure time controlled by the CPU 51 and the overall exposure controller 53 is set to 400 MHz. Originally, for the clock, the exposure amount Q (μC / cm ² ) of the resist was set to the current density of 250.
It is determined by dividing by A / cm ² . Therefore, since the sub-deflector 20 has 8 poles, 10 [μC /
cm ² ] / 250 [A / cm ² ] / 8 = 5 ns, and the frequency
It is 200 MHz, which is 400 MHz in the case of double exposure.

The clock is the clock setting circuit 63.
Can be set arbitrarily.

Therefore, FIG. 6 shows a diagram for explaining the operating principle. FIG. 6 exemplifies a situation in which the electron beam with which the sample 22 is irradiated is subjected to main deflection (main diff) and sub deflection (sub diff) by the scanning signals supplied to the main deflector 19 and the sub deflector 20, respectively. , Y direction is the stage movement direction.

That is, as shown in FIG. 6A, X
The width in the direction is, for example, 5 μm, and the length in the Y direction is, for example, 100 μm.
The subfield area SF ₁ on the sample surface, which is set to m,
The electron beam (line beam) is raster-scanned for 5 μsec in the direction indicated by the arrow in the figure, and then the adjacent subfield area SF ₂ is moved at the same speed as the above and in the opposite direction (as indicated by the arrow in the figure). ) Is raster-scanned. In this way, the subfield areas SF ₁ ,
The SF ₂ , SF _3, ... Are raster-scanned in the directions shown by the arrows in the figure, and the width in the X direction is 100 μm and the length in the Y direction is 100 μm.
Raster scanning is performed on the subfield area of m in a time of 5 μsec × 20 = 100 μsec. Such an operation is performed by the scanning signal from the sub deflector 20.

Next, the line beam is moved to a subfield area SF ₂₁ adjacent to the subfield area by the scanning signal from the main deflector 19 (main differential), and in the same manner as above, the subfield areas SF are sequentially arranged.
Raster scan ₂₁ , ₂₁ , SF ₂₂ ... in the direction of the arrow in the figure, and again scan X
Raster scanning is performed for a subfield area having a width of 100 μm in the direction and a length of 100 μm in the Y direction in a time of 100 μsec. Hereinafter, similarly, the scanning of the main deflector 19 is repeated,
A field region F having a scanning width of 2 mm in the X direction and a length of 100 μm in the Y direction is raster-scanned in a time of 100 μsec × 20 = 2 msec.

That is, the length in the Y direction is 2 msec.
Since the field area of 100 μm is processed, the scanning width in the X direction is 2 mm and the length in the Y direction is 100 μm × 50 in 1 second.
An area of 0 = 50 mm is processed. Therefore, the stage 20
The moving speed may be 50 mm / sec in the Y direction, and the processing area per second is 2 mm × 50 mm = 1 cm ² .

Further, in FIG. 6B, the above-mentioned width of 5 μm
Subfield area (eg SF ₁) To line beam
(One beam size is 0.05 μm □ as described above.
The beam becomes circular on the sample surface. ) Lined up
A₁Is the first opening stage (for example, 1
Beam passing through aperture in A), A₂Is the first opening
Beam passing through an aperture in a step (eg 1B). This
100 subfield regions with a width of 5 μm
Line beam A₁, A₂… A₁₀₀Scanned by
It In addition, the first opening stage (for example, 1A) and the second opening stage
In the above example, the sum of the number of holes in the hole stage (for example, 1B) is 128.
Therefore, the subfield area (for example,
SF₁) Crossing the left and right boundaries, 14 each
Line beam A_1LThrough A_14LAnd A_1RThrough A_14RIs average
It will be a mess. Therefore, the length of the 128 line beams
Is 6.4 μm.

FIG. 6C shows the sub deflector 20.
Is a scanning waveform diagram in the Y direction, in which the horizontal axis represents the elapsed time and the vertical axis represents the beam position in the Y direction on the sample surface.
That is, it shows that the beam exposure is linearly (constantly) performed in one subfield region.

