JP5422725B2 - Electron microscope system and resist pattern film thickness reduction evaluation method using the same - Google Patents

Description

  The present invention relates to a method for evaluating and managing a resist pattern formed on a wafer in a semiconductor manufacturing process, and in particular, by using an electron microscope image of the resist pattern, the amount of resist film reduction (reduction in resist pattern height). It is related with the technology which measures and evaluates.

  Conventionally, in a lithography process, a length measuring SEM (Scanning Electron Microscope), which is an electron microscope for the purpose of pattern dimension measurement, has been widely used as a process management tool. The length measurement SEM is capable of imaging at a high magnification of 100,000 to 300,000 times, and can measure the size of a fine pattern on the order of several tens of nanometers with an accuracy of 1 nanometer or less. Non-Patent Document 1 discloses a basic configuration of a length measurement SEM.

  The lithography process consists of transferring the circuit pattern onto the wafer by resist exposure / development and etching treatment according to the transferred resist pattern. The length measurement SEM is the transferred resist pattern or the pattern after etching. It is used for measuring dimensions. In particular, wiring patterns such as transistor gate wirings are closely related to the pattern width and device operating characteristics, and therefore strict dimension management is performed.

  FIG. 2A shows, in order from the left, the cross-sectional shape of the pattern before starting etching, during etching, and after etching. The length measurement SEM measures the resist pattern width W1 or the pattern width W2 after etching.

  In the conventional lithography process, if the resist pattern width W1 is within the standard, the relationship that the pattern width W2 after etching is also within the standard is established. Management was possible.

Tatsuhiko Higashiki, "Optical Lithography Technology II -Measurement and Control-", ED Research, pp. 31-41 (2003) Mikio Takagi, Yoshihisa Shimoda, "Image Analysis Handbook", The University of Tokyo Press, Function Part I, Chapter 2 (1991)

  In order to meet the needs for pattern miniaturization, in the exposure process, the introduction of high NA exposure technology for obtaining the resolution necessary for fine pattern formation is progressing. As a result, the margin of the exposure process is reduced, and slight variations in exposure parameters such as the dose amount during exposure and the focus position cause a reduction in the resist pattern film thickness, that is, a reduction in height relative to normal exposure. It was.

  As shown in FIG. 2B, when the pattern height is low (h1 <H1) even if the pattern width before the start of etching is equal (w1 = W1) compared to the case of FIG. The pattern width becomes smaller (w2 <W2). In the middle of the etching process, the resist film thickness gradually decreases. Therefore, if the original pattern height is low, the resist pattern almost disappears in the middle as shown in the center diagram of FIG. This is because, thereafter, the film to be processed itself is etched. As described above, the pattern width after etching is directly related to the operating characteristics of the device. The situation as shown in FIG. 2B, which causes a decrease in the pattern width after etching, should not be present.

  Under the introduction of high NA exposure technology, from the viewpoint of process management, the measurement at the resist pattern stage is not sufficient for the measurement of the pattern width (W1, w1), and the pattern height (H1, h1). In addition, it is desirable to measure together. The pattern height can be measured by dividing the wafer using an electron microscope for cross-sectional observation, or using an AFM (Atomic Force Microscope), but the former requires the wafer to be broken. Obviously, it cannot be applied to in-line daily process management, and the latter is not a suitable method for process monitoring that requires high throughput.

  An object of the present invention is to provide a system that can be applied as part of in-line process management and realizes high-throughput resist pattern film loss detection or film loss measurement.

  In order to achieve the above object, the present invention provides an electron beam image acquiring means for acquiring an image of a sample having a resist pattern formed on the surface using a scanning electron microscope, and processing the acquired image to obtain the resist pattern. Quantifying means for quantifying the variation characteristics of the brightness of the image in the desired region, and a reduction amount of the characteristics of the variation in the brightness of the image quantified by the quantification means from the resist pattern film thickness reference value An electron microscope system comprising: an index value calculating unit that calculates an index value to be associated with the display unit; and a display unit that displays information on the index value calculated by the index value calculating unit on a screen.

  According to the present invention, since it is possible to perform film thickness reduction detection or film thickness reduction measurement using an image used in the conventional line width measurement, it is possible to perform critical processing without reducing the throughput of conventional process management. It is possible to monitor the film loss where the process is defective. In addition, if the present invention is applied to the exposure process condition determination, that is, the dose amount and the focus position optimization, a more effective process window can be set as compared with the case of using only the conventional line width measurement result. Become.

