JPH09293669A - Charged particle-beam drawing device and method - Google Patents

Abstract

PROBLEM TO BE SOLVED: To enable a charged particle-beam drawing device to precisely draw a slant pattern without deteriorating it in throughput by a method wherein a correction of a dimensional error originating in an angle of slant is made to reflect drawing data. SOLUTION: A slant angle recognition section 31 detects the slant part of drawing data, and angle data are added to drawing data related to the detected slant part. The drawing data where the angle data are added are sent to a drawing data correcting section 32. The drawing data correcting section 32 refer to an irradiation dose required for correcting a dimensional error which originates in an angle of slant, getting access to a magnetic disc 21 through a control computer 20. At this point, a correction irradiation dose is previously determined by an experiment and stored as a table in the magnetic disc 21. An electronic beam is controlled in size by impressing a prescribed deflection signal to a beam size varying deflector 14 of an electron-optical system from a beam forming driver 26.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a charged beam drawing apparatus and a drawing method for drawing a pattern such as an LSI on a sample such as a mask or a wafer with high speed and accuracy.

[0002]

2. Description of the Related Art LSI patterns have become finer and more complex in recent years. As a technique for drawing such a pattern on a sample such as a wafer with a charged beam, for example, a resist is formed along a target pattern with a circular cross-section electron beam having a diameter of about ½ to ¼ of the minimum dimension of the pattern. There is a method for drawing (first prior art).

However, since this method is a method in which the pattern is sequentially filled with a beam having a fine circular cross section, there is a problem that the drawable area per unit time is small and the drawing speed is slow.

As a second conventional technique for improving this point, a variable shaped beam type electron beam drawing method has been developed in recent years. This second conventional technique, as shown in FIG. 8, includes first and second rectangular apertures 51, 5
The two images are superimposed to form a rectangular projection beam image 53, and the relative positions of the rectangular apertures 51 and 52 are controlled to change the dimensions of the projection beam image 53 according to the desired pattern shape. It is designed to be exposed while being exposed.

According to this method, for example, in the first conventional technique, 24 times of exposure with a circular beam was required to expose the pattern as shown in FIG. As shown in b), only two exposures are required.

However, this method has a problem that the exposure time becomes long for a pattern including diagonal lines. That is, as shown in FIG. 10, the pattern in the shaded area is
A method of performing exposure by dividing into a plurality of rectangular patterns 54a to 54d and 54c to 54h and performing rectangular approximation is adopted.
In this case, a staircase shape occurs in the shaded portion, and the dimensional accuracy of the pattern deteriorates. Therefore, in order to obtain the target accuracy, it is necessary to increase the number of divisions until the staircase shape can be ignored, which causes a problem that the exposure time becomes long.

As a technique (third conventional technique) capable of solving such a problem, an electron beam drawing method using first and second beam shaping apertures 55 and 56 as shown in FIG. 11 has been proposed. .

The first beam forming aperture 55 in this method is an aperture formed in a rectangular shape as shown in FIG. 11 (a), and the second beam forming aperture 56 is shown in FIG. 11 (b). ), A rectangular shape 3 parallel to either side of the first beam shaping aperture 55
An aperturer shaped to have a total of seven sides, four sides of a square having an angle of 45 ° to any of these sides.

Then, these beam shaping apertures 5
5 and 56, as shown in FIG. 9C, a rectangular projection beam image 57 and an isosceles right triangle projection beam image 58a.
~ 58d are formed, and these projection beam images 57, 58a ~
If a pattern composed of five patterns 59 and 60a to 60d is drawn using 58d, as shown in FIG.
A pattern similar to that shown in FIG. 10 can be drawn with a small number of exposures and without impairing the dimensional accuracy of the pattern.

In this method, the patterns 60a-60d are
As shown in the figure, in the case where the X side and the Y side sandwiching the right angle of the triangle have the same length, that is, the oblique side of the triangle is 45 °, the pattern can be formed without impairing the dimensional accuracy of the pattern as described above. .

However, it has been difficult to obtain a highly accurate pattern in a short drawing time for a pattern including a hypotenuse other than 45 °. As a technique (fourth conventional technique) capable of solving such a problem, a method in which approximation of oblique line division is improved has been proposed (JP-A-1-175737).

