JPH0521323A - Charged particl beam drawing device - Google Patents

Abstract

PURPOSE:To provide a charged beam drawing device which can well correct secondary or more misregistration caused by strain of a mirror, strain of a substrate, etc., of a laser length measurement system and improve pattern generation accuracy. CONSTITUTION:A charged beam drawing device which generates a desired pattern on a sample by irradiating with charged beam the sample, is provided with a memory 39 which stores high-order function (position correction function data) 47 of one parameter alone to an arbitrary position on the sample, an origin position correction circuit 36 which carries out correction operation of an origin position referring to the position correction function data 47 stored in the memory 39 for each sub-field and a deflection control circuit 33 which changes a drawing position in accordance with operation results obtained by the circuit 36.

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 for drawing a fine pattern of a semiconductor integrated circuit or the like using a charged beam, and more particularly to a function for changing and correcting a drawing position of drawing data of a desired pattern to be drawn. The present invention relates to a charged beam drawing apparatus having the same.

[0002]

2. Description of the Related Art In recent years, various electron beam drawing apparatuses have been used for forming a fine pattern on a sample such as an integrated circuit pattern transfer mask and a semiconductor wafer. In an electron beam drawing apparatus, an electron beam emitted from an electron gun is focused and accelerated to irradiate a sample placed on a stage, a deflection system scans the electron beam on the sample, and a blanking system is used. A desired pattern is drawn by turning the beam on and off.

However, this type of device has the following problems. That is, an error occurs in the position measurement due to the tilt or distortion of the mirror of the laser measuring system for measuring the drawing position mounted on the stage on which the sample is placed, and the pattern actually drawn on the sample is the desired drawing. It becomes tilted or distorted compared to the position. The position measurement error caused by the tilt of the mirror makes a primary contribution to the deviation of the drawing position on the sample. Further, the position measurement error due to the distortion of the mirror makes a secondary or higher contribution to the deviation of the drawing position on the sample. These deviations are close to 1 μm in a large area and are not allowed not only in the case of direct writing but also in the case of quintuple mask writing.

When supporting a mask or the like with a cassette,
Deflection of the supported substrate causes pattern distortion. The distortion of this pattern also becomes a function of the second or higher order of the drawing position. Conventionally, sufficient correction has been made for the primary positional deviation, but sufficient correction has not been made for secondary or higher positional deviation.

[0005]

As described above, in the related art, the drawing position measurement error occurs due to the tilt and distortion of the mirror of the laser measuring system, the deflection of the cassette, etc., and this is a major factor that reduces the drawing accuracy. It was. And 2
It has been difficult to make sufficient corrections for positional deviations of the following and above.

The present invention has been made in consideration of the above circumstances, and an object thereof is to sufficiently prevent a positional deviation of a second order or more due to distortion of a mirror of a laser measuring system or distortion of a substrate. It is an object of the present invention to provide a charged beam drawing apparatus that can be corrected and can improve drawing accuracy.

[0007]

The gist of the present invention is to represent the amount of distortion at an arbitrary position as a higher-order function of only one variable, and to use it as a correction function to obtain a corrected drawing position using that correction function. It is in.

That is, according to the present invention, in a charged beam drawing apparatus for irradiating a charged beam onto a sample to draw a desired pattern on the sample, a memory for storing position correction function data for an arbitrary position on the sample and each drawing. The present invention is characterized in that means for correcting the drawing position by referring to the function data stored in the memory at the position and means for changing the drawing position according to the calculation result obtained by the means are provided.

More specifically, according to the present invention, the distortion of the pattern of the sample drawn without correction is divided into an x component and ay component, and a function f (y) of only y in the x direction and a distortion in the y direction are obtained. It is possible to obtain the drawing data in which the drawing position is corrected by expressing it as a function g (x) of only x and performing the correction calculation with respect to the origin of the small area (subfield) as a correction function. Is.