It is a flowchart explaining the 1st Embodiment of this invention. It is explanatory drawing of resist film reduction. It is a figure explaining the calculation method of a film | membrane reduction index value. It is a figure explaining the situation in various film reduction states. It is a GUI screen for registering the relationship between the film reduction amount and the film reduction index value. It is an output screen of the estimation result of the film reduction amount. It is a flowchart explaining the 1st Embodiment of this invention. It is a flowchart explaining the 1st Embodiment of this invention. It is a GUI screen for registering the relationship between the etch bias and the film reduction index value. It is an output screen of the estimation result of etch bias.

  Hereinafter, a resist film reduction measuring method using an electron microscope according to the present invention will be described with reference to the drawings.

  Regarding the resist film reduction measurement method using the electron microscope according to the present invention, after describing the configuration and overall flow of the SEM system, each step will be described in detail, and further, the SEM system equipped with this method will be described.

[SEM system]
FIG. 1A shows a configuration of a length measurement SEM system having a film thickness reduction measuring function according to this embodiment. The length measuring SEM system according to the present embodiment includes an SEM main body 10, an image processing / overall control unit 109, an arithmetic processing unit 110, and a database unit 113, and is connected to a data server 116 via a network.

  The SEM body 10 includes an electron gun 101, an acceleration electrode 103 that accelerates the electron beam 102 emitted from the electron gun 101, a focusing lens 104, a deflection electrode 105 that deflects the trajectory of the electron beam 102, and a focal position at which the electron beam 102 converges. Is positioned on the surface of the sample 107 on which the pattern is formed, the objective lens 106 that controls the focal position of the electron beam 102, and some of the secondary electrons generated from the sample 107 irradiated with the electron beam 102 are detected by the detector 108. Is detected. The detection signal of the detector 108 is sent to the image processing / overall control unit 109 for processing, and an SEM image is obtained. Using this SEM image, processing is performed with reference to information stored in the database unit 113 by the arithmetic processing unit 110, and information related to film reduction is extracted. The result is sent to the data server 116 via the communication line and stored.

  The sample 107 is placed on a table (not shown), and the table (not shown) is controlled by the image processing / overall control unit 109 so that a desired region on the sample is positioned in the irradiation region of the electron beam 102.

  The arithmetic processing unit 110 includes a film reduction index value measurement unit 111, a database collation unit 112, a film reduction amount estimation unit 114, and an input / output unit 115 including a display screen.

[Overall flow]
FIG. 1B shows an overall flow of film reduction measurement performed by the arithmetic processing unit 110.
(S1): A resist pattern is imaged by the SEM 10, and a signal obtained by the imaging is processed by the image processing / overall control unit 109 to obtain an SEM image of the sample.
(S2): An index value for film reduction is calculated from the acquired image by the method described later.
(S3): Collation with the database 20 in which the correspondence between the film reduction amount of the actually formed pattern and the film reduction index value with respect to the resist pattern formed under normal exposure conditions stored in the database unit 113 is registered. To do. A method for creating the database 20 will be described later with reference to FIG.
(S4): Estimate the film reduction amount of the actually formed pattern with respect to the resist pattern formed under normal exposure conditions based on the collated result. The estimated result is displayed on the display screen of the input / output unit 115. The above is the overall flow of film reduction measurement. Details of each flow will be described below.

[Calculation of film loss index value]
The step S2 in FIG. 1B will be described in detail.
The film thickness of the actually formed pattern with respect to the resist pattern formed under normal exposure conditions is caused by excessively irradiating light on the resist pattern part that should not be irradiated with light, especially on the resist pattern. To do. Excessive light irradiation on the top of the resist pattern can also occur due to a focus position shift during exposure or an increase in dose.

  In the case of a resist material that is generally used, the surface roughness (surface roughness) increases as the amount of irradiation light increases. The film loss occurs when light is excessively applied to the upper part of the resist pattern, so that the roughness of the upper surface of the resist is inevitably accompanied.

  The present invention uses this mechanism to quantify the roughness change of the upper surface of the pattern accompanying the film reduction on the SEM image and use it as a film reduction index value.

