JP2008242112A - Mask pattern evaluation device and manufacturing method of photomask - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a mask pattern evaluation device which can accurately extract a hot spot by reflecting the formation result of a mask pattern actually formed on a mask pattern, and the manufacturing method of the photomask. <P>SOLUTION: The mask pattern evaluation device includes: a data correcting means of correcting mask data on the basis of the measured value of the mask pattern of the photomask; a pattern generating means of performing lithography simulation on the basis of the mask data corrected by the data correcting means; a wafer pattern generated through the lithography simulation performed by the pattern generating means; and a pattern comparing means of comparing a design pattern generated on the basis of design data to extract the hot spot. <P>COPYRIGHT: (C)2009,JPO&INPIT

Description

  The present invention relates to a mask pattern evaluation apparatus and a photomask manufacturing method, and more particularly to a mask pattern evaluation apparatus and photomask manufacturing method for extracting hot spots.

  A highly accurate and fine device such as a large scale integrated circuit (LSI) is manufactured through an exposure process in which a mask pattern formed on a photomask is transferred onto a wafer. In this process, a portion where the “exposure margin” of the mask pattern is small is called a “hot spot” or the like. Here, the “exposure margin” means an allowable range with respect to variation in exposure amount and focus. That is, the “hot spot” means a portion where an abnormality of a pattern formed on the transferred wafer is likely to occur when the exposure condition is deviated.

It is desirable to reduce such hot spots as much as possible. In order to inspect a hot spot of a mask, a pattern having a large variation in line width with respect to a variation in exposure amount or focus is extracted by lithography simulation based on mask pattern data before conversion into mask drawing data. At this time, the mask pattern data used in the simulation is based on the assumption that the mask is manufactured with the designed line width. In addition, in such a mask pattern correction method, a technique is disclosed in which correction is performed in consideration of a change in an exposure margin in an exposure process on a wafer without significantly increasing the burden of optical proximity effect correction processing on the mask pattern. (Patent Document 1).
Japanese Patent Laid-Open No. 2001-13668

However, the line width of the mask pattern actually formed on the photomask is often different from the design value. Furthermore, the deviation of the line width of the mask pattern from the design value is not uniform within the mask and often has a distribution. As a result, it has often been difficult to accurately extract hot spots.
The present invention provides a mask pattern evaluation apparatus and a photomask manufacturing method capable of accurately extracting hot spots by reflecting the performance of a mask pattern actually formed on a photomask.

According to one aspect of the present invention, a data correction unit that corrects mask data based on a measurement value of a mask pattern of a photomask, and a lithography simulation that is performed based on the mask data corrected by the data correction unit. Pattern generation means, pattern comparison means for extracting hot spots by comparing the wafer pattern generated by the lithography simulation executed by the pattern generation means with the design pattern generated based on the design data, There is provided a mask pattern evaluation apparatus characterized by comprising the above.
According to another aspect of the present invention, a step of forming a mask pattern on a mask to form a photomask, a step of evaluating the formed photomask using the mask pattern evaluation apparatus, And a step of executing a process relating to the photomask based on the result of the evaluation.

  ADVANTAGE OF THE INVENTION According to this invention, the mask pattern evaluation apparatus and photomask manufacturing method which can extract a hot spot correctly by reflecting the performance of the mask pattern actually formed in a photomask are provided.

Hereinafter, embodiments of the present invention will be described with reference to the drawings.
FIG. 1 is a block diagram of a mask pattern evaluation apparatus according to an embodiment of the present invention.
FIG. 2 is a flowchart for explaining the operation of the mask pattern evaluation apparatus shown in FIG.
In this embodiment, when the mask pattern is inspected, as shown in steps S100 and S110 in FIG. 2, a mask is actually manufactured, and the deviation of the pattern line width from the design value is measured. Then, hot spots on the wafer are extracted by lithography simulation based on the mask pattern reflecting the measured deviation. By doing so, the hot spot can be extracted more accurately by reflecting the performance of the mask pattern actually manufactured.

Hereinafter, before describing the mask inspection apparatus of the present embodiment, a comparative example will be described. FIG. 3 is a flowchart illustrating a hot spot extraction method as a comparative example.
4 to 6 are conceptual diagrams illustrating design data, a mask pattern, and a pattern transferred onto the wafer.
As shown in FIG. 4, when manufacturing an LSI or the like, first, device design data 70 is created. The design data 70 includes information on the shape, size, and arrangement of various design patterns 75 formed on the manufactured device. Next, a mask pattern 76 formed on the mask 71 is formed based on the design data 70. In FIG. 4, the design pattern 75 and the mask pattern 76 are represented as the same shape, but in actuality, the mask pattern 76 is designed to correct an optical proximity effect (OPE) or the like. It may have a shape different from 75.

