JP2512056B2 - Electronic beam exposure method - Google Patents

Description

The present invention relates to an electron beam exposure method of forming a pattern on a sample on an stage by an electron beam while continuously moving the stage, and performing pattern drawing with high accuracy and in a short time. For this purpose, while continuously moving the stage, the main deflector is deflected in the direction orthogonal to the continuous movement direction of the stage while the sub-deflector sequentially scans a plurality of small deflection areas on the sample surface to form a pattern. In the electron beam exposure method for forming, before the pattern is actually exposed, the plurality of small deflection regions are grouped into a plurality of strip-shaped regions whose longitudinal direction is a direction orthogonal to the continuous movement direction of the stage, and the strip-shaped regions are formed. After counting the number of patterns, the number of shots, and the number of small deflection areas for each, at least those, beam shot time, pattern division processing time, The exposure time is calculated by the calculating means on the basis of the deflector settling time, and while the stage is continuously moved at a moving speed capable of exposing the belt-shaped area within the calculated exposure time for each belt-shaped area, only the pattern existing area is electronically moved. The beam is deflected and scanned to form a pattern.

[Industrial applications]

The present invention relates to an electron beam exposure method, and more particularly to an electron beam exposure method for forming a pattern on a sample on a sedge by an electron beam while continuously moving a stage.

[Conventional technology]

Scanning methods of electron beam exposure apparatuses are roughly classified into a raster scanning method in which an electron beam is turned on / off according to the presence or absence of a pattern, and a vector scanning method in which an electron beam is deflected and scanned only where a pattern exists.
There is a method. Also, regarding the stage movement method,
It is roughly divided into a continuous movement method and a step-and-repeat (intermittent movement) method.

As shown in FIG. 9 (a), the data designation method in the raster scan method designates data at a constant pitch on the sample, and the value of the data is "1" where patterns A and B are present,
The areas where the patterns A and B do not exist are set to “0”, and the electron beam is controlled to be turned on / off to form a pattern.

On the other hand, as shown in FIG. 9 (b), the data specifying method in the vector scan method is such that the starting point coordinates (X ₁ , Y ₁ ) and (X ₁ ′, Y ₁ ′) of patterns C and D and Side distance X ₂ and
Specify Y ₂ , X ₂ ′ and Y ₂ ′, and irradiate the electron beam only at the given coordinates.

Regarding the beam deflection direction, in the raster scan method, as shown in FIG. 10A, the electron beam is deflected back and forth with respect to the patterns A and B to be formed (however, the return path is always an electron beam). The beam is off). Therefore, where patterns A and B are formed (where data is "1")
Only the electron beam is turned on, and the electron beam is turned off at other places (where the data is "0"), and the electron beam control system has a relatively simple configuration.

On the other hand, in the vector scan method, as shown in FIG. 10 (b), the electron beam is deflected and scanned only in the regions C and D to be formed as shown by the broken line. Therefore, since the electron beam is not scanned where the patterns C and D do not exist, the drawing time can be shortened as compared with the raster scan method.

On the other hand, regarding the two stage movement methods of the electron beam exposure apparatus, the continuous movement method is a method of drawing a pattern while continuously moving the stage, and it is necessary to synchronize the stage movement speed and the beam deflection scanning. is there.

On the other hand, in step-and-repeat, the stage is not moved until the drawing of one field (drawing range) is completed, and after the drawing of one field is completed, the stage is moved for drawing one field next. Therefore, the stage is moved intermittently, and the stage movement and the pattern drawing are sequentially asynchronous.

As described above, in the raster scan method, the electron beam control system has a relatively simple configuration, and since the electron beam is deflected and scanned regardless of the presence or absence of a pattern, a constant exposure speed (drawing time) is always required. , As shown in FIG. 10 (a), the stage on which the sample on which the patterns A and B are to be formed is to be continuously linearly moved upward in the figure, and the electron beam is deflected and scanned. Synchronization with speed is relatively easy.

Therefore, the raster scan system often employs a continuous stage moving system, and most of the current electron beam exposure apparatuses use this combination.