  FIG. 3A shows an SEM image when no film reduction occurs, and FIG. 3B shows an SEM image when film reduction occurs. In the SEM image, the flat part is dark and the inclined part is detected brightly due to the inclination angle effect, and therefore, if there is roughness, the rough edge is detected brightly in a streak shape. The portion surrounded by the dotted line in FIG. 3A corresponds to the upper surface of the resist pattern, but FIG. 3B is a region with uneven brightness compared to FIG. 3A. Is due to the existence of roughness. Several examples are given as methods for quantifying the features on the image.

  (A): Projected waveform on the y-axis in the region corresponding to the upper part of the pattern of the SEM image (the region surrounded by the dotted lines in FIGS. 3A and 3B) (FIG. 3C): the vertical axis SE Intensity is The brightness of the SEM image is shown, the horizontal axis Y) is generated, and the brightness variation of the projection waveform expressed by the standard deviation σ is used as the roughness index value, that is, the resist film reduction index value. The greater the roughness, the more uneven the brightness, so the standard deviation σ takes a larger value. FIG. 4B shows y-axis projection waveforms at various film reduction amounts (from (i) to (v), the larger the film number, the larger the film reduction). As shown in the figure, this index value increases as the film thickness decreases. According to the present method, since the contribution of high-frequency noise peculiar to SEM can be reduced by averaging effect by taking a certain range of projections, it is possible to detect brightness unevenness due to surface roughness with high sensitivity. Further, by performing the averaging in the pattern line width direction, it is possible to reduce the influence of the brightness change due to the pattern shape change other than the film reduction. If the region in the pattern longitudinal direction for generating the projection waveform is made longer, the accuracy as a more accurate film reduction index value can be improved. In the graphs shown in i) to (v) of FIG. 4B, the description of the vertical axis and the horizontal axis is omitted, but these are the vertical axis and horizontal axis of the graph shown in FIG. Same as axis.

(B): A brightness histogram is created in the region corresponding to the upper part of the pattern of the SEM image (the region surrounded by the dotted lines in FIGS. 3A and 3B) (FIG. 3D). The normal distribution is applied, and the σ value at that time or the center of the normal distribution is used as the resist film decrease index value. Specifically, the appearance frequency is determined for each brightness on the vertical axis from the projection waveform on the y-axis obtained in (a). The idea is the same as (a) above, and quantification is based on the fact that the greater the roughness, the greater the uneven brightness on the average. FIG. 4C shows histograms for various film reduction amounts (from (i) to (v), the larger the number, the greater the film reduction). This index value also increases as the film thickness decreases. In the graphs shown in i) to (v) of FIG. 4C, the description of the vertical axis and the horizontal axis is omitted, but these are the vertical axis and horizontal axis of the graph shown in FIG. Same as axis.
(C): A projection waveform on the x-axis of the SEM image is generated, and the minimum value at the center (min in FIG. 3E) is used as the film reduction index value. As the roughness increases, rough edges are detected brightly in a streak pattern, so quantification is based on the fact that the larger the roughness, the brighter the average. FIG. 4D shows projection waveforms on the x-axis at various film reduction amounts (from (i) to (v), the larger the number, the smaller the film reduction). This index value also increases as the film thickness decreases. According to this method, there is an advantage that the SEM noise reduction effect by the averaging of the wide area is very good. However, since it is easily affected by a shape change other than the film reduction, particularly when it is accompanied by a shape change at the pattern top portion, it is an effective index value for an object having a small shape change other than the film reduction. In addition, in the graphs shown in i) to (v) of FIG. 4D, the description of the vertical axis and the horizontal axis is omitted, but these are the vertical axis and horizontal axis of the graph shown in FIG. Same as axis.

  (D): Other texture analysis methods such as those described in Non-Patent Document 2, that is, a power spectrum in a region corresponding to the upper part of the pattern of the SEM image is obtained, and the roughness of the texture is quantified from the frequency features. A method, a combination of the various methods described above, or a method of weighting and adding various index values can also be applied.