[0010]

According to the present invention, the distortion of the drawing pattern caused by the distortion of the mirror of the laser measuring system for measuring the drawing position and the distortion of the drawing pattern caused by the deflection of the substrate supported by the cassette are eliminated. Is expressed as a correction function for, for example, a function value is obtained at the origin position of the subfield, the origin position is changed according to the correction formula, and the pattern is drawn according to that, so that a pattern without distortion can be obtained.

A method of correcting the drawing pattern distortion caused by the distortion of the laser measuring system mirror will be described. FIG. 1 shows drawing data and a drawing result when no correction is applied. The deviation at an arbitrary position (x _i , y _i ) on the substrate is shown in FIG.
F (y _i) is the x-direction as for the y-direction represents a g (x _i).

Therefore, an arbitrary position (x _O ,
y _O ) is displaced by f (y _O ) in the x direction and g (x _O ) in the y direction. Therefore, (x _O + f (y _O ), y _O + g (x _O ))
Is drawn at the position. Here, it is assumed that f (y) and g (x) are each represented by a function of quadratic or higher in y and x. The position to be actually drawn (x ', y') _{if, x '= x O + f} (y O) y' = y O + g (x O) since it is ‥‥ (1), actually drawn When the drawing data position for the position to be changed to (x _O , y _O ) is (x ″, y ″), x _O = x ″ + f (y ″) y _O = y ″ + g (x ″). It becomes (2). Therefore, the correction formula is f (x ″) to f (y ₀ ), g (x ″), assuming that the function value is constant in the small area.
Because made ~g (x _0), is given by _{x "= x O -f (y} O) y" = y O -g (x O) ‥‥ (3). By changing the position (x _O , y _O ) of the drawing data to (x ″, y ″), the position is actually drawn on the substrate at the position (x _O , y _O ).

Next, a method of correcting the distortion of the drawing pattern due to the bending of the substrate supported by the cassette will be described.
In many cases, the distortion caused by the bending of the substrate is distorted in mirror symmetry as shown in FIG. When distorted as shown in FIG. 3, a mirror surface exists at the position shown in FIG. Due to this distortion, displacement on the mirror surface occurs only in one direction of x or y. Further, there is no displacement at the point where the mirror surfaces overlap.

Utilizing the above-mentioned distortion property, I of FIG.
It suffices to make the correction only within the area of. Supplement
The form is given by the above equation (3), so that
Point (x_I, Y_I) Correction value is x_I′ = X_I-F (y_I) y_I′ = Y_I-G (x_I) ... (4) Becomes The point on the mirror surface that forms the boundary between I and II is x_{I, II}′ = X_{I, II}-F (y_{I, II}) y_{I, II}′ = Y_{I, II}                      ‥‥ (Five) Becomes The point on the mirror surface that bounds I and III is x_{I, III}′ = X_{I, III}     y_{I, III}′ = Y_{I, III}-G (x_{I, III}) ‥‥‥ (6)

[0015] Also, points (x _II , y _II ), (x _III , y _III ), (x _IV , y _IV ) in the regions II, III, IV.
Is first mirror-reversed and moved into the region I, then corrected according to the equation (4) and returned to the original region. Figure 5
The flow is shown in.

That is, when (x _II , y _II ) is transferred to I (x _II , -y _II ), the correction value of this is (x _II -f (-
y _II ), −y _II −g (x _II )), and when transferred to the region II again, it becomes (x _II −f (−y _II ), y _II + g (x _II )). By performing the correction calculation as described above, it is possible to obtain a pattern with sufficient accuracy without distortion.

[0017]

The details of the present invention will be described below with reference to the illustrated embodiments.

