JP2000100708A - Method of electron beam exposure - Google Patents

Abstract

PROBLEM TO BE SOLVED: To obtain a high overlay accuracy by re-sizing a pattern, based on deviation information between unit exposure regions in a data conversion in an electron beam exposure system, when there is a poor connection occurs due to the lens distortion. SOLUTION: When no connection part of a hole pattern, etc., is provided on the boundary line of a sub-deflection region under inspection (S11 No), a measured position correction is superposed on a data at each sub-deflection region as it is to end a data conversion (S14). If a line type pattern such as gates exists on the sub-deflection boundary line of the sub-deflection region (S11 Yes), since a pattern exists among a plurality of sub-deflection regions, the positional correction quantity difference between the sub-deflection regions among which the pattern exists is obtd. (S12). A connection part of the pattern existing among the sub-deflection regions is re-sized so as to compensate the position correction quantity (S13).

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, and more particularly, to a method of manufacturing a semiconductor device by using an optical exposure system and an electron beam exposure system together. The present invention relates to a pattern dimension correction method for correcting a displacement of a base pattern caused by lens distortion in a typical exposure system.

[0002]

2. Description of the Related Art Conventionally, in a method of manufacturing a semiconductor device, an optical exposure system and an electron beam exposure system have been generally used. The electron beam exposure system has a high resolution, but has a drawback that the throughput is low because the pattern is sequentially processed and drawn.

On the other hand, in an optical lithography system using a light source such as i-line or KrF, that is, an optical exposure system, the batch transfer using a mask (reticle) has high throughput, but the resolution is low. It has the disadvantage of being limited by wavelength. Therefore, in order to achieve both high resolution of the electron beam exposure system and high productivity of the optical lithography system, a so-called MIX & MATCH technology using both the electron beam exposure system and the optical exposure system has been proposed.

Such a technique is disclosed, for example, in Japanese Patent Publication No. 3-4552.
Publication No. 7, T. Ohiwa et al. Ext. Abstr. 17th Con
f. Solid State Devices and Materials (1985) 345
Or Y. Gotoh et al. Jpn. J. Appl. Phys. 12B (199
7) It is disclosed in 7541 and the like. In such a conventional technique, an electron beam exposure system is used for a process with strict rules, and an optical lithography system is used for a process with relatively loose rules. Further, in order to secure a high overlay accuracy between the two, the pattern displacement mainly caused by the lens distortion of the optical stepper is corrected by an electron beam exposure apparatus.

Here, regarding the configuration of the above-mentioned conventional example,
This will be described in detail below. That is, FIG. 6 shows a flowchart for explaining the above-mentioned conventional electron beam exposure method technology. First, after the start, in step (1), a measurement pattern is exposed on a wafer by a stepper.

More specifically, as shown in FIG. 7, a test mark which can be measured by an appropriate electron beam exposure apparatus or an appropriate coordinate measuring device, that is, a pattern in which position measurement marks M are arranged at an appropriate pitch is used. Then, an arbitrary sample 1 is exposed using an appropriate stepper for measuring lens distortion. FIG. 7A is an overall view of a sample 1 to be tested, and FIG. 7B is a plan view showing an example of a test pattern 2 provided in a unit exposure area S of the sample 1. It is.

The mark group M on the pattern 2 is desirably arranged over the entire maximum transfer area of the stepper so that distortion over the entire surface of the lens of the stepper can be measured. Next, the process proceeds to step (2), and the position of the mark M on the wafer is measured by the electron beam exposure apparatus or the coordinate measuring means.

That is, the position of each mark exposed on the wafer as the sample 1 is measured by an electron beam exposure apparatus or a coordinate measuring device or the like, and as shown in FIG. The position shift amount of each mark M is obtained from the measurement result 4 indicated by. Next, the process proceeds to step (4), where the primary components such as the shift component, the rotation component, and the magnification component are removed from the measurement result.

That is, from the measurement results, it is possible to remove primary components such as shift components, rotation components, and magnification components that can be corrected by the alignment of the electron beam exposure apparatus. FIG.
Can obtain rotation component information 5 with respect to the ideal grating 3 from the measurement result of FIG. However, since the curve components that cannot be removed from the measurement result of FIG. 8 remain until the end, information 6 as shown by a solid line in FIG. 10 is finally obtained. I do.