[How to create database 20]
FIG. 1C shows a flow for creating the database 20 stored in the database unit 113.
(S11): In order to create the database 20, samples having resist patterns formed under various exposure conditions are prepared. For example, an FEM (Focus Exposure Matrix) wafer in which the focus position and the dose amount of the exposure conditions are changed in a range where there is a possibility of fluctuation may be used on one wafer.
(S12): SEM images of the prepared resist pattern samples formed under various exposure conditions are acquired, and any one of (i) to (2) described above in [Calculation of film reduction index value] The film loss index value is calculated by the method.
(S13): For the same sample, the height of the resist pattern is measured by height measurement means such as AFM (atomic force microscope), scatterometry (light wave scattering measurement), or pattern cross-section observation by SEM, Determine the amount of resist film loss. Instead of actually measuring the height of the pattern, the height obtained by calculating the cross-sectional shape of the resist pattern under the above various exposure conditions by exposure simulation may be used.
(S14): The relationship between the film loss and the film thickness index value thus obtained is shown in graph 21. As described above, the relationship is such that the greater the amount of film reduction, the greater the film reduction index value. For this result, an approximate function is calculated as the relationship between the film reduction amount and the film reduction index value. At the same time, the variation of the film reduction amount with respect to the film reduction index value may be calculated. Further, a look-up table format may be used instead of the approximation function.
(S15): The approximate function calculated as the relationship between the film reduction amount and the film reduction index value is stored in the database. Further, the relationship of the variation in the amount of film reduction with the calculated film reduction index value is stored in a database.

[Comparison with database and estimation of film loss]
The steps S3 and S4 in FIG. 1) b) will be described in detail. In step S3 of FIG. 1B, the index value calculated in S2 is substituted into the approximate function calculated as the relationship between the film reduction amount and the film reduction index value in the database 20, or the film reduction amount and By referring to a lookup table based on the relationship between the film thickness reduction index values, the absolute value of the film thickness reduction amount is estimated. Further, from the relationship of the variation of the film loss amount with respect to the film decrease index value stored in the database 20, the index value calculated in S2 can be used to grasp the error of the film decrease amount estimation result.

[GUI]
FIG. 5 shows an example of a GUI (Graphic User Interface) screen 500 as an example of an input / output screen in the input / output unit 115 of the SEM system according to the present embodiment. This GUI screen is provided with an input screen necessary for storing the relationship between the film loss amount and the film thickness index value in advance to construct and update the database 20 of the database unit 113.

  The GUI screen 500 has an input portion 501 for the name of the database 20 to be newly registered or updated, and information selected from the database 20 according to the input name is displayed as a data list 506 and a graph 507. The Further, there is a part 502 for selecting a measurement recipe for acquiring a film thickness reduction index value for registration, and an image display area 504 for displaying an image 503 registered in the selected recipe. There is a portion 505 for selecting a region on the SEM image used for calculation of the film thickness reduction index value.

  After inputting these items, a cursor (not shown) is moved on the button 508 for executing the recipe and a mouse (not shown) is clicked to calculate the film thickness reduction index value based on the above set conditions. Is executed. The calculated film reduction index value data is automatically updated, and the data list 506 is also corrected. In addition, on the data list 506 displayed on the screen 500, film thickness reduction data acquired by another device or means can be input. When the input data is stored in the database 20, an update button 509 may be clicked. By executing this, the relationship between the film reduction amount and the film reduction index value is functionalized and registered in the database 20.

  In addition, the input / output unit 115 of the SEM system further displays a GUI screen 600 including a setting unit for estimating the amount of film loss and an output unit for displaying the result by collating with information in the database 20. To do. FIG. 6 shows an example of the GUI screen 600.

  When setting the measurement conditions, a portion 601 for selecting the database 20 for calculating the amount of film loss to be referred to and a region used for the film thickness reduction index value on the SEM image automatically acquired by the selection are displayed based on the database information. A portion 602. In addition, a part (film reduction evaluation image selection unit) 603 for selecting an image for evaluating the film reduction amount from the acquired image set, film thickness reduction index value calculation and film reduction amount for the selected image An execution button 604 for instructing execution of estimation is provided. A portion 606 for displaying the selected image 605 is also provided.

  Further, a measurement recipe selection part (film reduction evaluation recipe selection unit) 607 and an execution button 608 for executing film reduction index value calculation and film reduction amount estimation based on the selected recipe. By moving a cursor (not shown) on the execution button 608 on the screen 600 and clicking a mouse (not shown), calculation of a film reduction index value and estimation of a film reduction amount are performed based on the set conditions. And the result is displayed as a graph 610, and is displayed as a data list 609 with an estimation error.