FIG. 1 is a schematic configuration diagram showing an electron beam drawing apparatus according to an embodiment of the present invention. In the figure, 10 is a sample chamber, and a sample table 12 on which a sample 11 such as a semiconductor wafer is placed is accommodated in the sample chamber 10. Sample table 1
2 is a sample stage drive circuit 3 which receives a command from the computer 30
1 moves in the X direction (left and right direction on the paper surface) and the Y direction (front and back direction on the paper surface). The moving position of the sample table 12 is measured by the laser length measurement system 32 and the mirror 37, and the measurement information is sent to the computer 30 and the deflection control circuit 33.

On the other hand, above the sample chamber 10, the electron gun 2
1, various lenses 22a-22e, various deflectors 23-26
And an electron optical lens barrel 20 including beam forming aperture masks 27a and 27b. The deflector 23 is a blanking deflector for turning the beam on and off, and a blanking signal from a blanking control circuit 34 is applied to the deflector 23. Deflector 2
Denoted at 4 is a beam size varying deflection plate which variably controls the beam size by utilizing the optical aperture overlap of the aperture masks 27a and 27b. The deflector 24 receives a deflection signal from the variable beam size control circuit 35. Is applied. The deflectors 25 and 26 are beam scanning deflection plates that scan the beam on the sample 11.
A deflection signal is applied from the deflection control circuit 33.

The deflector 25 is a main deflector which largely deflects the beam on the sample, and the deflector 26 is a sub deflector which deflects the beam slightly on the sample. Then, the beam position is determined by the main deflection plate 25, and a small region (subfield) in the deflectable region of the deflection plate is drawn by the sub deflection plate 26.

Next, a drawing method using the apparatus of this embodiment, especially a method of correcting the drawing position will be described. The drawing position correction function is defined as a function f (y) of y for a shift in the x direction and a function g (x) of x for a shift in the y direction, as shown in FIG. The measured value is compared with the drawing data and is determined as a higher-order (second-order or higher) function by mathematical means such as the least square method. The f (y) and g (x) are stored in the buffer memory 39 as function form and coefficient data.

The correction of the drawing position is applied to the origin of the subfield shown in FIG. The origin of the subfield 2 of the drawing data is defined with respect to the origin of the frame 1, and the origin of the frame 1 is defined with respect to the origin on the drawing sample. Although each figure is defined with respect to the origin of the subfield 2, the position of the figure is not a correction target. The reason for this will be described with reference to FIG. In the case of FIG. 8, it is assumed that the distortion correction function is expressed by the following equation in each of the x and y directions. f (y) = ^-4 × 10 ^-11 × y ² +0.1 (x direction) g (x) = ^-4 × 10 ^-11 × x ² +0.1 (y direction)

In the above equation, y, f (y), x, g (x)
Are all defined in μm units. If the size of the subfield is 30 × 30 μm, the rate of change of the correction function within the subfield is 10 ⁻⁶ μ even at the maximum.
It can be seen that the order is m and it does not make sense to correct the figure position in the subfield. Therefore, the correction symmetry is limited to the origin of the subfield.

The flow of drawing position correction will be described below with reference to FIG. First, the CAD data 41 is converted into the EB drawing data 43 by the data conversion computer 42 and dropped on the magnetic tape. The drawing data on the magnetic tape is read by the control computer 30 and stored in the magnetic disk 44.
It is stored in the main memory 45 of the control computer 30. Then, the drawing data 46 and the origin position correction function data 47 are stored in the buffer memory 39. The drawing data 46 stored in the buffer memory 39 is expanded by the data expanding circuit 38 into the drawing data of the apparatus.

Here, as the origin position correction function data 47, a pattern is drawn without correction in advance, the formed pattern is measured to obtain the distortion at each position, and a primary variable described later is obtained from the distribution of the distortion. Shall define a higher-order function of This setting is for the device (mirror 3
(7, cassette, etc.) does not change, it is sufficient to perform only the first time.