Next, the exposure method of the present invention will be described. The setting of the exposure conditions in the above-mentioned electron beam exposure apparatus is as described above, and is summarized in Table 1.

[0065]

[Table 1]

FIG. 7 shows a diagram for explaining the first exposure method of the present invention. The exposure method in the present invention is
The exposure is performed twice.

That is, in FIG. 7A, when the electron beam is positioned by the main deflector 19 on the subfield region SF ₁ in one cell of the sample 22, the XY stage 21 is moved in the Y direction and , The sub-deflector 20 scans the electron beam in the Y direction to perform the first exposure with the first exposure pattern.

When the first exposure is completed, as shown in FIG. 7B, the XY stage 21 is returned without moving the main differential position, and the second exposure is performed with the second exposure pattern.

Then, as shown in FIG.
Set the differential position to the subfield area SF ₂To jump to
First exposure with the first exposure pattern as shown in FIG.
Then, as shown in FIG. 7 (D), as in FIG. 7 (B),
The second exposure is performed with the second exposure pattern.

Then, in the same manner, the sub-field areas in the next and subsequent steps are sequentially exposed twice.

The first and second exposure patterns are shown corresponding to the number of exposures in each subfield region, and the exposure patterns are the same or different for each subfield. Further, the second exposure pattern is for performing exposure as a proximity effect correction pattern.

Therefore, FIG. 8 shows a diagram for explaining the state of double exposure. FIG. 8A shows the first exposure amount, and FIG.
8B shows the second exposure amount, and FIG. 8C shows the combined exposure amount of these.

Now, the exposure pattern is a minute pattern,
Line and space pattern with a fill ratio of 50%,
And a large pattern.

As shown in FIG. 8A, the first exposure is performed with a normal pattern. The normal pattern is shown in FIG.
Based on the exposure pattern data stored in the buffer memory 52A ₁ of the data expansion circuit 52 _A in the electron beam apparatus shown in FIG. Here, the exposure clock is 1.0 T. In this case, the fine pattern has an exposure amount of 1.0 T, but the line and space pattern and the large pattern have a total energy amount of 1.5 T and 2.0 T including backscatter due to the proximity effect.

In the two-division exposure as in the present invention, the total number of on-beams can be kept below a certain value, and low current scanning with a small Coulomb interaction can be performed.
If less than 200 out of 1000 (20% or less), Coulomb interaction is less blurred. Further, in the case of a large pattern with large filling, a large current scan in which the total current exceeds a fixed value is performed, and a low current scan and a large current scan can be alternately performed in the same cell stripe.

Subsequently, the second exposure (exposure clock 1.0 T) is performed with the proximity effect correction pattern as shown in FIG. 8 (B). This proximity effect correction pattern is based on the exposure pattern data stored in the buffer memory 52 _B1 of the data expansion circuit 52 _B in the electron beam apparatus shown in FIG.

That is, the minute pattern is given an auxiliary exposure amount of 1: 1 and the large pattern is given an exposure amount of zero. Generally, in the vicinity of 30 KV, the total amount of reflected electrons from the large pattern is almost equal to the total amount of energy of the incident beam. Therefore, the exposure amount of 2T is required for the exposure of the minute pattern, and the irradiation amount of 1.0 T is required for the large pattern. . Therefore, the fine pattern is exposed twice, and the large pattern is exposed once.

In the line-and-space pattern, for example, when the filling ratio is 50%, the backscatter gives energy of 0.5 × T.
In the filled area, the total exposure energy is T + 0.5 T = 1.5 T. To match this with the total energy 2T near the pattern, 2T / 1.5 T = 1.333
The dose may be doubled. Since the first exposure amount is 1, the second exposure may be performed by applying an electron beam having an area ratio of 0.333 to the pattern. That is, the second exposure pattern is exposed with a smaller dimension than the first exposure pattern.