  As described above, according to the present invention, since it is possible to perform film thickness reduction detection or film thickness reduction measurement using the image used for the conventional line width measurement, the throughput of the conventional process management can be reduced. It is possible to monitor film loss without reducing it. Further, the present invention may be applied to the determination of exposure process conditions, that is, optimization of the dose amount and focus position during exposure. Compared to the conventional method using only the line width, a more effective process window can be set by incorporating the conditions without causing film loss.

  A second embodiment of the present invention is shown in FIG. In the first embodiment, after calculating the film reduction index value from the SEM image (S2), the absolute value of the film reduction amount is estimated (S4) by collating with the database (S3), and the film reduction index is calculated. In the second embodiment, the film reduction index value is calculated only from the SEM image (S2), the film reduction amount estimation in S3 and S4 is not performed, and the film reduction is performed. Only information related to the index value is output (S6). The method for calculating the film thickness reduction index value from the SEM image in this embodiment (S2) is the same as the method described in the first embodiment.

  The GUI screen of the input / output unit in the present embodiment is basically the same as that described with reference to FIGS. 5 and 6, and there is no column for the film reduction amount in the data list 506 of FIG. 5. The data list 609 is different in that the film reduction amount estimation result column and the estimation error column are missing. Another example of the output according to the present embodiment is shown in FIG.

  According to the present embodiment, since the index value represents a relative change in the amount of film loss, it is possible to monitor the process variation by monitoring the trend of the index value as shown in FIG. 7B. It is.

  A third embodiment of the present invention is shown in FIG. In the first and second embodiments, the method of measuring the resist pattern formed on the wafer has been described. In this embodiment, the resist pattern is etched using the resist pattern as a mask as the resist pattern is reduced. Another object of the present invention is to detect a reduction in the line width of a circuit pattern formed on a wafer.

  In the present embodiment, instead of the database in which the correspondence relationship between the film loss amount and the film reduction index value in the first embodiment is registered, the resist line width, the etching bias, that is, the line width of the resist pattern, Etching bias is estimated from the film thickness index value measurement result using the database 30 in which the correspondence relationship between the pattern line widths after etching is registered (S23, 24).

  Hereinafter, a method for creating the database 30 and an SEM system equipped with this method will be described.

[SEM system]
FIG. 8A shows the configuration of a length measuring SEM system having a film thickness reduction measuring function according to this embodiment. The length measurement SEM system according to the present embodiment includes an SEM main body 80, an image processing / overall control unit 81, an arithmetic processing unit 82, and a database unit 83, and is connected to a data server 84 via a network. The configurations of the SEM main body 80 and the image processing / overall control unit 81 are the same as those described in the first embodiment with reference to FIG.

  In this embodiment, the SEM image output from the image processing / overall control unit 81 is used to perform processing while referring to the information stored in the database unit 83 by the arithmetic processing unit 82 to extract information on the etch bias. The The result is sent to the data server 116 via the communication line and stored.

  The arithmetic processing unit 82 includes a film thickness reduction index value measuring unit 821, a database collating unit 822, an etch bias estimating unit 823, and an input / output unit 824 including a display screen.

[Overall flow]
FIG. 8B shows an overall flow of etch bias estimation performed by the arithmetic processing unit 82.
(S21): A resist pattern is imaged by the SEM 80, and a signal obtained by the imaging is processed by the image processing / overall control unit 81 to obtain an SEM image of the sample.
(S22): An index value for film reduction is calculated from the acquired image by the same method as in step S2 in the first embodiment.
(S23): The correspondence between the line width of the resist pattern stored in the database unit 83 and the pattern line width after etching is collated with the registered database 30. A method for creating the database 30 will be described later with reference to FIG.
(S24): Estimate the etch bias based on the collation result.
(S25): The estimated result is displayed on the display screen of the input / output unit 824. The data is sent to the data server 84 via the communication line and stored.

  The above is the overall flow of etch bias estimation in this embodiment. Details of the main flow will be described below.

