JP2005115110A - Method and program for correcting figure of photomask, recording medium having the program recorded thereon, apparatus and method for forming photomask pattern, photomask, and method for manufacturing semiconductor integrated circuit device - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To correct the figure of a photomask on the basis of the reference data relating to the deviation of a pattern dimension and to decrease fluctuation in the pattern dimension which much influences device characteristics. <P>SOLUTION: When the deviation of a pattern dimension is determined by subtracting the pattern dimension of a designed pattern in a processed layer of a semiconductor wafer from the pattern dimension of a real transfer pattern of the semiconductor wafer transferred by using a photomask, the relationship between the deviation of the pattern dimension and the pattern spacing between the designed pattern and adjacent patterns in either side of the designed pattern in the width direction is measured so as to obtain reference data. Whether the deviation of the pattern dimension is equal to or higher than the upper limit F<SB>H</SB>or equal to or lower than the lower limit F<SB>L</SB>is judged by using the reference data. In a region where the pattern dimension is judged as out of the above limit, the figure data of the photomask are corrected. A photomask pattern is formed on the basis of the corrected figure data, and the obtained photomask is used to form a pattern in the layer to be processed of a semiconductor integrated circuit device. <P>COPYRIGHT: (C)2005,JPO&NCIPI

Description

  The present invention relates to a photomask used in a lithography process in the manufacture of a semiconductor integrated circuit device (LSI), a photomask graphic correction method, a photomask graphic correction program, a recording medium recording the program, a photomask pattern forming apparatus, and a photomask pattern. The present invention relates to a forming method and a manufacturing method of a semiconductor integrated circuit device.

  In general, when manufacturing an LSI, a resist is applied on a semiconductor integrated circuit substrate (semiconductor wafer), a resist pattern that is a transfer image of a photomask is formed by a lithography process, and an unnecessary portion of a processed layer of a semiconductor wafer is formed by an etching process. The process of removing is performed. In each of these processes, it is desirable that the shapes of the mask pattern, the resist pattern, and the etching pattern of the layer to be processed are exactly the same except when the shape is intentionally changed. However, in practice, shape changes due to transfer characteristics in each process occur, and in particular, in the formation of a fine pattern that is lower than the exposure wavelength used in the lithography process, unfavorable shape changes occur (that is, the workpiece is processed). The layer pattern is different from the original design position. (For example, see Patent Document 1).

  One method for reducing the shape change is rule-based PPC (Process Proximity Correction). This method is a method of correcting a mask pattern by predicting the shape of another similar pattern by actually measuring several tens to several hundreds of reference patterns. Specifically, when the difference between the edge position of the actual pattern of the semiconductor wafer transferred using a photomask and the design edge position of the design pattern is the deviation amount of the edge position, both sides of the design pattern and its width direction Using the reference data obtained by actual measurement of the relationship between the pattern interval between adjacent patterns and the deviation amount of the edge position, the deviation amount of the edge position of the pattern is determined for the figure of the photomask to be corrected. The photomask figure is corrected for the area determined to be necessary.

FIG. 8 is a diagram for explaining a conventional photomask graphic correction method. In FIG. 8, the horizontal axis indicates the pattern interval (μm), the vertical axis indicates the deviation amount (nm) of the edge position, the solid line 111 indicates the deviation amount (nm) of the edge position before PPC processing, and the broken line 112 indicates The deviation amount (nm) of the edge position after PPC processing is shown. The PPC process is executed as follows. As indicated by a solid line 111 in FIG. 8, an area where the deviation amount of the edge position is determined to be equal to or greater than the upper limit (+5 nm) (area where the pattern interval is d ₃₁ or less) is indicated by a broken line 112 in FIG. 8. In addition, the edge position of the photomask pattern is changed so as to reduce the deviation amount of the edge position by a predetermined value (the minimum drawing dimension (minimum correction unit) when forming the photomask pattern, for example, 10 nm). In addition, the edge position of the photomask pattern is set so that the deviation amount of the edge position is increased by a predetermined value for an area where the deviation amount of the edge position is determined to be lower than the lower limit (−5 nm) (not shown). change.