  A wafer pattern 77 is formed on the wafer 72 by an exposure process and an etching process using the mask 71 thus formed. The wafer pattern 77 is ideally the same as the design pattern 75 based on the design data. However, in practice, a hot spot 78 having a shape and size different from the design pattern 75 may be formed. In FIG. 4 (d), the hot spot 78 is exemplified by a shape in which the corners of the pattern are rounded or swollen, but this is merely for convenience of explanation, and the actual hot spot is Different.

  Such a hot spot causes a defect of the device and becomes one of factors that reduce the yield in the manufacturing process. Therefore, by examining the location where hot spots appear by inspecting the photomask, improve the photomask pattern, and manage the exposure conditions in the manufacturing process so that hot spots do not occur at those locations. Is needed.

  In contrast to this, in the mask inspection method of the comparative example shown in FIG. 3, first, mask data 90 is generated from the design data 70. The mask data 90 is data related to the mask pattern 76 shown in FIG. Then, lithography simulation is performed based on the mask data 90 (step S200). In this method, the actual transfer process of the wafer 72 to the resist is reproduced by simulation, and a resist wafer pattern 77 formed on the wafer by exposure and development using a photomask is obtained by simulation.

  When the wafer pattern 77 is obtained in this way, it is compared with the design pattern 75 (step S210). And a part where pattern shape, size, and arrangement exceed a predetermined range is extracted as a hot spot.

  As described above, according to this comparative example, a lithography simulation is performed based on the mask data 90, and a hot spot is extracted by comparing the wafer pattern 77 and the design pattern 75 obtained as a result.

However, the mask pattern of a photomask that is actually manufactured is not always as designed. That is, in the case of the comparative example shown in FIG. 3, the photomask 71 is created based on the mask data 90 (step S220), but the line width of the mask pattern 76 of the created photomask 71 is shifted from the design value. However, the information is not reflected in the lithography simulation in step S200.
FIG. 5 is a schematic view illustrating the case where the line width of the mask pattern of the photomask is different from the design value.
When the photomask 71 is manufactured as shown in FIG. 5B corresponding to the design data 70 as shown in FIG. 5A, the actual photomask is caused by a deviation in conditions in the formation process. The line width d2 of the mask pattern may be different from the design value (represented by a broken line). Thus, when the line width d2 of the mask pattern 76 deviates from the design value, the location and frequency at which hot spots appear also change.

  However, in the case of the comparative example shown in FIG. 3, since the performance of the mask pattern actually formed is not taken into consideration, the wafer pattern 77 obtained by lithography simulation may be different from the actual one. For example, even if the mask pattern 76 is designed as designed, it may become a hot spot because the line width of the mask pattern 76 deviates from the design value even if it is not a hot spot. Or vice versa. As a result, there may be a case where a location to be a hot spot cannot be extracted.

On the other hand, according to the present embodiment, it is possible to extract hot spots more accurately by performing lithography simulation by reflecting the actual photomask performance.
Hereinafter, the embodiment of the present invention will be described with reference to FIGS. 1 and 2 again.
The mask pattern evaluation apparatus 5 of this embodiment includes an optical unit 40, a dimension calculation unit 12, a dimension comparison unit 15, a dimension error calculation unit 17, a data correction unit 20, a pattern generation unit 25, and a pattern comparison unit. 30. Note that these elements do not need to be integrated. For example, the optical means 40 may be separate from other elements.

  In the present embodiment, as shown in FIG. 2, the photomask 71 is created based on the mask data 90 created based on the design data 70 (step S100). Then, the mask pattern 76 of the created photomask 71 is measured.

  The optical means 40 shown in FIG. 1 has a role of imaging the mask pattern 76 of the photomask 71 created based on the design data. The optical means 40 includes a stage 27 on which a photomask is placed, a light source 51, a condenser lens 52a, an objective lens 52b, and an image sensor 55. The stage 27 can be moved in a plane by the stage control means 13. As a result, the mask pattern can be imaged with high resolution over the entire photomask 71.