[How to create a database]
FIG. 8C shows a flow of creating the database 30 in which the correspondence relationship between the line width of the resist pattern and the pattern line width after etching is registered in this embodiment.
(S31): As described in the description of S11 in the first embodiment, a sample on which a resist pattern is formed under various exposure conditions is prepared.
(S32): Similar to S12 in the first embodiment, the film reduction index value is calculated and the line width of the pattern is measured.
(S33): Similar to the description of S13 in the first embodiment, the height of the resist pattern is measured to determine the resist film reduction amount.
(S34): The wafer on which the resist pattern is formed is etched.
(S35): The line width of the pattern formed on the wafer by performing the etching process is measured. The wafer used in this step may be the same wafer used in S32 and S33, or may be a different wafer produced by the same process. The measured line width data of the pattern formed on the etched wafer is stored in correspondence with the resist pattern dimension measured in S32 and the resist pattern height data measured in S33.
(S36): The correspondence relationship between the etching bias, which is the difference between the line width of the resist pattern obtained in S32 and the line width of the etched pattern obtained in S35, and the resist film reduction obtained in S33 is obtained. . An example of this determination is shown in graph 31. Based on this result, an approximate function is calculated, and the relationship between the etching bias and the film reduction index value is obtained.
(S37): The relationship between the obtained etching bias and the film reduction index value is stored in the database 30. At the same time, the film decrease index value at which the etching bias increases is stored as a threshold, and the variation of the etching bias with respect to the film decrease index value is calculated and stored in the database 30.

[Estimation of etching bias]
In the step of estimating the etch bias in S24 of FIG. 8B, the etching bias is estimated by substituting the film reduction index value calculated in S22 into the approximate function on the database 30 calculated in S36. Alternatively, it is determined whether or not the etching bias exceeds a threshold value.

[GUI]
FIG. 9 shows an example of the GUI screen 900 as an example of the input / output screen in the input / output unit 84 of the SEM system according to the present embodiment. This GUI screen is the same as the GUI screen of the first embodiment shown in FIG.

  More specifically, the GUI screen 900 includes an input portion 901 for the name of the database 30 to be newly registered or updated, and information selected from the database 30 according to the input name is displayed in the data list 906 and Displayed as a graph 907. In the data list 906 in this embodiment, in addition to the resist pattern film reduction index value, the resist CD value (SEM) is used instead of the film reduction amount column in the data list 506 of the first embodiment shown in FIG. The value of the resist pattern dimension obtained from the image) and the etch CD value (the value obtained by obtaining the dimension of the pattern formed on the wafer after etching from the SEM image), the line width of the resist pattern (resist CD value) and the post-etching value The etching bias value that is the difference from the line width (etch CD value) of the pattern is displayed. The graph 907 displayed in the present embodiment is different from the case of the first embodiment shown in FIG. 5 in that the relationship between the etch bias and the film reduction index value is displayed.

  On the other hand, a part 902 for selecting a measurement recipe for acquiring a film reduction index value for registration, an image display area 904 for displaying an image 903 registered in the selected recipe, and a film on the displayed image 903 A portion 905 for selecting an area on the SEM image used for calculation of the reduction index value, a button 908 for executing these recipes after inputting these items, and calculation of the film reduction index value based on the set conditions, The point that the data is automatically updated and the data list 906 is corrected is the same as in the first case. In addition, on the data list 906 displayed on the screen 900, it is possible to input data on the amount of film loss acquired by another device or means. When the input data is stored in the database 30, an update button 909 may be clicked. By executing this, the relationship between the film reduction amount and the film reduction index value is functionalized and registered in the database 30.

  Further, the input / output unit 824 of the SEM system further displays a GUI screen 1000 having a setting unit for estimating the amount of film loss and an output unit for displaying the result by collating with information in the database 30. To do. The GUI screen 1000 has a configuration similar to the GUI screen 600 of FIG. 6 described in the first embodiment.

  As described in the first embodiment, when setting the measurement conditions, a portion 1001 for selecting a reference film reduction amount database 30 and a film reduction index value on the SEM image automatically acquired by the selection. A region 1002 is displayed based on the database information. In addition, a part (film reduction evaluation image selection unit) 1003 for selecting an image for evaluating the film reduction amount from the acquired image set, and a film reduction index value calculation and a film reduction amount for the selected image. An execution button 1004 for instructing execution of estimation is provided. Further, a portion 1006 where the selected image 1005 is displayed is also provided.

  The GUI screen 1000 further includes a part for selecting a measurement recipe (recipe selection unit for film reduction evaluation) 1007 and an execution button 1008 for executing film reduction index value calculation and film reduction amount estimation based on the selected recipe. Have. By moving a cursor (not shown) on the execution button 1008 on the screen 1000 and clicking a mouse (not shown), the film thickness index value and the etch bias are calculated based on the set conditions. Then, the estimation error of the film reduction amount and the NG determination are executed, the result is displayed in the data list 1009, and the relationship between the etch bias and the film reduction index value is displayed as a graph 1010.