JP-A-8-32450 (paragraphs 0002 to 0004)

  However, in the conventional photomask graphic correction method described above, the determination as to whether or not to correct the photomask graphic is made based on the deviation amount of the edge position, so that the characteristics of the semiconductor integrated circuit device are greatly affected. The pattern size (pattern width) of the photomask that gives the above was not corrected to an optimum value.

  Further, as shown by the broken line 112 in FIG. 8, even if the deviation amount of the edge position is corrected within the range from −5 nm to +5 nm on one side of the pattern, the deviation amount of the pattern dimension is represented by the broken line 122 in FIG. As shown in the figure, the range is from −10 nm to +10 nm, the pattern dimension variation is a maximum of 20 nm, and is a value larger than the minimum correction unit (for example, 10 nm) when forming a photomask pattern. As described above, the photomask graphic correction method based on the deviation amount of the edge position is not an appropriate correction method from the viewpoint of the accuracy of the pattern size that greatly affects the performance of the semiconductor integrated circuit device.

  Accordingly, the present invention has been made to solve the above-described problems of the prior art, and its purpose is to reduce variations in the pattern size of the transfer pattern of the semiconductor wafer, which has a great influence on the characteristics of the semiconductor integrated circuit device. Provided are a photomask graphic correction method, a photomask graphic correction program, a recording medium recording the program, a photomask pattern forming apparatus, a photomask pattern forming method, a photomask, and a method for manufacturing a semiconductor integrated circuit device. There is.

In one aspect of the photomask graphic correction method according to the present invention, a value obtained by subtracting the design pattern dimension of the semiconductor wafer from the pattern dimension of the actual transfer pattern of the semiconductor wafer transferred using the photomask is defined as a pattern dimension deviation amount. When correcting the figure of the photomask using reference data obtained by actual measurement of the relationship between the pattern interval between the design pattern and adjacent patterns on both sides in the width direction and the deviation amount of the pattern dimension There,
[1A] A step of determining whether or not a deviation amount of a pattern dimension is not less than a predetermined upper limit value or not more than a predetermined lower limit value with respect to a photomask figure to be corrected using the reference data;
[1B] The step of correcting the figure of the photomask for the area determined in step [1A] where the deviation amount of the pattern dimension is greater than or equal to the upper limit value or less than the lower limit value.

According to another aspect of the photomask graphic correction method of the present invention, a value obtained by subtracting the design pattern dimension of the semiconductor wafer from the pattern dimension of the actual transfer pattern of the semiconductor wafer transferred using the photomask is used as the pattern dimension. When taking the deviation amount, the photo mask figure is corrected using reference data obtained by actual measurement of the relationship between the pattern interval between the design pattern and adjacent patterns on both sides in the width direction and the deviation amount of the pattern dimension. A way to
[5A] The deviation of the pattern dimension of the asymmetric pattern in which the pattern interval between the design pattern and the adjacent patterns on both sides thereof is d ₁ and d ₂ is F (d ₀ ), and the design pattern is adjacent to both sides of the design pattern. F (d ₁ ) is the deviation amount of the pattern dimension of the symmetrical pattern where the pattern spacing between the patterns is d _1, and the pattern spacing between the design pattern and the adjacent patterns on both sides is d _2. When the deviation amount of the pattern dimension of the symmetric pattern is F (d ₂ ), the deviation amounts F (d ₁ ) and F (d ₂ ) of the pattern dimension are called from the reference data, and the pattern dimensions of the asymmetric pattern the deviation is approximated by _{F (d 0) = F (} d 1) / 2 + F (d 2) / 2, the shapes of the corrected photomask, the deviation amount of the pattern size F _(d 0) Determining whether equal to or less than a predetermined upper limit value or more or a predetermined lower limit value,
[5B] The method includes a step of correcting the figure of the photomask for an area determined in step [5A] where the pattern dimension deviation amount F (d ₀ ) is greater than or equal to the upper limit value or less than the lower limit value. is there.

  According to the present invention, it is possible to reduce variations in the pattern size of the transfer pattern of a semiconductor wafer that greatly affects the characteristics of the semiconductor integrated circuit device, and to provide the semiconductor integrated circuit device with desired characteristics. Is obtained.