Next, proceeding to step (4), the obtained positional deviation data is approximated by a high-order polynomial as a function of the X and Y coordinates in the chip using, for example, the least square method. this is,
This is for obtaining the amount of displacement at each coordinate in the chip, and may be performed by an interpolation method between the fixed points on each side. Next, in step (5), when the pattern data to be drawn is converted into a format for electron beam exposure, the displacement correction data calculated from the polynomial is superimposed on the pattern data.

That is, the unit for superimposing the correction data is for each pattern or for each deflection area of the electron beam exposure apparatus.
Usually, an electron beam exposure apparatus uses multi-stage deflection, and the lower deflection area is about several tens μm square to about 100 μm square.
In the above-mentioned conventional method, the correction amount corresponding to the position coordinates is added to the original position coordinates of each pattern or each continuous deflection region on the electron beam exposure data. .

Therefore, according to the conventional method, it is not necessary to mount a new storage device, a new system such as an adder and application software in the electron beam exposure apparatus itself, and the distortion can be corrected only by a computer used for data conversion. Therefore, it is possible to construct a system at low cost.

[0013]

However, in such a conventional method, when there is a pattern connection between adjacent correction units, that is, between unit exposure areas, the difference between the position correction amounts of the two is corrected. Only the patterns are separated, overlapped, or shifted. Therefore, problems such as disconnection and partial deterioration of dimensional accuracy occur in the wiring pattern.

FIG. 11 illustrates a typical pattern data shift phenomenon at a connection portion. That is, FIG.
When there is a pattern formed straddling between a plurality of unit exposure regions arranged continuously, the data on the case where a positional shift due to lens distortion occurs between the unit exposure regions 3A and 3B show a pattern and a dimensional variation of a pattern obtained with an actual resist, respectively.

That is, FIG. 11A shows an example in which the connection portions of the pattern are separated in the longitudinal direction of the pattern, and FIG. 11B shows a case in which the connection portions of the pattern overlap each other. FIG. 11C shows a case where the connection portion of the pattern is shifted in a direction orthogonal to the longitudinal direction of the pattern, that is, in a lateral direction. FIG.
In the case of (A), when the shift amount increases, the resist pattern is disconnected at the connection portion. On the other hand, FIG.
In the cases of (C) and (C), if the overlap or the lateral displacement is large, the line width increases, or there is a risk of contacting an adjacent line pattern in a displaced state.

The relationship between the deviation between patterns and the change in resist dimensions in the conventional example is generally as shown in a graph before correction (● solid line) in FIG. In FIG.
Investigation was conducted on isolated line-shaped patterns of 0.15 μm, and a minus indicates that the patterns are separated from each other, and a plus indicates that the patterns are overlapped.

As can be seen from the graph, in the portion where the deviation between the unit exposure regions is 15 nm or more, the dimensional fluctuation of the resist exceeds 10% of the designed size. On the other hand, Japanese Patent Application Laid-Open No. H08-15141 describes an electron beam exposure apparatus using a three-stage deflector, but merely discloses a structural improvement, and uses a lens distortion. There is no disclosure or suggestion about a technique for calibrating the displacement in software during data conversion.

Japanese Patent Application Laid-Open No. 3-45527 discloses a technique in which a projection position of an electron beam is corrected by using a distortion amount, which is lens distortion data obtained in advance, to perform exposure. Japanese Patent Application Laid-Open No. Sho 62-57216 discloses a method of correcting a mask or a reticle in consideration of the amount of distortion, but none of the above-mentioned technical ideas of the present invention disclose or suggest. It has not been.

Accordingly, it is an object of the present invention to improve the above-mentioned drawbacks of the prior art and to make no improvement or addition of the device,
It is an object of the present invention to provide an electron beam exposure method capable of performing an efficient electron beam exposure process with high overlay accuracy and little pattern dimensional variation.