  A condenser lens 52a and a light source 51 are provided in this order substantially below the stage 27, and an objective lens 52b and an image sensor 55 are provided in this order substantially above the stage 27. These are disposed on the optical axis of the light emitted from the light source 51. In the optical means 40, the light emitted from the light source 51 is condensed by the condenser objective lens 52a and imaged on the photomask. The light transmitted through the photomask diverges and is collected by the objective lens 52b, and is imaged by an image sensor 55 such as a CCD.

  In addition, the mask pattern evaluation apparatus 5 of this embodiment does not necessarily include the optical means 40. In other words, any process may be used as long as the process described below is executed based on data captured by the optical means 40 provided as a separate body.

  The dimension calculation means 12 is connected to the optical means 40 and the stage control means 13. The mask pattern obtained by the optical means 40 is subjected to image processing, and then converted into coordinate data to calculate a size or size distribution (step S100). At this time, the position information of the mask pattern 76 formed on the photomask 71 can be measured by associating the position information of the photomask from the stage 27. Here, by measuring the dimensions over the entire photomask 71, a size distribution can also be obtained.

  The dimension comparison unit 15 is connected to the dimension calculation unit 12 and the database 10. The size and size distribution of the mask pattern obtained from the size calculation means 12 is compared with the size and size distribution on the mask data obtained from the database 10, and the mask data and the actually measured mask pattern do not match. Can be detected (step S120).

  The dimension error calculation means 17 is connected to the dimension comparison means 15 and calculates the “dimension error” of the mismatched portion between the mask data obtained from the dimension comparison means 15 and the actually measured mask pattern 76 (step S125). ). At this time, for example, after dividing the mask 71 into a plurality of regions, the dimensional error of the line width d2 of the mask pattern 76 can be averaged and calculated for each region. Thereby, the process for every area | region of the photomask 71 can be performed, and it can suppress that calculation data amount becomes excessive. Here, the dimension error means an amount of deviation from the dimension on the mask data.

  The data correcting unit 20 is connected to the dimensional error calculating unit 17 and corrects the mask data based on the dimensional error obtained from the dimensional error calculating unit 17 (step S130).

  The pattern generation unit 25 is connected to the data correction unit 20, performs lithography simulation using the corrected mask data obtained from the data correction unit 20, and creates a wafer pattern 77 to be transferred onto the wafer 72. (Step S140).

  The pattern comparison unit 30 is connected to the pattern generation unit 25 and the database 10. The wafer pattern 77 simulated by the pattern generating means 25 is compared with the design pattern 75 based on the design data (step S150). Then, a portion where the deviation of the wafer pattern 77 viewed from the design pattern 75 exceeds a predetermined range is extracted as a hot spot 78.

  As described above, according to the present embodiment, by creating the photomask 71 and measuring the line width of the mask pattern 76, the workmanship of the mask pattern 76 can be reflected in the lithography simulation. As a result, a hot spot corresponding to the actual photomask 71 can be accurately extracted. For example, as described above with reference to FIG. 5, even when the location where the hot spot appears changes because the line width of the mask pattern 76 deviates from the design value, this can be accurately extracted.

In addition, according to the present embodiment, even when there is a distribution of workmanship within the surface of the photomask 71, a lithography simulation reflecting this can be performed.
FIG. 6 is a schematic view illustrating the case where the line width of the mask pattern 76 has a distribution in the plane of the photomask 71.
That is, even if the line width d of the pattern is uniform in the design data 70, the line width d2 of the mask pattern 76 may have a distribution in the plane of the photomask 71 when the photomask 71 is created. In the specific example illustrated in FIG. 6, the line width d2 of the mask pattern 76 is large near the center of the photomask 71, and the line width d2 of the mask pattern 76 decreases near the end of the mask pattern 76 near the center.
When distribution occurs in the line width of the mask pattern 76 in this way, the location and frequency at which hot spots appear correspondingly change.
On the other hand, according to the present embodiment, the distribution of the line width of the mask pattern 76 can be reflected. That is, in the process described above with reference to FIG. 2, first, the line width of the mask pattern 76 is measured and calculated over the entire photomask 71 (step S110). Then, the measured dimensional distribution of the mask pattern is compared with the dimensional distribution on the mask data obtained from the database 10, and the mismatch between the mask data and the actually measured mask pattern is detected over the entire photomask 71. (Step S120).