  FIG. 1 shows a photomask graphic correction method according to the present invention (a graphic correction program for causing a computer to execute this method, and a computer-readable recording medium on which this program is recorded), a photomask pattern forming method (this method). 1 is a flowchart schematically showing a pattern forming apparatus to be implemented, a photomask on which a pattern is formed, and a method for manufacturing a semiconductor integrated circuit device.

  The photomask graphic correction method according to the present invention belongs to rule-based PPC (process proximity effect correction). In the photomask graphic correction method according to the present invention, reference data to be used in carrying out the photomask graphic correction method is acquired in advance (step ST11 in FIG. 1). The reference data is acquired by actually measuring the relationship between the pattern interval and the pattern dimension deviation amount for the reference pattern. Here, the pattern interval is a distance between a design pattern (line pattern) and adjacent patterns on both sides in the width direction. The pattern dimension deviation amount is a deviation amount caused by an optical proximity effect, etc., and the design pattern dimension of the semiconductor wafer is subtracted from the pattern dimension of the actual transfer pattern of the semiconductor wafer transferred using the photomask. Value.

  When carrying out the photomask graphic correction method according to the present invention, the photomask figure to be corrected (that is, graphic data) using the reference data acquired in advance (step ST21 in FIG. 1), pattern dimensions. The amount of deviation is determined, and the photomask figure is corrected for the area determined to be corrected (step ST22 in FIG. 1). These processes are executed by a computer (computer) constituting a part of the photomask pattern forming apparatus according to the present invention. Further, a program for causing a computer to execute these processes and a computer-readable recording medium (for example, a CD-ROM) on which the program is recorded are also included in the present invention.

  Further, when carrying out the photomask pattern forming method according to the present invention, a substrate provided with a mask forming layer (for example, a chromium film) is prepared (step ST31 in FIG. 1), and the corrected photomask figure is prepared. Based on the data, the photomask is processed by a photomask pattern forming device (pattern drawing device) to form a mask pattern (step ST32 in FIG. 1).

  Furthermore, when carrying out the method for manufacturing a semiconductor integrated circuit device according to the present invention, a resist is applied on a layer to be processed of a semiconductor wafer (step ST41 in FIG. 1), and a photomask having a pattern formed by the method described above. A resist pattern is formed by a lithography process using (step ST42 in FIG. 1), and an unnecessary portion of the processed layer is removed by an etching process using the resist pattern as an etching blocking portion to form a pattern of the processed layer (FIG. 1). 1 step ST43).

<First Embodiment>
FIG. 2 is a diagram for explaining a photomask graphic correction method according to the first embodiment of the present invention. In FIG. 2, the horizontal axis indicates the pattern interval (μm) and indicates the range from 0 μm to 40 μm. The vertical axis represents the pattern dimension deviation amount (nm), indicating the range from -15 nm to +20 nm. In FIG. 2, the deviation amount of the pattern dimension is a value obtained by subtracting the design pattern dimension of the semiconductor wafer from the pattern dimension of the actual transfer pattern of the layer to be processed of the semiconductor wafer transferred using the photomask. In FIG. 2, a solid line 101 indicates a change in the deviation amount of the pattern dimension before the photomask graphic correction process is executed, and a broken line 102 indicates after the photomask graphic correction process according to the first embodiment is executed. The change of the deviation amount of the pattern dimension is shown. However, the solid line 101 and the broken line 102 show an example of the pattern dimension deviation amount and the figure correction content in order to explain the photomask figure correction method according to the first embodiment. And the correction contents of the figure are not limited to those shown in the figure.

As can be seen from the solid line 101 in FIG. 2, the deviation amount of the pattern size when the pattern spacing d ₁₃ becomes 0, the deviation amount of the pattern dimension when the pattern spacing is narrower than the d ₁₃ is a positive value, the pattern interval There deviation of the pattern dimensions is a negative value when wider than d _13. This indicates that when the pattern interval is narrow, the width of the transfer pattern of the processed layer is widened, and when the pattern interval is wide, the width of the transfer pattern of the processed layer is narrowed.