[0020]

In order to achieve the above-mentioned object, the present invention employs the following basic technical structure. That is, in a method of manufacturing a semiconductor device by using both an optical exposure system and an electron beam exposure system, a displacement of a base pattern caused by a lens distortion in the optical exposure system is determined by using the electron beam exposure system. When correcting at the time of the data conversion in the case where there is a pattern formed over the individual unit exposure areas that are continuous with each other in the electron beam exposure system, and the continuous pattern is the lens When a connection failure occurs due to distortion, an electron beam exposure method for resizing the pattern based on the displacement information between the unit exposure regions during data conversion in the electron beam exposure system. As a specific example, a semiconductor device is manufactured by using an optical exposure system and an electron beam exposure system together. In correcting the position shift of the underlying pattern caused by the lens distortion in the optical exposure system during data conversion in the electron beam exposure system, When there is a pattern formed straddling between the individual unit exposure regions to be performed, a difference value between the position correction amounts of the individual unit exposure regions that are continuous with each other is calculated, and according to the difference value, An electron beam exposure method for correcting the dimension of a pattern formed over the continuous individual unit exposure areas.

[0021]

DESCRIPTION OF THE PREFERRED EMBODIMENTS That is, the electron beam exposure method according to the present invention employs the above-described technical configuration.
In the electron beam exposure method in which the stepper lens distortion correction is performed at the time of electron beam exposure data conversion, it is possible to perform dimensional correction on a pattern located at a boundary portion between each correction unit by a difference in a position correction amount between each correction unit. I can do it.

[0022]

BRIEF DESCRIPTION OF THE DRAWINGS FIG. 1 is a block diagram showing an embodiment of an electron beam exposure method according to the present invention. That is, as the electron beam exposure method according to the present invention, in a method of manufacturing a semiconductor device by using both an optical exposure system and an electron beam exposure system, the method is performed by lens distortion in the optical exposure system. When correcting the displacement of the underlying pattern during data conversion in the electron beam exposure system, there is a pattern formed over the individual unit exposure regions that are continuous with each other in the electron beam exposure system. And when the continuous pattern has a connection failure due to the lens distortion, the data is converted in the electron beam exposure system based on the displacement information between the unit exposure regions. This is an electron beam exposure method characterized by resizing the pattern, and as a result, it is especially used in the past. High overlay accuracy can be obtained by simply changing the software without structurally improving or adding an electron beam lithography system that uses both a lithography system and an electron beam lithography system. It is possible to obtain an electron beam exposure method capable of performing an efficient electron beam exposure process with little fluctuation.

The connection failure of the pattern used in the present invention means that a plurality of consecutively arranged unit exposure regions, particularly between adjacent unit exposure regions. In the case where there is a pattern that is formed by the above, the continuity of the pattern is broken by the result of the lens distortion in the optical exposure system, and the portions to be continued of the pattern are separated from each other. Or, information in which a part of a pattern formed over an adjacent unit exposure region is overlapped.

Further, in a special case, a case where a predetermined pattern in each of the unit exposure regions is two-dimensionally shifted in the vertical direction and the horizontal direction is also included. Further, the electron beam exposure apparatus used in the present invention is not particularly limited, and any electron beam exposure apparatus can be used. Therefore, for example, the deflector for deflecting the electron beam may be a one-stage electron beam exposure apparatus,
An electron beam exposure apparatus in which a sub-deflector is added to the main deflector in one stage or an electron beam exposure apparatus in which the sub-deflector is added in two or more stages can be advantageously used.

The electron beam exposure method according to the present invention will be described more specifically. The electron beam exposure method according to the present invention basically uses both an optical exposure system and an electron beam exposure system. In the method of manufacturing a semiconductor device, when correcting the position shift of the underlying pattern caused by lens distortion in the optical exposure system, at the time of data conversion in the electron beam exposure system,
In the case where there is a pattern formed over the individual unit exposure regions that are continuous with each other in the electron beam exposure system, the difference value between the position correction amounts of the individual unit exposure regions that are continuous with each other. Is calculated, and in accordance with the difference value, dimensional correction of a pattern formed over the continuous individual unit exposure regions is performed.

In the present invention, in order to obtain a high overlay accuracy when an overlay exposure is performed by an electron beam lithography apparatus on a base exposed using photolithography, a stepper used for the base exposure is used. Lens distortion must be corrected. Therefore, the lens distortion of the stepper is measured in advance, and the position correction amount of each pattern in the chip is obtained. The correction data is used when converting the pattern data into the electron beam exposure data, and the correction amount is superimposed on the original position information for each figure in the electron beam exposure data.