On the other hand, the vector scan method is often combined with step-and-repeat, but in recent years, an electron beam exposure apparatus combined with a stage continuous movement method has also appeared in order to further shorten the exposure time.

[Problems to be Solved by the Invention]

However, as miniaturization of large-scale integrated circuits (LSI) progresses, the minimum input unit (LSB) to the electron beam exposure system
However, due to the nature of the method, the raster scan method of continuous stage movement has an enormous amount of data (for example, a 10 mm square chip has an LSB of 0.01 μm).
With m, the data amount becomes 10 ¹² bits), which is a problem that it cannot support.

On the other hand, in the vector scan method in which only the necessary pattern data is input data and only the data existing area is deflected and scanned, the pattern amount can be formed with high accuracy without increasing the data amount even when the stage is continuously moved.

However, in this vector scanning method of continuous stage movement, when the density of the exposure pattern is sparse and dense, the drawing time is long in the dense part, and the drawing time is short in the sparse part.
The exposure speed becomes uneven depending on the density of the exposure pattern, and the continuous stage moving speed must be suppressed to a slow speed at which the portion with the longest exposure time can be drawn, and as a result, the exposure time can be shortened. It wasn't enough.

The present invention has been made in view of the above points, and an object of the present invention is to provide an electron beam exposure method capable of performing pattern drawing with high accuracy and in a short time.

[Means for solving the problem]

According to the electron beam exposure method of the present invention, while continuously moving the stage, the main deflector is deflected in the direction orthogonal to the continuous movement direction of the stage, and the sub-deflector sequentially scans a plurality of small deflection regions on the sample surface. In the electron beam exposure method of forming a pattern, before the actual exposure of the pattern, a plurality of small deflection regions are collected into a plurality of strip-shaped regions whose longitudinal direction is the direction orthogonal to the continuous movement direction of the stage, After counting the number of patterns, the number of shots, and the number of small deflection regions for each band-shaped region, the exposure time is calculated by the calculation means based on at least those and the beam shot time, the pattern division processing time, and the main deflector settling time, and the calculation The electron beam is polarized only in the pattern existing area while continuously moving the stage at a moving speed capable of exposing the strip area within the exposure time. It is obtained so as to scan.

[Action]

In the vector scan method, the electron beam is deflected and scanned only in the pattern existing region, but the movement of the stage is stopped during the writing period of one field. Therefore, as described above, the exposure time changes depending on the density of the exposure pattern. To do. .

On the other hand, in the present invention, the electron beam is deflected and scanned only in the pattern existence region, and the stage is continuously moved. Since the moving speed of the stage is calculated based on the exposure time for each strip-shaped area calculated before the actual exposure and is determined based on the calculated exposure time, the dead time due to the unevenness of the exposure time can be reduced.

〔Example〕

FIGS. 1A and 1B are flowcharts for explaining the operation of an embodiment of the present invention, respectively, which are applied to the electron beam exposure apparatus having the configuration shown in FIG. Therefore, first, the configuration of an example of the electron beam exposure apparatus shown in FIG. 8 will be described. Ninth
In the figure, 1 is a central processing unit (CPU), 2 is a magnetic disk,
A magnetic tape 3 is connected to each other via a bus 4 and is also connected to a data memory 6 and a stage control circuit 7 via a bus 4 and an interface circuit 5, respectively.

On the other hand, 8 is a housing, and inside is an electron gun 9, a lens 10, a blanking electrode 11, a lens 12, a slit deflector electrode 13, a sub deflector electrode 14, a lens 15, a main deflector coil 16, a stage 17 and a sample. 18 are arranged. The sample 18 is placed on the stage 17, and the stage 17 is controlled to move in the X and Y directions by the output signal of the stage control circuit 7.

Further, the data read from the data memory 6 is supplied to the pattern correction circuit 20 through the pattern generation circuit 19. The pattern correction circuit 20 applies a blanking signal to the blanking electrode 11 via an amplifier 21,
Signals are applied to the electrodes 13 and 14 and the coil 16 via converters (DAC) 22, 24 and 26 and amplifiers 23, 25 and 27.