Or In a first embodiment, by using the reference data which has been previously measured, the graphic to be corrected photomask, the deviation amount of the pattern size equal to or less than a predetermined upper limit value F _H above or a predetermined lower limit value F _L Determine whether or not. Then, the region where the deviation amount of the pattern size is determined to be less than the upper limit value F _H above or the lower limit value F _L, to correct the shape of the photomask.

  Next, the photomask figure correction method according to the first embodiment will be described in detail. The photomask graphic correction method according to the first embodiment is executed by, for example, a photomask pattern forming apparatus. FIG. 3 is a block diagram schematically showing the configuration of a photomask pattern forming apparatus. As shown in FIG. 3, a photomask pattern forming apparatus 200 forms (draws) a central processing unit 201, a storage unit 202, a medium reading unit 203 that reads information on a recording medium 204, and a mask pattern using an electron beam or the like. A drawing unit 205 and an operation unit 206. The photomask pattern forming apparatus 200 is installed with a program (for example, a program recorded on the recording medium 204) for executing the photomask graphic correction method according to the first embodiment. In addition, the storage unit 202 of the photomask pattern forming apparatus 200 stores in advance reference data used when performing the photomask graphic correction method according to the first embodiment. The central processing unit 201 performs processing for determining whether or not correction is required for the graphic data of the photomask line pattern and correction processing for the photomask graphic data. The drawing unit 205 forms a mask pattern based on the graphic data corrected by the central processing unit 201.

FIG. 4 is a diagram for explaining a photomask graphic correction method according to the first embodiment. In the first embodiment, the reference data is acquired as follows. As shown in FIG. 4, the pattern interval between the design pattern 301 and adjacent patterns 302 and 303 on both sides in the width direction is d ₁ and d _2, and the width of the design pattern 301 is W _m , corresponding to the design pattern 301. the actual pattern width of the transfer pattern 401 of the work layer and _{d 1} and _{d 2} of the function _{W e (d 1, d 2} ), the function F _(d 1 of the deviation of the pattern dimensions _{d 1} and _{d 2,} When d ₂ ) = W _e (d ₁ , d ₂ ) −W _m , the relationship between the pattern spacings d ₁ and d ₂ and the pattern dimension deviation amount F (d ₁ , d ₂ ) is measured and used as a reference. Get the data. Here, when a design pattern is closely match the pattern of the photomask, the width W _m of the design pattern 301 corresponds to the width of the corresponding line patterns of the photomask.

Next, using the reference data, the pattern dimension deviation amount F (d ₁ , d ₂ ) is greater than or equal to a predetermined upper limit (F _H = + 5 nm in FIG. 2) or predetermined It is determined whether or not the lower limit value (F _L = −5 nm in FIG. 2) is reached.

Next, the region where the deviation amount of the pattern size F _{(d 1,} d ₂₎ is determined to be less than the upper limit value F _H above or the lower limit value F _L, corrects the graphic data photomask. Correction of the graphic data, such that the pattern spacing of Figure 2 is indicated by dashed lines 102 to d ₁₂ below a region, the region where the deviation amount of the pattern size is determined to be less than the upper limit value F _H, the deviation amount of the pattern size Is reduced by a predetermined value F _A. Further, as the corrected again corrected shapes of photomask, a determination again, so that the pattern spacing of Figure 1 is shown by a broken line 102 in d ₁₁ less area, the deviation amount of the pattern size or larger than the upper limit value F _H For the area determined to be, the figure correction is repeated so that the deviation amount of the pattern dimension is decreased again by a predetermined value F _A. Furthermore, the correction of the graphic data, such that the pattern spacing of Figure 2 is indicated by dashed lines 102 to d ₁₄ or d ₁₅ less region, the region where the deviation amount of the pattern size is determined to be less than the lower limit value F _L, This is executed so that the deviation amount of the pattern dimension is increased by a predetermined value F _A.