  According to the present embodiment, as shown by a graph 31 in FIG. 8, the relationship between the film reduction amount and the pattern size after etching is not linear, and when the film reduction amount exceeds a certain value, the etch bias rapidly increases. There is a characteristic. According to the present embodiment, it is possible to determine whether or not the generated film loss is at a level that should be a problem in terms of device characteristics.

DESCRIPTION OF SYMBOLS 10,80 ... Electron microscope 109, 81 ... Image processing and whole control part 110 ... Arithmetic processing part 113 ... Database part 116 ... Server 500 ... GUI 600 at the time of database update ... GUI for setting film reduction estimation condition and outputting result 82... Arithmetic processing unit 83... Database unit 84... Server 900... For registering the relationship between etch bias and film reduction index value GUI 1000... GUI for outputting an etch bias estimation result

Claims (14)

Image acquisition means for acquiring an image of a sample having a resist pattern formed on the surface;
Quantifying means for processing the image acquired by the image acquiring means to quantify the brightness variation characteristics of an image of a desired region of the resist pattern;
An electron microscope system, comprising: an index value calculation unit that calculates an index value that associates the characteristics of variations in image brightness quantified by the quantification unit with an etching bias of a circuit pattern when the resist pattern is used as a mask. The electron microscope system according to claim 1,
An electron microscope system, further comprising: an estimation unit that estimates an etching bias of the circuit pattern using the index value calculated by the index value calculation unit. The electron microscope system according to claim 1 or 2,
The index value calculating means uses the etching bias of the circuit pattern when the characteristics of the variation in brightness of the image quantified by the quantifying means are etched as a mask under the normal condition of the resist pattern as a reference value An electron microscope system characterized in that an index value associated with an amount of decrease from the reference value is calculated. The electron microscope system according to claim 2,
The estimating means estimates the etching bias by applying the index value calculated by the index value calculating means to a database indicating the relationship between the etching bias index value and the etching bias value that has been created and stored in advance. An electron microscope system characterized by: The electron microscope system according to any one of claims 1 to 4,
An electron microscope system characterized in that the quantification means processes the image acquired by the image acquisition means to quantify the degree of roughness of a desired region of the resist pattern. The electron microscope system according to any one of claims 1 to 4,
The quantification unit processes the image acquired by the image acquisition unit to quantify the average brightness of a desired region of the resist pattern. The electron microscope system according to any one of claims 1 to 4,
The quantification unit processes the image acquired by the image acquisition unit to quantify the brightness distribution of a desired region of the resist pattern. An image acquisition step of acquiring an image of a sample having a resist pattern formed on the surface;
A quantification step of processing the image acquired in the image acquisition step to quantify the brightness variation characteristics of an image of a desired region of the resist pattern;
An evaluation method comprising: an index value calculation step of calculating an index value that correlates an image brightness variation characteristic quantified in the quantification step with an etching bias of a circuit pattern when etching is performed using the resist pattern as a mask. The evaluation method according to claim 8,
And an estimating step of estimating an etching bias of the circuit pattern using the index value calculated in the index value calculating step. The evaluation method according to claim 8 or 9, wherein
When the index value calculating step uses the etching bias of the circuit pattern when the resist pattern is etched as a mask under the normal condition of the characteristics of the image brightness quantified in the quantifying step as a reference value An index value associated with the amount of decrease from the reference value is calculated. An evaluation method according to claim 9, wherein
The estimation step estimates the etching bias by applying the index value calculated in the index value calculation step to a database indicating the relationship between the etching bias index value and the etching bias value that has been created and stored in advance. An evaluation method characterized by: The evaluation method according to any one of claims 8 to 11,
In the quantification step, the image acquired in the image acquisition step is processed to quantify the degree of roughness of a desired region of the resist pattern. The evaluation method according to any one of claims 8 to 11,
In the quantification step, the image acquired in the image acquisition step is processed to quantify an average brightness of a desired region of the resist pattern. The evaluation method according to claim 8. Because
In the quantification step, the image acquired in the image acquisition step is processed to quantify the brightness distribution of a desired region of the resist pattern. .

Images