The electron beam emitted from the electron gun 9 passes through the lens 10, is transmitted or blocked by the blanking electrode 11, and is further shaped by the slit deflector electrode 13 into, for example, a rectangular beam having an arbitrary shot size of 3 μm or less in parallel. The sub-deflector electrode 14 causes small deflection at a speed of about 100 μm or less and at a high speed of about 0.6 μs / 100 μm.
Further, the electron beam passes through the projection lens 15 and is about 10 mm or less by the main deflector coil 16 and is 2 to 30 mm.
After being largely deflected at a low speed of about μs / 100 μm, the surface of the sample 18 is irradiated and scanned.

In this way, a pattern based on the data read from the data memory 6 is drawn on the surface of the sample 18.

In such an electron beam exposure apparatus, in the present embodiment, a plurality of small deflection areas (sub-deflector deflectable areas: 100 μm square, for example) are divided on the surface of the sample 18 from the control on the hardware of the apparatus. The "fields" are grouped on a band whose longitudinal direction is the direction orthogonal to the moving direction Y of the stage as shown in FIG. 2 (hereinafter referred to as "bands" B1, B2, B3, ...). Therefore, the longitudinal directions of the bands B1, B2, B3, ... Align with the deflection direction X of the main deflector.

In the present embodiment, the memory circuit in the data memory 6 has a third
As shown in the figure, it is composed of a main memory ((a) in the figure), a sub memory ((b) in the figure) and a band memory ((c) in the figure). The main memory has a band start flag (this value is " 1 "indicates the start position), a storage region for storing the relative position coordinates (X ₀ , Y ₀ ) of the subfield from the center of the chip, and the start and end addresses of the submemory (subfield) A pointer storage area or the like is provided.

As shown in FIG. 3 (b), the sub memory stores data such as the starting point (X ₁ , Y ₁ ) and the size (X ₂ , Y ₂ ) indicating the pattern in the sub field and the pattern shape data. A storage area is provided.

Further, as shown in FIG. 3 (c), the band memory is provided with respective storage areas for the number of subfields, the number of patterns and the number of shots of each band. The exposure time is calculated. This band memory is not used at the time of actual exposure, but can be used for confirmation after the pattern exposure is completed.

The data memory 6 has not only the above-mentioned memory circuit but also a structure as shown in FIGS. 4 and 5. In FIG. 4, 30 is a shot decomposition unit, 31 is a counter for counting the number of shots (shot counter), 32 is a counter for counting the number of patterns (pattern counter), 33 is a counter for counting the number of subfields (subfield counter). ), The outputs of these counters 31 to 33 are supplied to the band memory BM, where they are stored for each band.

FIG. 5A is a block diagram of an embodiment of the shot decomposition unit 30 described above, in which data X ₂ and Y ₂ indicating the size of the pattern are supplied from the sub memory SM. That is, one pattern in the sub-field has a size of X _{2 in} the horizontal direction and Y _{2 in} the vertical direction as shown by the solid line in FIG. 5B. Assuming that the above pattern is formed by arranging in a matrix, the shot decomposition part
Of the input data X ₂ and Y ₂ of 30, X ₂ is supplied to the memory 35,
Here, the number of shots N _X stored in advance is referred to from the value X ₂ and output.

The data of this shot number N _X is similarly Y ₂
Input terminal 37 showing the number of vertical shots N _Y obtained from
Is supplied to the multiplier 36 together with the data of No. 1, and is multiplied here and output to the output terminal 38 as data of the number N of shots of one pattern. The data of the shot number N is supplied to the shot counter 31 of FIG. 4, where it is integrated for each pattern and the integrated number per band is stored in the band memory BM as the shot number.