In FIG. 2, the predetermined value F _A is set to 10 nm which is equal to the difference between the upper limit value F _H and the lower limit value F _L. This value F _A is set to correctable minimum drawing size when forming a pattern on a photomask (minimum correction units). It is not always necessary to equalize the difference between the upper limit value F _H and the lower limit F _L and minimum drawing dimensions. For example, the upper limit value F _H and the lower limit value F _L are considered in consideration of the high sensitivity of the mask dimension in the fine pattern (for example, the pattern variation on the wafer more than the dimension change value of the mask size occurs). In many cases, each absolute value is set to be 1/2 or more of the minimum drawing size. Further, the edge for correcting the photomask figure may be either edge on either side of the photomask line pattern. As shown in d ₁₁ and d ₁₂ of FIG. 2, when performing multi-stage correction, the first stage changes the right edge (to widen or narrow the pattern interval), and the second stage The left edge may be changed (to make the pattern interval wider or narrower), and the correction (change of the edge position of the line pattern) may be distributed alternately on both edges.

  As shown in FIG. 8, in the conventional photomask graphic correction method, the determination of whether or not to correct the photomask pattern graphic is outside the range where the edge position deviation amount is from -5 nm to +5 nm. Was going depending on whether or not. For this reason, as shown in FIG. 9, there is an error within the range from −10 nm to +10 nm (maximum error of 20 nm) in the deviation amount of the pattern dimension after correction. On the other hand, in the photomask graphic correction method according to the first embodiment, whether or not to perform photomask graphic correction is determined based on the pattern dimension deviation amount. As shown in FIG. 2, the deviation amount can be within a range of 10 nm from −5 nm to +5 nm. As described above, by using the photomask graphic correction method according to the first embodiment, the dimensional accuracy (difference from the design pattern dimension) of the transfer pattern of the layer to be processed of the semiconductor wafer is improved to twice that of the conventional method. Can be made.

<Second Embodiment>
As described in the first embodiment, the pattern spacing between the design pattern and adjacent patterns on both sides in the width direction is d ₁ and d ₂ , the width of the design pattern is W _m, and the semiconductor corresponding to the design pattern When the width of the transfer pattern of the processed layer of the wafer is W _e (d ₁ , d ₂ ), the pattern dimension deviation amount F (d ₁ , d ₂ ) is F (d ₁ , d ₂ ) = W _It can be calculated by _e (d ₁ , d ₂ ) −W _m . However, in order to perform this calculation, it is necessary to grasp the pattern dimension deviation amount F (d ₁ , d ₂ ) for every combination of the pattern intervals d ₁ and d ₂ . For this reason, it is necessary to acquire a very large amount of reference data. Therefore, in the second embodiment, by using an approximate expression as a calculation formula for the deviation amount of the pattern dimension, the reference data to be acquired is reduced (only the symmetric pattern of d ₁ = d ₂ needs to be grasped. ) Makes it possible.

In the second embodiment, acquisition of reference data is performed as follows. As shown in FIGS. 6A and 6B, the pattern interval between the design pattern 511 or 521 and the adjacent patterns 512, 513 or 522, 523 on both sides in the width direction is set to d (= d ₁ or d ₂ , respectively). ), The width of the design pattern 511 or 521 is W _m , the pattern width of the actual transfer pattern 611 or 621 of the work layer corresponding to the design pattern 511 or 521 is W _e (d), and the pattern dimensions , Where F (d) = W _e (d) −W _m is calculated as F (d), the relationship between the pattern interval d and the pattern dimension deviation amount F (d) is measured. Get reference data. The number of reference data may be determined based on various factors such as an error that is allowed as a deviation amount of a dimension pattern and a processing capability of an apparatus that performs figure correction of a photomask.