According to the present invention, it is possible to prevent dimensional fluctuation and deformation of the resist pattern due to the distortion correction of the stepper. FIG. 1 is a flowchart showing an operation procedure of a specific example of the electron beam exposure method according to the present invention. In this specific example, a case is considered in which a gate layer is superimposed on an element isolation layer previously exposed to a base by a stepper using an electron beam exposure apparatus including a main deflector and a one-stage sub deflector.

Therefore, the unit in which the predetermined position correction data is superimposed on the predetermined position data in this specific example is the unit exposure area S, that is, the sub deflection area S. First, as shown in the flowchart of FIG. 1, after starting, in step (10), distortion data relating to position information generated by the lens distortion is measured, and each sub-deflection area S is determined based on the measured distortion data. Is calculated.

In the present invention, the position correction amount for each sub-deflection area S is obtained based on the measured distortion data of the stepper in each of the sub-deflection areas S.
Since the position correction amount differs depending on the position in the chip (which position of the lens has been used for exposure), the position correction amount differs between adjacent sub-deflection regions. In particular, like the four corners of the lens distortion data 4 shown in FIG.
2B with respect to the corners of the ideal lattice 3 shown in FIG.
As shown in (C), the difference in the position correction amount becomes large in a portion where the distortion amount is large.

In the present invention, each position correction amount in each of the unit exposure areas is calculated by a predetermined method and stored in a predetermined storage means. Next, proceeding to step (11), it is determined whether there is a pattern having a connection portion at the boundary portion 15 of each sub-deflection region S.

That is, an arbitrary pattern is formed in each of the unit exposure area, that is, the sub deflection area S in FIG. 3, and in the example of FIG.
A desired pattern is formed across a plurality of successively adjacent sub-deflection regions S. A boundary formed between two sub-deflection regions S adjacent to each other is referred to as a boundary portion 15, and particularly, in FIG.
a is formed with a pattern extending over another adjacent sub-deflection area S, and is a boundary line with a pattern. Since the pattern does not exist on the boundary line 15b, a boundary without a pattern is formed. Called line.

That is, in the above example, since the gate extends in the horizontal direction, the pattern straddles the horizontal boundary portion 15a, and there is no pattern at the vertical boundary portion 15b. If it is determined in step (11) that there is no connection such as a hole pattern at the boundary 15 of the sub-deflection area S under inspection, the process proceeds to step (14), and the data of each sub-deflection area is directly measured. The calculated position correction amount is superimposed and the data conversion ends.

However, the sub-deflection boundary line 15 of the sub-deflection region S
If there is a line pattern like a gate in
Since there is a pattern straddling a plurality of sub-deflection areas, the process proceeds to step (12), and the difference in the amount of position correction between the sub-deflection areas S and S 'straddling the pattern is determined. Next, the process proceeds to step (13), where the sub-deflection regions S and S '
Resizing is performed so as to compensate for the difference in the amount of position correction for the connection portion of the pattern extending between them.

In the electron beam exposure method according to the present invention, the resizing of the pattern, that is, the dimensional correction, is performed on the pattern which extends over the unit exposure regions which are continuous with each other, and is separated from each other, or This is to resize the superposed pattern portion. Also, in the present invention,
When the pattern is divided between the continuous unit exposure regions, the dimensional correction is performed by appropriately extending the pattern from one or both of the divided pattern portions toward the opposing pattern. In the case where the pattern is superimposed between the continuous unit exposure areas, the dimensional correction is performed from the superposed one or both of the pattern portions. It is desirable that the pattern be retracted in the opposite direction.

Further, in the electron beam exposure method according to the present invention, when the pattern is divided between the continuous unit exposure areas, the dimensional correction is performed on both of the divided parts. It is also preferable that an auxiliary pattern is generated between pattern portions. If the resizing according to the present invention is described in more detail, FIG.
As shown in the figure, when the wiring pattern 20 in the longitudinal direction is formed across the three consecutively arranged sub-deflection regions S, S ', and S ", As a result, each sub-deflection region S is separated, and thus the pattern 20 is also separated.