In this embodiment, the data memory 6 is configured as described above, and the CPU 1 and the stage control circuit 7 are operated according to the flowcharts shown in FIGS. 1A and 1B. The operation of the present invention will be described with reference to FIGS. 1A and 1B. In FIG. 1A, prior to the actual exposure, first, the pattern data is read from the magnetic disk 2 according to the instruction of the CPU 1, and the pattern data is supplied to the data memory 6 through the bus 4 and the interface circuit 5, respectively. Data is transferred to the sub memory SM (step S ₁ ). The transfer of the pattern data to the sub memory SM is carried out for all the objects on the surface of the sample 18 to be exposed (step S ₂ ).

When the transfer of all the pattern data is completed, the data for the main memory of the magnetic disk 2 is read out, and the data is transferred to the main memory in the data memory 6 through the bus 4 and the interface circuit 5, respectively (step).
S ₃ ). Then, the above operation is carried out until the data transfer of all the main memory data to the main memory is completed (step S ₄ ).

Here, the operation as shown in FIG. 1B is performed during the processing period of the above step S ₃ . In FIG. 1B, after initialization is performed to clear each counter in the band memory and the data memory 6 (step S ₁₁ ), the data (X ₀ , Y ₀ ) etc. stored in the main memory is set to 1 One read,
And it is incremented by one subfield counter (step S _12).

Next, the sub-memory end address corresponding to the data (X ₀ , Y ₀ ) is read (step S ₁₃ ), and it is determined whether all the data (X ₀ , Y ₀ ) have been read (step S ₁₃ ).
S _14), when the reading of all the data is not completed is determined whether the value of the band start flag is "1" (step S _15).

When the value of the band start flag is "0", it is determined whether the sub-memory end address (step S _16), when not yet reached the sub-memory end address, the sub-memory as described in conjunction with Figure 4 One piece of data (X ₂ , Y ₂ ) is read from that address and the pattern counter 32 is incremented by 1 (step S ₁₇ ).

Thereafter, the shot decomposition memory as shown by 35 in FIG. 5 (a) is accessed and the number of shots N _X , N _Y is obtained from the data X ₂ , Y ₂ showing the size of the pattern (step
S ₁₈ ), as described above, the multiplier 36 multiplies the two to calculate the shot number N (= N _X · N _Y ) (step S ₁₉ ), and sets the calculated value N in the shot counter 31 (step S ₁₉ ).
S ₂₀ ).

Next, after incrementing the sub memory address by one (step S ₂₁ ), it is again determined whether or not it is the sub memory end address (step S ₁₆ ). If it is not the sub memory end address, the above step S ₁₇ is again performed. ~ Perform the operation of S ₂₁ .

In this way, when one sub-field pattern data is read from the sub-memory, the sub-memory address becomes the sub-memory end address. Therefore, the next one sub-field data is read again from the main memory. And the subfield counter is incremented by 1 (steps S ₁₆ and S ₁₂ ).

Thus, the number of patterns, the number of shots, the counting of the number of subfields is performed from the pattern data of one sub-field that the step S ₁₃ to S ₂₁ again. Less than,
When the hand start flag read from the main memory becomes "1" by repeating the same operation as the above, the subfield is the first subfield of the band, so the subfield counter 33 and the pattern counter up to immediately before that. After the integrated values of 32 and the shot counter 31 are transferred to and stored in the band memory (steps S ₁₅ and S ₂₂ ), the number of counters is cleared (step S ₂₃ ). The integrated value indicates the number of subfields, the number of patterns, and the number of shots of the band immediately before that.

Hereinafter, the process returns to step S ₁₂ again, steps S ₁₃ ~
Operation of S ₂₃ is repeated, the operation is completed by the read completion of all the data stored data in the main memory (step S _24).

In this way, the number of subfields, the number of patterns, and the number of shots for each band are previously stored in the band memory before actual exposure.
It will be stored in BM.

Then, at the start of the actual exposure, the band memory BM is accessed by the CPU 1, and the exposure time T _n is calculated for each band based on the following equation, for example.