FIG. 5 is a diagram for explaining a photomask graphic correction method according to the second embodiment of the present invention. In FIG. 5, the pattern interval between the design pattern 501 and adjacent patterns 502 and 503 on both sides in the width direction is d ₁ and d ₂ , the pattern width of the design pattern 501 is W _m, and the workpiece corresponding to the design pattern 501 is processed. The width of the actual transfer pattern 601 of the layer was defined as W _e (d ₀ ), and the pattern dimension deviation amount was defined as F (d ₀ ). In the second embodiment, the deviation amount of the pattern dimension is calculated by F (d ₀ ) according to the following approximate expression.
F (d ₀ ) = F (d ₁ ) / 2 + F (d ₂ ) / 2
Here, F (d ₁ ) and F (d ₂ ) are deviation amounts of the pattern dimensions of the symmetrical pattern shown in FIGS. 6A and 6B, and are data stored as reference data. . Therefore, the deviation amount of the pattern dimension of the asymmetric pattern (d ₁ ≠ d ₂ ) as shown in FIG. 5 can be approximated by the above formula. Incidentally, the deviation D ₅₀₁ of the left edge of the asymmetric pattern 501 as shown in Figure 5, the deviation amount D ₅₁₁ of the left edge of the symmetrical pattern 511 shown in FIG. 6 (a), that is not strictly coincide Known for its process characteristics. However, when the magnitude of the error amount is sufficiently smaller than the accuracy assumed in the PPC process, it is considered that they are equal and there is no practical problem even if the above approximate expression is used. It is done.

For graphic photomask determines whether the deviation amount of the pattern size F (d ₀₎ is equal to or less than the upper limit value F _H above or the lower limit value F _L. Then, the region where the deviation amount of the pattern size F (d ₀₎ is determined to be less than the upper limit value F _H above or the lower limit value F _L, to correct the shape of the photomask.

In order to achieve the first embodiment, it is necessary to obtain in advance, as reference data, the deviation amount of the pattern dimension with respect to all combinations (asymmetric patterns) of the pattern intervals d ₁ and d ₂ . However, in order to do so, it is necessary to measure a very large number of pattern types and analyze and grasp their characteristics, which requires a great deal of time and labor. Was a barrier. In contrast, according to the mask pattern graphic correction method according to the second embodiment, the dimensional deviation amount of the asymmetric pattern can be approximated using the reference data of the dimensional deviation amount of the symmetric pattern. For this reason, according to the figure correction method of the mask pattern according to the second embodiment, it becomes easy to grasp the dimensional deviation amount of the asymmetric pattern, which is a barrier for achieving the first embodiment. In the second embodiment, points other than those described above are the same as those in the first embodiment.

<Third Embodiment>
FIG. 7 is a diagram for explaining a photomask graphic correction method according to the third embodiment of the present invention. In the orthogonal coordinate system shown in FIG. 7, the horizontal axis and the vertical axis respectively indicate pattern intervals d ₁ and d ₂ on both sides in the width direction of the design pattern (line pattern). In FIG. 7, the pattern interval in the range from 0 μm to 40 μm is shown in the horizontal axis direction, and the pattern interval in the range from 0 μm to 40 μm is shown in the vertical axis direction. In FIG. 7, a plurality of contour lines (arc-shaped curves) having the same pattern dimension deviation amount F (d ₁ , d ₂ ) are plotted. Numerical values −9 nm to +12 nm attached to the contour lines indicate pattern dimension deviation amounts F (d ₁ , d ₂ ).

In the case of the first embodiment, a region 701 in which the pattern dimension deviation amount F (d ₁ , d ₂ ) is equal to or higher than the upper limit (+5 nm) with one of the contour lines as a boundary (lower left hatching in FIG. 7). If it is in a region 702 (hatched region in the upper right in FIG. 7) that is equal to or lower than the lower limit (−5 nm) with another one of the contour lines as a boundary, the photomask figure is corrected. For this reason, it is necessary to clearly indicate the range of the areas 701 and 702 to the computer. However, it is very difficult to accurately indicate such a curved area in the computer, and even if it can be achieved, the subsequent calculation becomes very complicated and the processing time is greatly increased. Was not practical.

Therefore, in the third embodiment, the ranges of the regions 701 and 702 are cut out by a plurality of squares having various sizes. Here, the quadrangle has a side parallel to the horizontal axis and a side parallel to the vertical axis in FIG. As described above, since the ranges of the regions 701 and 702 having curved boundaries are approximated by a combination of a plurality of quadrangles, the conditions given to the computer are only the upper and lower limits of the pattern intervals d ₁ and d _2. It is only necessary to prepare an upper limit value and a lower limit value for the number of squares.