In this case, it is necessary to prevent the center deviation of the exposure pattern from occurring, and according to the above-mentioned example, the total number of the reflected electrons and the number close to 2 are selected, and preferably 2 or more are selected. For example, shown in Table 2.

[0080]

[Table 2]

Then, a combination of the first exposure and the second exposure is shown in FIG. 8 (C). That is,
By setting the phenomenon condition to the position indicated by the broken line, the exposure for which the proximity effect correction is performed is performed by two times of exposure.

For example, when exposure is performed on a wafer at a rate of 0.5 cm ² per second, a wafer having an 8-inch diameter (240 cm ² ) can perform 12 exposures per hour in 300 seconds in one exposure. It takes 540 seconds for double exposure, and 6.66 sheets can be exposed per hour. In the case of a wafer having a diameter of 6 inches (135 cm ² ), it is possible to perform exposure processing of 195 seconds for one exposure and 18.4 wafers per hour, 330 seconds for two exposures and 10.9 wafers per hour. it can.

A pattern such as a contact hole pattern having a rough pattern that does not require proximity effect correction is exposed once and is drawn at high speed. It is necessary to correct the proximity effect so that the pattern is exposed twice so that the accuracy can be obtained even if the throughput is slightly lowered.

Here, FIG. 9 shows a diagram for explaining another exposure method in the first exposure method. FIG. 9 shows the case of exposing a small pattern and a large pattern, as shown in FIG.
T) is performed, and in the second exposure, as shown in FIG. 9B, the fine pattern and the large pattern are exposed at 0.5T. Then, a combined exposure pattern as shown in FIG. 9C is obtained. That is, the exposure amount of the minute pattern is increased and the exposure amount of the large pattern is decreased.

Next, FIG. 10 shows a diagram for explaining the second exposure method of the present invention. 10A is the first exposure amount, FIG. 10B is the second exposure amount, and FIG.
10C shows the combined exposure amount, and FIG.
It shows the exposure energy distribution of (C).

Here, when the final exposure pattern has a pattern of a predetermined size, as shown in FIG.
In the first exposure pattern for the first time, the exposure (exposure amount 1.2 T) is performed with the leading edge pattern and the trailing edge pattern of the predetermined large pattern including the minute pattern. The operation in this case is similar to that of the first exposure method.

Subsequently, in the second exposure pattern for the second time, exposure is performed with an exposure amount of 0.8 T including a fine pattern and a connection pattern for connecting the leading edge pattern and the trailing edge pattern. As a result, a combined exposure pattern as shown in FIG. 10C is obtained. In this case, FIG.
As shown in (D), the exposure energy summed with the reflected electrons from the sample 22 is obtained, and an exposure pattern corrected by the proximity effect is obtained by the double exposure by setting the broken line portion as the development condition. Is.

Although FIG. 10 shows the case where the edging pattern is formed for a pattern having a certain size, the edging may be performed for all patterns of the first exposure pattern with a predetermined width. In this case, in the second exposure pattern, the large pattern may be trimmed at the edge, the medium pattern may be trimmed at the edge slightly, and the minute pattern may be trimmed at all.

Although the basic clock for exposure is 400 MHz as described above, the above embodiment has been described, but the number of clocks may be arbitrarily changed.

Furthermore, the first and second exposure patterns are
Although the case of the same density has been described, the exposure amount may be decreased by making the density of one pattern rougher than that of the entire pattern. The concentration is
Although not shown, each pattern is set by a binary value when the bit map data is expanded in the data expansion circuits 52 _A and 52 _B of the electron beam exposure apparatus in FIG. 1, and the density is coarsened. Therefore, the exposure amount can be reduced. In this case, when the density is to be roughened, "1" is performed by appropriately setting 1 to 0 at the points in the bitmap.

[0091]

As described above, according to the present invention, the first exposure pattern is exposed on the same field region of the sample,
Subsequently, a second exposure pattern for correcting the first exposure pattern is overlapped and exposed, and the second exposure pattern is exposed twice to perform proximity effect correction with minimum scanning,
High-speed exposure can be performed.