T _n = T _M・ M _n ＋ T _P・ P _n ＋ T _S・ S _n (1) where T _M : Main deflector settling time T _P : Pattern division processing time T _S : Beam jot time M _n : Number of subfields P _n : number of patterns S _n : number of shots FIG. 6 shows an example of the relationship between each band and the calculated exposure time. In this way, the exposure time T _n varies depending on each band.

Next, based on the following equation, T _n · v _n = constant (2), the stage moving speed v _{n for} each band is calculated in a form inversely proportional to the exposure time T _n . For example, band width 100μ
m, exposure time T _n = 10 ms, stage movement speed v _n is 10
mm / s (= 100 μm / 10 ms). An example of the relationship between the stage moving speed and the band is shown in FIG. As can be seen from FIGS. 6 and 7, the stage moving speed v _n is slow in a band with a long exposure time (for example, band No. 2) and the stage moving speed v _n is in a band with a short exposure time (for example, band No. 3). _n becomes faster. The slowest stage movement speed V there
_{By controlling} the stage 17 to move at a constant speed as indicated by I in FIG. 7 with _nMIN , a desired pattern can be formed in a short time. At this time, the electron beam is deflected and scanned only in the area where the pattern exists.

〔The invention's effect〕

As described above, according to the present invention, the exposure time for each band-shaped area (band) is calculated in advance before the actual exposure, and the electron beam is patterned while continuously moving the stage at the stage moving speed determined based on the exposure time. Since only the existing area is deflected and scanned, it is possible to perform pattern formation (drawing) with higher accuracy than the conventional raster scan method and in a shorter time than the conventional vector scan method, and for each band. It is also possible to perform a smoothing process on the change in the exposure time and perform the pattern drawing with the stage at a variable speed instead of the constant speed. In this case, the drawing period can be further shortened.

[Brief description of drawings]

FIG. 1A is a flow chart for explaining the operation of the embodiment of the present invention, FIG. 1B is a flow chart for explaining the operation of the embodiment of the main part of the present invention, FIG. 2 is a band explanatory view, and FIG. 3 is the present invention. FIG. 4 is an explanatory view of a memory map of an embodiment of a main part, FIG. 4 is a block diagram of an embodiment of the main part of the present invention, FIG. 5 is a shot exploded view of FIG. 4, and FIG. FIG. 7 is a diagram showing an example of a relationship between time and band, FIG. 7 is a diagram showing an example of a relationship between band and stage moving speed, FIG. 8 is a configuration diagram of an example of an electron beam exposure apparatus to which the present invention can be applied, and FIG. FIG. 10 is an explanatory view of the data specifying method, and FIG. 10 is an explanatory view of the stage movement and beam deflection direction of each method. In the figure, 1 is a central processing unit (CPU), 6 is a data memory, 7 is a stage control circuit, 9 is an electron gun, 11 is a blanking electrode, 13 is a slit deflector electrode, 14 is a sub deflector electrode, and 16 is a A main deflector coil, 17 is a stage, 18 is a sample, 30 is a shot disassembly unit, 31 is a shot counter, 32 is a pattern counter, 33 is a subfield counter, and B1 to B5 are band areas.

Claims (1)

(57) [Claims] 1. A main deflector is deflected in a direction (X) orthogonal to a continuous movement direction of the stage while continuously moving a stage (17), and a sub-deflector moves within a plurality of small deflection regions on a sample surface. In an electron beam exposure method of sequentially scanning to form a pattern, before the actual exposure of the pattern, the plurality of small deflection regions are
The number of patterns and the number of shots are grouped into a plurality of strip-shaped regions (B1, B2, ...) Having a direction perpendicular to the continuous movement direction of the stage (17) as its longitudinal direction. After counting the number of small deflection areas, the exposure time is calculated based on at least those, the beam shot time, the pattern division processing time, and the main deflector settling time.
6,7), and the pattern resident area while continuously moving the stage (17) at a moving speed capable of exposing the belt-shaped areas (B1, B2, ...) Within the calculated exposure time for each belt-shaped area. An electron beam exposure method characterized in that only an electron beam is deflected and scanned to form a pattern.