In this state, in the electron beam exposure method according to the present invention, the difference (ΔX, ΔX) between the position correction amounts formed between the adjacent sub-deflection regions S in step (12).
Y). In this specific example, it is obtained by calculating only the deviation ΔX in the X-axis direction. Thereafter, the process proceeds to step (13), and the pattern size on the boundary of each sub-deflection area S is resized by the difference ΔX between the position correction amounts.

More specifically, as shown in FIG. 4B, the difference between the position correction amounts is positive, and two adjacent sub-deflection areas S and S ′ are located between the sub-deflection areas that are away from each other. In this case, the resizing of the pattern 20 extending between the sub-deflection areas S and S 'is executed. That is, in the drawing, a black-painted portion 30 is a portion where resizing has been performed. That is, in this specific example, the dimensional correction and resizing are performed by extending in the connection direction by the difference ΔX in the correction amount of the connection portion, thereby performing resizing.

As described above, such resizing may be such that the pattern 20 is extended from one sub-deflection area S to the other sub-deflection area S ', or both sub-deflection areas S, S , The patterns 20 may be respectively extended toward the opposing sub-deflection regions S and joined at the intermediate points. In this specific example, resizing is not performed in a direction other than the connection direction of the pattern (Y direction in this example).

FIG. 4C shows that the difference between the position correction amounts is negative, and the two adjacent sub-deflection areas S and S ′ are located between the sub-deflection areas in the direction of overlapping with each other. An example in which resizing of the pattern 20 extending between the sub deflection areas S and S ′ is performed is shown. In such a specific example, the pattern is resized in a direction to shrink the black painted portion 40.

The above processing is performed for all the sub-deflection areas, and in step (14), the conversion of the pattern data into the electron beam exposure format is completed. Due to the resizing, it is possible to suppress the overlapping and separation of unnecessary patterns due to the MIX & MATCH correction. Referring to FIG. 5 described above, when the electron beam exposure method according to the present invention is employed, the thinning and fattening of the line at the connection portion after the correction are prevented as compared with the graph of the conventional example. I understand.

Next, another specific example of the electron beam exposure method according to the present invention will be described below. That is, in the previous embodiment, the pattern 20 extending over the sub-deflection areas S and S ′ is used.
Although the resizing is used for the dimension correction in this example, in this specific example, an auxiliary pattern prepared separately is automatically generated.

That is, the pattern 20 extending between the sub-deflection areas S and S 'is separated between the sub-deflection areas S and S' in the direction in which the difference .DELTA.X of the position correction amount is separated in the X direction. Therefore, in this specific example, when data is converted into an electron beam exposure format, the position of such a sub-deflection area is searched, and a rectangular pattern for compensating the position is automatically generated between distant patterns. It is to be.

Specifically, the black-painted portion 3 in FIG.
The same effect can be expected by automatically generating a pattern of 0 as an auxiliary pattern. In this specific example, although the number of shots at the time of electron beam exposure increases as compared with the resize method of the previous specific example, the shot division does not change as compared with the conventional method. According to another aspect of the present invention, there is provided a method of manufacturing a semiconductor device by using an optical exposure system and an electron beam exposure system in combination, wherein an underlying pattern generated by lens distortion in the optical exposure system is provided. When correcting the positional deviation of the electron beam exposure system at the time of data conversion, there is a pattern formed over the individual unit exposure areas continuous with each other in the electron beam exposure system. In addition, the difference value of each position correction amount of the individual unit exposure region that is continuous with each other is calculated, and according to the difference value,
There is a recording medium in which a program for causing a computer to perform an operation of correcting a dimension of a pattern formed over the continuous individual unit exposure areas is provided.

[0044]

As described above, the technical configuration as described above is adopted, so that the drawbacks of the prior art can be improved, and no improvement or addition in terms of equipment can be made. Thus, an electron beam exposure method capable of performing an efficient electron beam exposure process with little fluctuation can be obtained.

[Brief description of the drawings]

FIG. 1 is a flowchart illustrating an operation procedure of a specific example of an electron beam exposure method according to the present invention.

FIG. 2 is a diagram illustrating an example of a position shift state of a sub deflection area S;

FIG. 3 is a plan view showing an example of a pattern used in a specific example of the electron beam exposure method according to the present invention.