That is, as shown in FIG. 13A, the first step of dividing the drawing pattern into basic figures including rectangles and trapezoids, and as shown in FIG. 13B, the X direction and the Y direction.
The second step of dividing the basic figure including the shaded portion with respect to the direction by line segments parallel to the X direction or the Y direction to form an assembly of rectangles and right triangles; and, as shown in FIG. Among them, a right-angled triangle whose side length in the X direction and side length in the Y direction are different is divided by a line segment parallel to the larger side in the X direction and the Y direction and elongated along the parallel line segment. An approximation method has been proposed which has a third step of forming a group of trapezoidal figures and further dividing each of the trapezoidal figures into a rectangle and a right triangle.

As a result, a right-angled triangle 76 which can be formed by overlapping two molding apertures at the end of the shaded portion.
By embedding a, it is possible to alleviate the step generated by the approximation without reducing the width of the oblique line division, that is, to draw the oblique line pattern at an arbitrary angle without lowering the drawing speed and the dimensional accuracy.

However, even this oblique line approximation method is not always sufficient to satisfy the ever-increasing demand for higher precision of the electron beam drawing apparatus using the variable shaped beam.

The state will be described below. FIG. 15 shows L
The pattern used for evaluating the diagonal dimension accuracy of the electron beam drawing apparatus for SI mask production is shown. As can be seen from the numerical values of the angle of the diagonal pattern around the pattern, lines and spaces are incorporated in 5 ° increments from 5 ° to 180 °. There are four types of patterns with different pattern widths for each angle. The pattern width is
l. 0 [μm], 2.0 [μm], 4.0 [μm],
There are four values of 8.0 [μm].

Using this pattern, the present inventors have evaluated the diagonal dimension accuracy of the electron beam drawing apparatus. FIG. 16 shows the evaluation result when the pattern width is 2.0 μm. In the figure, the vertical axis represents the error of the pattern width from the design dimension (2.0 μm),
The horizontal axis represents the angle.

From the figure, 0 °, 45 °, 90 °, 135
Since it is not necessary to approximate the oblique lines of 180 ° and 180 °, it can be seen that dimensional error hardly occurs. On the other hand, it can be seen that the dimensional error is large for other angles to be approximated. Similar results were obtained for other pattern widths.

In order to reduce the dimensional error, that is, to improve the dimensional accuracy of the diagonal line pattern, the division width in the diagonal line approximation may be reduced, but the drawing speed is inevitably reduced. Since the figure division by the oblique line approximation is symmetrical with 90 ° in between, the dimensional error is also symmetrical within the range of measurement error.

[0019]

As described above, various electron beam drawing methods (apparatuses) have been proposed in the past, but their drawbacks have been clarified, and in order to further improve drawing accuracy, There is nothing effective yet.

That is, the first conventional technique has a problem that the drawing speed is slow because the patterns to be drawn are sequentially filled with an electron beam having a circular cross section. Moreover, in the second conventional technique, since the drawing surface is performed with a rectangular projection beam image, it is difficult to draw a highly accurate pattern in a short time for a pattern including a diagonal line. was there.

The third conventional technique does not impair the dimensional accuracy for a pattern having a hypotenuse of 45 °,
Although it is possible to draw in a short time, there is a problem that it is difficult to obtain a highly accurate pattern in a short drawing time for a pattern including a hypotenuse other than 45 °.

Further, in the fourth conventional technique, it is possible to draw an oblique line having an arbitrary angle without slowing down the drawing speed, but the dimensional accuracy of the oblique line can sufficiently satisfy the demand for higher accuracy. There was a problem that it was not a thing. Further, there is a problem that the drawing speed must be reduced in order to guarantee the accuracy.

The present invention has been made in consideration of the above circumstances, and an object thereof is a charged beam drawing capable of drawing a slanted line pattern of an arbitrary angle with higher accuracy than before without causing a decrease in throughput. An object is to provide an apparatus and a drawing method.