  FIG. 7 shows an example in which the cut-out error amount is ± 1 nm. That is, a square boundary is located between the time of error +1 nm (ie, +6 nm contour) of the boundary line (ie, +5 nm contour) of the region 701 in FIG. 7 and the time of error −1 nm (ie, +4 nm contour). An example is shown. In this case, the region 701 can be approximated by five squares (shown by dotted lines). Similarly, a square boundary between the error +1 nm (ie, −4 nm contour) of the boundary line (ie, −5 nm contour) of the region 702 in FIG. 7 and the error −1 nm (ie, −6 nm contour). An example where the line is located is shown. In this case, the region 702 can be approximated by three squares (shown by dotted lines).

  In the first embodiment, it is necessary to clearly indicate to the computer the area formed by the curve, which increases the processing time. However, by using the third embodiment, the load on the computer is reduced. Therefore, the processing time can be greatly reduced. Thereby, the achievement of the first embodiment is facilitated. In the third embodiment, points other than those described above are the same as those in the first embodiment.

3 is a flowchart for explaining a photomask pattern correcting method, a photomask pattern forming method, and a semiconductor integrated circuit device manufacturing method according to the present invention. It is a figure for demonstrating the figure correction method of the photomask which concerns on the 1st Embodiment of this invention. 1 is a block diagram schematically illustrating a configuration of a photomask pattern forming apparatus that performs a photomask graphic correction method according to a first embodiment; It is a figure for demonstrating the figure correction method of the photomask which concerns on 1st Embodiment. It is a figure for demonstrating the figure correction method of the photomask which concerns on the 2nd Embodiment of this invention. (A) And (b) is a figure for demonstrating the reference | standard data used when implementing the figure correction method of the photomask which concerns on 2nd Embodiment. It is a figure for demonstrating the figure correction method of the photomask which concerns on the 3rd Embodiment of this invention. It is a figure for demonstrating the figure correction method of the conventional photomask. It is a figure for demonstrating the subject of the figure correction method of the conventional photomask.

Explanation of symbols

301, 501, 511, 521 Design pattern, 401, 601, 611, 621 A transfer pattern of a layer to be processed of a semiconductor wafer, 302, 303, 502, 503, 512, 513, 522, 523 An adjacent pattern of the design pattern.

Claims (14)