FIG. 4 is a diagram illustrating an operation example of position correction in a specific example of the electron beam exposure method according to the present invention.

FIG. 5 is a graph illustrating a difference in effect between the conventional electron beam exposure method and the electron beam exposure method according to the present invention.

FIG. 6 is a flowchart illustrating an operation procedure of a specific example of a conventional electron beam exposure method.

FIG. 7 is a diagram showing an example of a pattern for measuring lens distortion used in a conventional electron beam exposure method.

FIG. 8 is a diagram showing a measurement result of a lens distortion in a conventional electron beam exposure method.

FIG. 9 is a diagram in which only a rotation component is extracted from the measurement data of FIG. 8;

FIG. 10 is a diagram showing displacement data after removing a primary component from lens distortion measurement data in a conventional electron beam exposure method.

FIG. 11 is a diagram showing a data shift and a resist pattern shape at a connection portion in a conventional electron beam exposure method.

[Explanation of symbols]

 DESCRIPTION OF SYMBOLS 1 ... Sample 2 ... Test pattern 3 ... Ideal grating 4 ... Lens distortion data 5 ... Rotation component 6 ... Position shift data by lens distortion 15 ... Sub-deflection boundary part 20 ... Pattern 30 ... Extended resize part, auxiliary pattern 40 ... Reduction resize part

Claims (9)

[Claims] In a method of manufacturing a semiconductor device by using an optical exposure system and an electron beam exposure system together, a displacement of a base pattern caused by a lens distortion in the optical exposure system is determined by using the electronic exposure system. When correcting at the time of data conversion in the line exposure system, when there is a pattern formed over individual unit exposure regions that are continuous with each other in the electron beam exposure system, and the continuous pattern is In the case where a connection failure has occurred due to the lens distortion, the pattern is resized based on displacement information between the unit exposure areas during data conversion in the electron beam exposure system. Electron beam exposure method. 2. The electron beam exposure method according to claim 1, wherein the connection failure of the pattern is separation or superposition of the pattern. 3. A method of manufacturing a semiconductor device by using an optical exposure system and an electron beam exposure system in combination, wherein the position shift of a base pattern caused by lens distortion in the optical exposure system is determined by using the electronic exposure system. When correcting at the time of data conversion in the line exposure system, if there is a pattern formed over the individual unit exposure regions that are continuous with each other in the electron beam exposure system, Calculating a difference value between the respective position correction amounts of the unit exposure regions, and performing dimensional correction of a pattern formed across the continuous individual unit exposure regions according to the difference value. Electron beam exposure method. 4. The apparatus according to claim 3, wherein each position correction amount in each unit exposure area is calculated by a predetermined method and stored in a predetermined storage unit. The electron beam exposure method according to the above. 5. The dimensional correction of the pattern is characterized in that the pattern is a pattern that straddles the unit exposure areas that are continuous with each other and that is resized to be divided or overlapped with each other. The electron beam exposure method according to claim 3 or 4, wherein 6. When the pattern is divided between the continuous unit exposure areas, the dimensional correction is performed by shifting the pattern from one or both of the divided pattern portions toward an opposing pattern. The electron beam exposure method according to any one of claims 3 to 5, wherein the length is appropriately extended. 7. When the pattern is superimposed between the continuous unit exposure regions, the dimensional correction is performed from one or both of the superimposed pattern portions in a direction opposite to the opposing pattern. 6. The electron beam exposure method according to claim 3, wherein the pattern is moved backward. 8. When the pattern is divided between the continuous unit exposure areas, the dimensional correction is to generate an auxiliary pattern between the two divided pattern parts. The electron beam exposure method according to any one of claims 3 to 5, wherein: 9. A method of manufacturing a semiconductor device using an optical exposure system and an electron beam exposure system in combination, wherein the displacement of a base pattern caused by a lens distortion in the optical exposure system is controlled by the electronic exposure system. When correcting at the time of data conversion in the line exposure system, if there is a pattern formed over the individual unit exposure regions that are continuous with each other in the electron beam exposure system, Calculates the difference value of the position correction amount of each of the unit exposure areas of the above, and performs, according to the difference value, the operation of performing dimensional correction of the pattern formed across the continuous individual unit exposure areas. A recording medium with a built-in program for executing it.