[0024]

[Means for Solving the Problems]

[Outline] In order to achieve the above object, a charged beam drawing apparatus according to the present invention (claim 1) is necessary for drawing a pattern from a charged beam based on drawing data of a pattern to be drawn on a sample. A charged beam drawing apparatus comprising means for forming a plurality of kinds of variable shaped beams and means for drawing the pattern on the sample using these variable shaped beams, wherein a hatched pattern existing in the drawing data Is included in the drawing data, and the drawing data is corrected using the corrected drawing data obtained by the data correcting means. And

Here, the data correction means, for example, generates data in which the design size, the irradiation amount, or the shot position is corrected so that the hatched pattern actually drawn has a predetermined size.

Further, a charged beam drawing method according to the present invention (claim 2) is based on drawing data of a pattern to be drawn on a sample, and a plurality of types of variable shaped beams required for drawing the pattern from the charged beam. In the charged beam drawing method for forming the pattern on the sample by using these variable shaped beams, the data related to the oblique line pattern existing in the drawing data is corrected according to the angle of the oblique line pattern. It is characterized by using the applied data.

A charged beam drawing method according to the present invention (claim 3) is the charged beam drawing method (claim 2).
In, in the data related to the diagonal pattern, the diagonal pattern actually drawn has a predetermined size,
It is characterized in that data obtained by correcting the design dimension of the diagonal pattern corresponding to the angle is used.

A charged beam drawing method according to the present invention (claim 4) is the charged beam drawing method (claim 2).
In, in the data related to the diagonal pattern, the diagonal pattern actually drawn has a predetermined size,
It is characterized by using data in which the irradiation amount of the variable shaping beam is corrected in accordance with the angle.

A charged beam drawing method according to the present invention (claim 5) is the charged beam drawing method (claim 2).
In, in the data related to the diagonal pattern, the diagonal pattern actually drawn has a predetermined size,
It is characterized in that data obtained by correcting the shot position of the variable shaping beam is used according to the angle.

[Operation] The present inventors have found that there is a certain reproducible relationship between the dimensional error of the diagonal pattern and the angle of the diagonal pattern. For example, the distribution shown in FIG. 15 is the same each time within the range of measurement error, no matter how many times the same measurement is repeated.

Therefore, if the correction is performed according to the angle, the dimensional accuracy can be improved without reducing the division width in the oblique line approximation. Further, since it is not necessary to reduce the division width at the time of the diagonal approximation, there is no problem that the drawing speed (throughput) is reduced.

The present invention (Claims 1 to 5) was created based on the above facts, and the present invention (Claim 1)
Is a charged beam drawing apparatus characterized in that the above correction is performed inside the apparatus, and another invention (Claims 2 to 5) uses the drawing data that has been corrected in advance as drawing data, and uses the same charged beam drawing as the conventional one. This is a charged beam drawing method characterized by performing drawing with a beam drawing apparatus.

In any of the present inventions (claims 1 to 5), in order to improve the dimensional accuracy of the diagonal line pattern, it is not necessary to reduce the division width in the diagonal line approximation, so that the throughput is lowered. Instead, it becomes possible to draw a diagonal pattern with high accuracy.

[0034]

BEST MODE FOR CARRYING OUT THE INVENTION Embodiments of the present invention (hereinafter referred to as embodiments) will be described below with reference to the drawings. (First Embodiment) FIG. 1 is a schematic view showing the schematic arrangement of an electron beam drawing apparatus according to the first embodiment of the present invention.

In the figure, reference numeral 1 denotes a sample chamber in which a stage 3 on which a sample 2 such as a mask substrate is mounted.
Is housed. The stage 3 is driven by a stage drive circuit 4 in the X direction (left and right direction on the paper) and the Y direction (front and back direction on the paper). The moving position of the stage 3 is measured by the position detection circuit 5 using a laser length meter or the like.

Above the sample chamber 1, an electron beam optical system 1
0 is set. The electron beam optical system 10 is
Electron gun 6, various lenses 7, 8, 9, 11, 12, blanking deflector 13, beam size varying deflector 14, beam scanning main deflector 15, beam scanning sub-deflector 16,
And two beam forming apertures 17, 18 and the like. Two beam shaping apertures 1
Reference numerals 7 and 18 are apertures formed similarly to the beam forming apertures 55 and 56 shown in FIG. 11, respectively.