  Next, the distribution of “dimensional error” of the mismatched portion between the mask data obtained from the dimension comparison means 15 and the actually measured mask pattern 76 is calculated (step S125). The data correction unit 20 corrects the distribution of the mask data based on the distribution of the dimensional error obtained from the dimensional error calculation unit 17 (step S130).

  The pattern generating means 25 performs lithography simulation using the corrected mask data distribution obtained from the data correcting means 20, and creates a distribution of the wafer pattern 77 transferred onto the wafer 72 (step S140).

The pattern comparison unit 30 compares the distribution of the wafer pattern 77 simulated by the pattern generation unit 25 with the design pattern 75 based on the design data (step S150). Then, a portion where the deviation of the wafer pattern 77 viewed from the design pattern 75 exceeds a predetermined range is extracted as a hot spot 78.
As described above, even when there is a distribution in the workmanship of the photomask 71, it is possible to accurately extract hot spots by performing lithography simulation reflecting this.

  After extracting the hot spot in this way, processing related to the photomask is executed according to the content of the hot spot. Specifically, for example, if the photomask is a non-defective product, it is used or shipped. Further, if the repair of the mask pattern 76 is possible, the repair can be executed, the mask pattern 76 can be improved, or the mask can be treated as a defective product when the frequency of hot spots is high. . Further, the photomask 71 may be formed again by improving the formation conditions of the photomask 71. Then, the non-defective photomask is attached to the exposure apparatus and the exposure process is executed.

  Alternatively, it is also effective to monitor the location where the hot spot is extracted by the mask pattern evaluation apparatus of the present embodiment as a process monitor in the mass production process and manage the exposure conditions so that the hot spot does not occur.

  The embodiments of the present invention have been described above with reference to specific examples. However, the present invention is not limited to these specific examples.

For example, in this embodiment, the dimension is measured over the entire surface of the mask pattern, but the present invention is not limited to this. In the mask pattern, the dimension of a predetermined portion may be measured and compared. In this way, it is possible to investigate the distribution of the photomask performance in a shorter time.
Moreover, in this embodiment, although the dimension was represented by the line width, it is not limited to this. Similar effects can be obtained by using rounded L-shaped corners, dot areas, and the like.

  Furthermore, in the mask pattern evaluation apparatus of the present invention, the elements constituting the optical means, dimension calculating means, dimension comparing means, dimension error calculating means, data correcting means, pattern generating means, pattern comparing means, etc. Even those appropriately modified by a trader are included in the scope of the present invention as long as they include the gist of the present invention.

It is a block diagram of the mask pattern evaluation apparatus which is embodiment of this invention. 3 is a flowchart for explaining the operation of the mask pattern evaluation apparatus shown in FIG. 1. It is a flowchart which illustrates the hot spot extraction method as a comparative example. It is a conceptual diagram which illustrates design data, a mask pattern, and a pattern transferred onto a wafer. It is a conceptual diagram which illustrates design data, a mask pattern, and a pattern transferred onto a wafer. It is a conceptual diagram which illustrates design data and a mask pattern.

Explanation of symbols

  DESCRIPTION OF SYMBOLS 5 Mask pattern evaluation apparatus, 10 Database, 12 Dimension calculation means, 13 Stage control means, 15 Dimension comparison means, 17 Dimension error calculation means, 20 Data correction means, 25 Pattern generation means, 27 Stage, 30 Pattern comparison means, 40 Optical Means, 51 light source, 52a condenser lens, 52b objective lens, 55 image sensor, 70 design data, 71 photomask, 72 wafer, 75 design pattern, 76 mask pattern, 77 wafer pattern, 78 hot spot, 90 mask data

Claims (4)

Data correction means for correcting the mask data based on the measurement value of the mask pattern of the photomask;
Pattern generation means for performing lithography simulation based on the mask data corrected by the data correction means;
Pattern comparing means for extracting a hot spot by comparing a wafer pattern generated by the lithography simulation executed by the pattern generating means with a design pattern generated based on design data;
A mask pattern evaluation apparatus comprising:   2. The mask pattern evaluation apparatus according to claim 1, wherein the data correction unit corrects the mask data based on measured values of the mask pattern at a plurality of portions of the photomask.   The mask pattern evaluation apparatus according to claim 1, wherein the measurement value of the mask pattern is a line width of the mask pattern. Forming a mask pattern on the mask to form a photomask; and
Using the mask pattern evaluation apparatus according to claim 1 to evaluate the formed photomask;
Executing a process related to a photomask based on the result of the evaluation;
A method for manufacturing a photomask, comprising:

Images