[Brief description of drawings]

FIG. 1 is a diagram illustrating the principle of the method of the present invention.

FIG. 2 is a configuration diagram of an embodiment of the present invention.

FIG. 3 is a diagram for explaining the BAA of FIG.

FIG. 4 is a diagram for explaining the BAA of FIG.

FIG. 5 is a diagram for explaining the BAA of FIG.

FIG. 6 is a diagram for explaining the operation principle.

FIG. 7 is a diagram for explaining the first exposure method of the present invention.

FIG. 8 is a diagram for explaining a state of double exposure.

FIG. 9 is a diagram for explaining another exposure method in the first exposure method.

FIG. 10 is a diagram for explaining a second exposure method in the present invention.

[Explanation of symbols]

DESCRIPTION OF SYMBOLS 1 exposure part 5 control part 16 BAA 19 main deflector 20 sub deflector 21 XY stage 22 sample 31 blanking aperture 32 main line part 33 constant voltage applying wire 34 constant voltage applying electrode 35 constant voltage electrode 38 variable voltage applying wire 51 CPU 52 _A , 52 _B Data development circuit section 53 Whole exposure controller 54 Data emission circuit 55 BAA register 56 Blanking circuit 63 Clock setting circuit

Claims (8)

[Claims] 1. Beam generating means for generating a beam (1
1), a blanking aperture array (16) having two-dimensionally arranged blanking apertures (31) for shaping the generated beam, a stage (21) supporting a sample (22), Deflection means (1) for deflecting the beam by a predetermined amount in order to irradiate a predetermined position on the sample (22) with the pattern of the beam transmitted through the blanking aperture array (16).
9, 20), and an electron beam exposure method using an apparatus composed of a predetermined number of field regions (SF ₁ , SF ₂ , ...) For exposing the sample (22). (S
F ₁ ), exposing the first exposure pattern with a predetermined exposure amount, and forming the first exposure pattern on the same predetermined field region (SF ₁ ).
Of the second exposure pattern, which overlaps the first exposure pattern with a predetermined exposure amount, to obtain a combined exposure pattern, and a predetermined field area (SF ₁ ) and subsequent steps. An electron beam exposure method comprising: sequentially performing two exposures of the field regions (SF ₂ , SF ₃ , ...) With the first and second exposure patterns. 2. The first exposure pattern includes a minute pattern (), and the second exposure pattern has a minute pattern () having the same shape or a small shape as the minute pattern ().
The electron beam exposure method according to claim 1, further comprising: 3. The line-and-space pattern () is included in the first exposure pattern, and the line-and-space pattern smaller than the line-and-space pattern () is included in the second exposure pattern. The electron beam exposure method according to claim 1 or 2. 4. The electron beam exposure method according to claim 1, wherein the first exposure pattern includes a large pattern (), and the large pattern is excluded from the second exposure pattern. 5. The synthetic exposure pattern includes a large pattern, and the large pattern is removed from the first exposure pattern,
4. The electron beam exposure method according to claim 1, wherein the second exposure pattern includes a large pattern. 6. The composite exposure pattern includes a pattern of a predetermined size, and the first exposure pattern includes a leading edge pattern () and a trailing edge pattern () corresponding to a leading edge and a trailing edge of the pattern of the predetermined size. 6. The electron beam exposure method according to claim 1, further comprising: a connection pattern () in which the leading edge pattern () and the trailing edge pattern () are connected to each other in the second exposure pattern. . 7. The exposure is performed by reducing the density of a predetermined pattern in at least one of the first and second exposure patterns and configuring individual patterns to reduce the exposure amount. The electron beam exposure method according to claim 1, which is characterized in that. 8. The electron beam exposure method according to claim 1, wherein an exposure amount by the first and second exposure patterns is set by an exposure clock, and the exposure clock is variable.