Then, the main deflector 15 positions the beam in a predetermined sub-deflection region (subfield), the sub-deflector 16 positions the figure drawing position in the subfield, and the beam size varying deflector 14 and The beam shape is controlled by the beam forming apertures 17 and 18, and the subfield is drawn while the stage 3 is continuously moved in one direction. When drawing of one subfield is completed in this way, the process proceeds to drawing of the next subfield.

Further, when the drawing of a frame which is a set of a plurality of sub-fields is completed, the stage 3 is moved stepwise in a direction orthogonal to the continuous movement direction, and the above processing is repeated to sequentially draw each frame area. There is.

Here, the frame is a strip-shaped drawing area determined by the deflection width of the main deflector 15, and the subfield is a unit drawing area determined by the deflection width of the sub-deflector 16. On the other hand, the control computer 20 uses the magnetic disk 2 which is a storage medium.
The drawing data of the mask recorded in No. 1 is read, and the read drawing data is temporarily stored in the pattern memory 22 for each frame area.

The drawing data for each frame area stored in the pattern memory 22, that is, the frame information composed of the drawing position and the drawing figure data is sent to the oblique line angle recognizing section 31.

The slanted line angle recognition unit 31 detects a slanted line portion in the drawing data, and adds angle information to the drawn data related to this slanted line portion. The drawing data to which the angle information is added is sent to the drawing data correction unit 32. In the present embodiment, the oblique line angle recognition unit 31 and the drawing data correction unit 3
Although the two are separated, they may be integrated.

The drawing data correction unit 32 accesses the magnetic disk 21 via the control computer 20 and refers to the irradiation amount for correcting the dimensional error caused depending on the oblique line angle.

Here, the corrected irradiation amount is previously determined by experiment and is stored as a table inside the magnetic disk 21. Therefore, information regarding the corrected irradiation amount is added to the drawing data in the shaded area.

Drawing data (drawing data relating to the shaded area,
(Drawing data relating to the non-hatched portion) is passed through the pattern data decoder 23 and the drawing data decoder 24, which are data analysis units, to the blanking circuit 25, the beam shaper driver 26, the main deflector driver 27, and the sub deflector driver. Sent to 28.

That is, the pattern data decoder 23 creates blanking data based on the drawing data, and the blanking data is used as the blanking circuit 2.
Sent to 5. Further, desired beam size data is created based on the drawing data, and this beam size data is sent to the beam shaper driver 26.

Then, a predetermined deflection signal is applied from the beam shaper driver 26 to the beam size varying deflector 14 of the electron optical system 10, whereby the size of the electron beam is controlled.

In the drawing data decoder 24, subfield positioning data is created based on the drawing data, and this subfield positioning data is sent to the main deflector driver 27.

Then, a predetermined deflection signal is applied from the main deflector driver 27 to the main deflector 15 of the electron optical system 10,
As a result, the electron beam is deflected and scanned at a designated subfield position.

Further, the drawing data decoder 24 generates a control signal for sub-deflector scanning based on the drawing data, and this control signal is sent to the sub-deflector driver 28.

Then, a predetermined sub-deflection signal is applied from the sub-deflector driver 28 to the sub-deflector 16, whereby the inside of the sub-field is drawn. As described above, according to the present embodiment, the oblique line angle recognition unit 31 and the drawing data correction unit 32 perform correction according to the angle on the drawing data, and use the corrected drawing data to approximate the oblique line. In other words, even if the division width is not reduced, in other words, the same number of drawing unit figures (diagonal division number) as in the related art can improve the dimensional accuracy of a fine diagonal pattern having an arbitrary angle. Further, since it is not necessary to reduce the division width at the time of the diagonal approximation, there is no problem that the drawing speed (throughput) is reduced.

In this embodiment, the irradiation amount is changed in order to correct the dimensional error caused by the oblique line angle, but the design size and the shot position may be changed. (Second Embodiment) A feature of the electron beam drawing method of the present embodiment is that drawing is performed using drawing data in which design dimensions are corrected so that actually drawn diagonal patterns have predetermined dimensions. Especially. That is, the dimensional error caused according to the angle of the diagonal line is reflected in the drawing data as the design dimension.