In the data expansion circuit 38, in the main deflection expansion circuit 48, the origin coordinates of the subfield defined for the frame in the drawing data 46 are made into absolute coordinates by the main deflection position calculation, and are defined as the main deflection position 55. . Further, the sub-deflection expansion circuit 49 converts shot data (beam position, beam shape size,
The shot time) is calculated, and the sub-deflection position 53, beam shape, size, and shot time 54 are obtained. The sub-deflection position 53 of the shot data, the beam shape and size, and the shot time 54 are subjected to predetermined processing by the figure dividing circuit 50, the mirror inversion scaling circuit 51, and the sorting circuit 52.

The sub-deflection position 53 of the shot data, the beam shape and size, and the shot time 54 are sent to the deflection control circuit 33 as they are. On the other hand, the main deflection position 55 is sent to the origin position correction circuit 36. Then, the origin position correction circuit 36
Refers to the position correction function data 47 stored in the buffer memory 39 to obtain the function value for the main deflection position, that is, the origin of the subfield, and then performs the correction calculation. The corrected main deflection position is sent to the deflection control circuit 33.

Next, details of the processing in the origin position correction circuit 36 will be described with reference to FIG. The origin position correction circuit 36 receives the main deflection position data 55 output from the data expansion circuit 38, that is, the origin position data of the subfield. Then, referring to the position correction function data 47 stored in the buffer memory 39, f (y) is set with respect to the origin (x _Ni , y _Ni ) of the subfield to be corrected.
_Ni), seeking g (x _Ni), subjected to correction operation by _{x Ni '= x Ni -f (} y Ni) y Ni' = y Ni -g (x Ni) ‥‥ (7). When the origin position (x _Ni , y _Ni ) of the subfield is rewritten to (x _Ni ′, y _Ni ′), the process for the next subfield is started.

The above description is mainly for the case of correcting the pattern distortion due to the distortion of the mirror in the laser measuring system, but the same can be applied to the pattern deflection due to the mask support of the cassette. . In this case, the position of the mirror surface also includes the correction functions F (y) and G (x).
In the same way as the above, the relationship between the origin position of the subfield to be corrected and the mirror surface position is checked, the correction functions F (y) and G (x) are subjected to inversion calculation, etc., and then (7 ) Perform the same correction calculation as the formula) and rewrite the origin position of the subfield. For example, a point (x ₃ ,
y ₃₎ are mirror-m _x, the relationship between the m _y, the correction function defined in the region I -F (y), - change the G (x), mirror surface (x _03, y ₀₃₎ The coordinate values (x ₃ ′, y ₃ ′) when m _x and m _y are the x and y axes are changed. _{x 3 "= x 3 '-} (- F (y 3')) y 3" = y 3 '- (- G (x 3')) ‥‥ (8)

After the correction calculation is performed, (x ₃ ″,
y ₃ ″) to the original absolute coordinates (x ₀₃ ″, y ₀₃ ″),
Change the subfield origin position from (x ₀₃ , y ₀₃ ) to (x ₀₃ ″, y ₀₃ ″). In addition, (x ₀₄ ,
y ₀₄ ) is converted into (x ₄ ′, y ₄ ′) and is corrected by x ₄ ″ = x ₄ ′ y ₄ ″ = y ₄ ′ − (− G (x ₄ ′)) (9) Then, (x ₄ ″, y ₄ ″) is returned to (x ₀₄ ″, y ₀₄ ″) in the original absolute coordinates, and the origin position of the subfield is rewritten.

Further, in the above case, if the correction function is switched depending on the cassette number and the type of sample (size, thickness, etc.), more detailed correction can be performed according to each distortion.

By obtaining the corrected position by the origin position correction circuit 36 and drawing the corrected position as a reference, the pattern actually obtained becomes free of distortion as shown in FIG. In FIG. 12, (a) shows drawing data, (b) shows drawing results, a solid line shows a corrected pattern, and a broken line shows a non-corrected pattern.