When the value obtained by subtracting the design pattern dimension of the semiconductor wafer from the pattern dimension of the actual transfer pattern of the semiconductor wafer transferred using the photomask is used as the deviation amount of the pattern dimension, the design pattern and both sides in the width direction thereof A method of correcting a figure of a photomask using reference data obtained by actual measurement of a relationship between a pattern interval between adjacent patterns and a deviation amount of the pattern dimension,
[1A] A step of determining whether or not a deviation amount of a pattern dimension is not less than a predetermined upper limit value or not more than a predetermined lower limit value with respect to a photomask figure to be corrected using the reference data;
[1B] A step of correcting a figure of the photomask for an area where it is determined in step [1A] that the pattern dimension deviation amount is greater than or equal to the upper limit value or less than the lower limit value. Figure correction method.   2. The photomask graphic correction method according to claim 1, wherein the process of steps [1A] and [1B] is repeated with the photomask graphic corrected in step [1B] as a correction target again. For the region in which the pattern dimension deviation amount is determined to be greater than or equal to a predetermined upper limit value in the step [1A], the pattern dimension deviation amount is decreased by a predetermined first value in the step [1B]. Correct the photo mask shape,
For the region in which the pattern dimension deviation amount is determined to be less than or equal to the predetermined lower limit value in the step [1A], the pattern dimension deviation amount is increased by a predetermined second value in the step [1B]. The photomask graphic correction method according to claim 1, wherein the photomask graphic is corrected. Based on the reference data, when plotting a plurality of contour lines in which the deviation amount of the pattern dimension is the same value in the orthogonal coordinate system with the one pattern interval as the horizontal axis and the other pattern interval as the vertical axis, A region that is greater than or equal to the upper limit value bounded by one of the contour lines and a region that is less than or equal to the lower limit value bounded by another one of the contour lines has a side parallel to the horizontal axis and a side parallel to the vertical axis. Approximate a region that combines multiple rectangles,
In the step [1A], when the combination of the pattern intervals is included in the region of the combination of the plurality of squares, it is considered that the deviation amount of the pattern dimension is not less than a predetermined upper limit value or not more than a predetermined lower limit value. 4. The method of correcting a photomask figure according to claim 1, wherein When the value obtained by subtracting the design pattern dimension of the semiconductor wafer from the pattern dimension of the actual transfer pattern of the semiconductor wafer transferred using the photomask is used as the deviation amount of the pattern dimension, the design pattern and both sides in the width direction thereof A method of correcting a figure of a photomask using reference data obtained by actual measurement of a relationship between a pattern interval between adjacent patterns and a deviation amount of the pattern dimension,
[5A] The deviation of the pattern dimension of the asymmetric pattern in which the pattern interval between the design pattern and the adjacent patterns on both sides thereof is d ₁ and d ₂ is F (d ₀ ), and the design pattern is adjacent to both sides of the design pattern. F (d ₁ ) is the deviation amount of the pattern dimension of the symmetrical pattern where the pattern spacing between the patterns is d _1, and the pattern spacing between the design pattern and the adjacent patterns on both sides is d _2. When the deviation amount of the pattern dimension of the symmetric pattern is F (d ₂ ), the deviation amounts F (d ₁ ) and F (d ₂ ) of the pattern dimension are called from the reference data, and the pattern dimensions of the asymmetric pattern the deviation is approximated by _{F (d 0) = F (} d 1) / 2 + F (d 2) / 2, the shapes of the corrected photomask, the deviation amount of the pattern size F _(d 0) Determining whether equal to or less than a predetermined upper limit value or more or a predetermined lower limit value,
[5B] The step of correcting the figure of the photomask for the area determined in step [5A] where the pattern dimension deviation amount F (d ₀ ) is not less than the upper limit value or not more than the lower limit value. Characteristic photomask correction method.   6. The photomask graphic correction method according to claim 5, wherein the process of steps [5A] and [5B] is repeated with the photomask graphic corrected in step [5B] as a correction target again. The deviation of the step [5A] Oite pattern dimension F (d ₀₎ is determined to be above a predetermined upper limit value area, the deviation amount of the step [5B] Oite pattern dimension F a (d ₀₎ Correcting the photomask graphic to reduce it by a predetermined first value;
The deviation of the step [5A] Oite pattern dimension F (d ₀₎ is determined to be below a predetermined lower limit value area, the deviation amount of the step [5B] Oite pattern dimension F a (d ₀₎ The photomask graphic correction method according to claim 5, wherein the photomask graphic is corrected so as to be increased by a predetermined second value. The first value and the second value are equal to a minimum drawing dimension when forming a mask pattern of a photomask,
The absolute value of the said upper limit value and the said lower limit value is respectively 1/2 or more of the said 1st value and the said 2nd value. The photomask figure in any one of Claim 3 or 7 characterized by the above-mentioned. Correction method.   9. A photo mask graphic correction program for causing a computer to execute a photo mask graphic correction process using the graphic correction method according to claim 1.   9. A computer-readable recording medium on which a program for causing a computer to execute a photo mask graphic correction process using the graphic correction method according to claim 1 is recorded. A figure of a photomask is corrected using the figure correction method according to any one of claims 1 to 8,
A photomask pattern forming apparatus, wherein a mask pattern is formed based on the corrected photomask figure. Correcting the figure of the photomask using the figure correcting method according to any one of claims 1 to 8,
Forming a mask pattern based on the corrected photomask figure. A photomask pattern forming method comprising:   A photomask having a shape based on a figure corrected by using the photomask figure correcting method according to any one of claims 1 to 8. Correcting the figure of the photomask using the figure correcting method according to any one of claims 1 to 8,
Forming a mask pattern using the corrected figure;
Forming a resist pattern on a work layer of a semiconductor wafer by a lithography process using a photomask having the mask pattern;
Forming a pattern on a layer to be processed of the semiconductor wafer by an etching process using the resist pattern. A method for manufacturing a semiconductor integrated circuit device, comprising:

Images