FIG. 2 shows a schematic view of an electron beam drawing apparatus used in this embodiment and the subsequent embodiments. This electron beam drawing apparatus has a structure in which the angle recognition unit 31 and the drawing data correction unit 32 are removed from the device of the first embodiment, and represents the structure of a conventional device. However,
The drawing data stored in the disk 21 is different from the conventional one.

FIG. 3 shows a data generation process for performing drawing processing. Design pattern data (LSI data) is designed by a CAD system, this LSI data is converted into drawing data by a host computer, and this drawing data is read and electron beam drawing is performed.

The LSI data designed by the CAD system is usually composed of several polygons,
Furthermore, the data system is such that there are connections between patterns.

Therefore, an LSI having such a data system
In order to make the data usable by the electron beam drawing apparatus (drawing data), first, the host computer performs contouring processing of the figure to remove the multiple exposure area of the beam.

Next, the host computer detects the shaded portion in the LSI data to obtain the shaded angle. Since the host computer prepares the relationship between the slanted line angle and the dimension error as a table in advance, the design dimension is changed by an amount necessary for reducing the dimension error by referring to the table, and converted into the drawing surface data.

FIG. 4 shows a process of converting the LSI data into drawing data. FIG. 4A shows LSI data of a pattern having a hatched portion which has been subjected to contour processing. In the figure, d _o is the design dimension, and θ is the angle of the diagonal line.

Here, since d _o is 2 μm and θ is 30 °, the amount 0.04 μm, which is reduced in advance from FIG. 16, is Δd.
As a result, one side is thickened by Δd / 2. For oblique line angles that do not exist in the table, data is appropriately interpolated from the values before and after the oblique line angle to obtain the dimensional error, and Δd is determined.

As a result, as shown in FIG. 4B, intermediate data in which the dimensions are changed only in the shaded portion is generated.
After that, as shown in FIG. 4C, the drawing is divided into basic figures consisting of rectangles and triangles as in the conventional case, and further the inside thereof is divided into drawing unit figures according to a conventional drawing unit figure dividing algorithm to create drawing data. . That is, after the intermediate data is generated, it is the same as the fourth conventional technique.

According to this embodiment, as the drawing data,
Since the data in which the design dimensions of the shaded area have been corrected to reduce the error dimension is used, it is possible to draw a pattern with an arbitrary shaded angle with the same number of divisions as before, that is, without lowering the throughput. It will be possible to perform with higher accuracy than before. (Third Embodiment) The electron beam writing method of this embodiment is characterized in that writing is performed using writing data in which the irradiation amount is corrected so that the hatched pattern actually drawn has a predetermined size. Especially. That is, the dimensional error that occurs depending on the angle of the oblique line is reflected in the drawing data as the irradiation amount.

In this embodiment, the host computer of FIG.
When converting the SI data into the drawing data, the relationship between the oblique line angle and the dimension error and the relationship between the variation of the irradiation amount and the accompanying dimension error are held in advance as a table. In other words, the relationship is dependent on the oblique line angle. Prepare the amount of change in the dose required to reduce the dimensional error as correction data, and when the shaded area and its shade angle are detected, change the dose by referring to the table above and convert it to drawing data. .

FIG. 5 shows a process of converting the LSI data into drawing data. FIG. 5A shows LSI data of a pattern having a hatched portion which has been subjected to contour processing. In the figure, d _o is the design dimension, and θ is the angle of the diagonal line. FIG. 5B shows a basic figure composed of triangles and rectangles with the design dimensions kept unchanged. At this stage, the triangular basic figure and the rectangular basic figure included in the slanted line are labeled as the basic figure forming the slanted line with an angle of θ, which is different from other horizontal and vertical directions. , The angle-dependent dose is assigned as shown in FIG. Then, the inside of each basic figure is divided into drawing unit figures according to the conventional drawing unit figure dividing algorithm, and drawing data is created.