As described above, according to the present embodiment, the distortion of the pattern of the sample drawn without correction is divided into the x component and the y component, and the function f (y) of only y in the x direction and the y direction. Is expressed as a function g (x) of only x. And
By performing a correction calculation on the origin of the subfield using these functions as a correction function, it is possible to obtain drawing data in which the drawing position is corrected. Therefore, it is possible to surely correct the drawing position measurement error, particularly the drawing position measurement error of the second order or higher, which is caused by the distortion of the mirror of the laser measuring system and the bending of the substrate due to the substrate support in the cassette. Therefore, the drawing accuracy can be improved, and its usefulness is great.

The present invention is not limited to the above embodiment. In the embodiment, the unit for correcting the drawing position is one subfield. However, when the distortion of the mirror or the deflection of the substrate is small, the position correction may be performed for each of the plurality of subfields. Further, the present invention is not limited to the two-stage deflection system including the main deflector and the sub-deflector, but can be applied to an apparatus having only the main deflector. Further, it can be applied to an ion beam drawing apparatus using an ion beam instead of an electron beam. In addition, various modifications can be made without departing from the scope of the present invention.

[0035]

As described above in detail, according to the present invention, position correction function data (higher-order function of only one variable) for an arbitrary position is stored in the memory, and this function data is stored at each drawing position. Since the drawing position is corrected with reference to change the drawing position according to the calculation result, it is possible to sufficiently correct the positional deviation of the second or higher order due to the distortion of the mirror of the laser measuring system or the distortion of the substrate. Can be corrected to
It is possible to realize a charged beam drawing apparatus that can improve drawing accuracy.

[Brief description of drawings]

FIG. 1 is a schematic diagram showing distortion of a pattern caused by distortion of a mirror of a laser measuring system,

FIG. 2 is a schematic diagram for explaining how to define a correction function,

FIG. 3 is a schematic diagram showing the distortion of the pattern due to the deflection of the substrate,

FIG. 4 is a schematic diagram showing a method for correcting pattern distortion due to bending of a substrate,

FIG. 5 is a schematic diagram showing a method for correcting pattern distortion due to bending of a substrate,

FIG. 6 is a schematic configuration diagram showing an electron beam drawing apparatus according to an embodiment of the present invention,

FIG. 7 is a schematic diagram showing the relationship between frames and subfields,

FIG. 8 is a schematic diagram showing an example of actual pattern distortion,

FIG. 9 is a circuit configuration diagram schematically showing the flow of drawing data.

FIG. 10 is a flowchart showing a processing flow of an origin correction circuit,

FIG. 11 is a schematic diagram for explaining a process of correcting the pattern distortion due to the deflection of the substrate,

FIG. 12 is a schematic diagram showing drawing data and a drawing result.

[Explanation of symbols]

1 ... frame, 2 ... subfield, 21 ... electron gun, 22a to 22e ... various lenses, 23-26 ... Various deflectors, 27a, 27b ... aperture mask, 30 ... control computer, 31 ... Sample stage drive circuit, 32 ... Laser length measurement system, 33 ... Deflection control circuit, 34 ... Blanking control circuit. 35 ... Variable shaped beam size control circuit, 36 ... Origin position correction circuit, 37 ... Mirror, 38 ... Data expansion circuit, 39 ... Buffer memory.

Claims (3)

[Claims] 1. A charged beam drawing apparatus for irradiating a sample with a charged beam to draw a desired pattern on the sample, comprising: a memory for storing position correction function data for an arbitrary position on the sample; A charged beam comprising: a means for performing a correction calculation of a drawing position by referring to the function data stored in the memory; and a means for changing the drawing position according to a calculation result obtained by the means. Drawing device. 2. The position correction function data stored in the memory is represented by a function of only y in the x direction and a function of only x in the y direction in the coordinate system on the sample. Charged beam drawing apparatus as described. 3. In a two-stage deflection system in which a small area in a drawing area is positioned by main deflection and drawing is performed in the small area by sub-deflection, correction operation is performed on drawing position data of main deflection,
2. The origin position of the small area is corrected.
Charged beam drawing apparatus as described.