According to this embodiment, as the drawing data,
Since the data is obtained by correcting the irradiation amount in the shaded area so that the error size becomes smaller, the same number of divisions as before can be used.
In other words, it is possible to draw a pattern having an arbitrary slanted line angle with higher accuracy than before without lowering the throughput. (Fourth Embodiment) In this embodiment as well, drawing is performed using drawing data in which the irradiation amount is corrected so that the diagonal pattern actually drawn has a predetermined size, but the third embodiment The difference is that only for the drawing unit graphic included in the diagonal design line, the irradiation amount is changed according to the diagonal angle.

Further, the present embodiment is different from the second and third embodiments in the method of drawing data generation. That is, in the second and third embodiments, the pattern data decoder 23 performs the step of supplying the basic figure to the magnetic disk 21 and dividing the basic figure into the drawing unit figure. The data is divided into drawing unit figures by a computer, and the data is then recorded on the magnetic disk 21.
F is supplied and drawn.

FIG. 6 shows a schematic diagram of a drawing unit figure generated by dividing the basic figure. Approximately diagonal design line 42
On the other hand, the drawing unit figure is divided so that the areas of the uneven portions are equal.

The host computer holds in advance as a table the relationship between the oblique line angle and the dimensional error and the relationship between the variation of the irradiation dose and the dimensional error accompanying it, in other words, the dimensional error depending on the oblique line angle is stored. The amount of change in the irradiation dose required to reduce the size is prepared as correction data, and when the drawing unit figure 41 (hatched drawing unit figure) including the diagonal design line 42 is detected, refer to the above table. The amount of change in the dose required to reduce the error dimension is added to the drawing unit graphic as information. Therefore, only the drawing unit figure included in the hatched design line 42 can be drawn by changing the irradiation amount. In this embodiment, the same effect as the third embodiment can be obtained. In the figure, reference numeral 40 denotes a drawing unit figure including the design line 42. (Fifth Embodiment) A feature of this embodiment is that drawing is performed using drawing data in which a drawing position (shot position) is corrected so that a diagonal pattern actually drawn has a predetermined size. It is in. More specifically, the drawing position is changed according to the oblique line angle only for the drawing surface unit graphic included in the oblique line design line. The drawing data generation process conforms to that of the fourth embodiment.

FIG. 7A is a schematic diagram when the diagonal pattern is divided into drawing unit figures. The approximation is divided by the drawing unit figure so that the area of the concavo-convex portion is equal to the hatched design line 45.

The host computer has a table in advance which holds the relationship between the oblique line angle and the dimensional error and the relationship between the change in the drawing position of the drawing unit figure and the dimensional error. In other words, the dimensional error depending on the oblique line angle is stored. The amount of change in the drawing position of the drawing unit graphic required to reduce the size is prepared as correction data, and when the drawing unit graphic 44 including the diagonal design line 45 is detected, only δ is shown in FIG. Changing the drawing position as shown is added as information to the drawing-unit graphic. Therefore, only the drawing unit figure included in the diagonal design line 45 can be drawn by changing the drawing position. In this embodiment, the same effect as the third embodiment can be obtained. In the figure, 43 indicates a drawing unit graphic not including the design line 45.

Here, a gap or overlap corresponding to δ is generated by drawing the drawing unit graphic 46 (hatched drawing unit graphic) whose drawing position has changed. Exists in
When the drawing pattern is developed, its effect is not seen at all.

The present invention is not limited to the above embodiment. For example, in the above-mentioned embodiment, the electron beam writing method is described as an example, but the charging beam is not limited to the electron beam, and the present invention can be applied to a charged beam writing method including an ion beam.

As for the drawing method, the present invention can be applied not only to the two-step deflection method in which the main and sub-deflection are combined, but also to the one-step deflection method. The present invention can also be applied.

The present invention can be applied to not only mask patterns but also general LSI patterns. Further, the shape of the beam forming aperture is not limited to that of the above embodiment, and can be changed as appropriate. In addition, various modifications can be made without departing from the scope of the present invention.

[0073]

As described above in detail, according to the present invention, the correction of the dimensional error depending on the oblique line angle is reflected in the drawing data to improve the dimensional accuracy of the oblique line pattern. Since there is no need to reduce the size, it is possible to draw the diagonal pattern with high accuracy without lowering the throughput.

[Brief description of drawings]

FIG. 1 is a schematic diagram showing a schematic configuration of an electron beam drawing apparatus according to a first embodiment of the present invention.

FIG. 2 is a schematic diagram showing a schematic configuration of an electron beam drawing apparatus used in an electron beam drawing method according to a second embodiment or later of the present invention.

FIG. 3 is a diagram showing a drawing data generation process.

FIG. 4 is a diagram for explaining an electron beam drawing method according to a second embodiment of the present invention.

FIG. 5 is a diagram for explaining a drawing beam drawing method according to a third embodiment of the present invention.

FIG. 6 is a diagram for explaining a drawing beam drawing method according to a fourth embodiment of the present invention.

FIG. 7 is a diagram for explaining a drawing beam drawing method according to a fifth embodiment of the present invention.

FIG. 8 is a diagram showing a rectangular aperture according to a second conventional technique.

FIG. 9 is a diagram showing a drawing pattern according to a second conventional technique.

FIG. 10 is a diagram for explaining the problem of the second conventional technique.

FIG. 11 is a diagram showing a beam shaping aperture in a third conventional technique.

FIG. 12 is a diagram showing a drawing pattern according to a third conventional technique.

FIG. 13 is a diagram for explaining an approximation method of oblique line division according to a fourth conventional technique.

FIG. 14 is a diagram for explaining an approximation method of oblique line division according to a fourth conventional technique.

FIG. 15 is a diagram showing a pattern used to evaluate the diagonal dimension accuracy of an electron beam drawing apparatus for manufacturing an LSI mask.

16 is a diagram showing an evaluation result of diagonal line dimension accuracy performed using the pattern of FIG.

[Explanation of symbols]

 DESCRIPTION OF SYMBOLS 1 ... Sample chamber 2 ... Sample 3 ... Stage 4 ... Stage drive circuit 5 ... Position detection circuit 6 ... Electron gun 7,8,9,11,12 ... Various lenses 10 ... Electron beam optical system 13 ... Blanking deflector 14 Beam deflecting deflector 15 Beam scanning main deflector 16 Beam scanning sub-deflector 17, 18 Beam shaping aperture 20 Control computer 21 Magnetic disk 22 Pattern memory 23 Pattern data decoder 24 Drawing data decoder 25 ... Blanking circuit 26 ... Beam shaper driver 27 ... Main deflector driver 28 ... Sub deflector driver 31 ... Oblique line angle recognition unit (data correction means) 32 ... Drawing data correction unit (data correction means) 40 ... Drawing unit graphic not including design line 41 ... Drawing unit graphic including design line 42 ... Design line 43 ... Design line Drawing unit figure 45 ... design line that contains the drawing unit figures 44 ... design line that does not include the emissions

Claims (5)

[Claims] 1. A means for forming a plurality of kinds of variable shaped beams required for drawing the pattern from a charged beam based on drawing data of a pattern to be drawn on the sample, and the sample using these variable shaped beams. A charged beam drawing apparatus comprising: a unit for drawing the pattern above; a data correction unit that detects an angle of a hatched pattern existing in the drawing data and applies a correction corresponding to the angle to the drawing data. And a charged beam drawing apparatus, which draws the pattern using the corrected drawing data obtained by the data correction unit. 2. A plurality of kinds of variable shaped beams required for drawing the pattern are formed from a charged beam based on drawing data of a pattern to be drawn on the sample, and the variable shaped beams are used to form on the sample. In the charged beam drawing method for drawing the pattern, the data related to the oblique line pattern existing in the drawing data is corrected data corresponding to the angle of the oblique line pattern is used. 3. The data related to the diagonal pattern is used in which the design dimension of the diagonal pattern is corrected corresponding to the angle so that the diagonal pattern actually drawn has a predetermined dimension. The charged particle beam drawing method according to claim 2. 4. The data relating to the diagonal pattern is data in which the irradiation amount of the variable shaping beam is corrected corresponding to the angle so that the diagonal pattern actually drawn has a predetermined size. The charged particle beam drawing method according to claim 2, wherein 5. The data relating to the diagonal pattern is used in which the shot position of the variable shaping beam is corrected corresponding to the angle so that the diagonal pattern actually drawn has a predetermined size. The charged particle beam drawing method according to claim 2, wherein