JP2009014739A - Irregularity determination method of sample, and charged particle radiation device - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide method and device especially suitable for determining irregularity of line & space pattern formed on sample. <P>SOLUTION: This method is performed as follows; tilt charged particle radiation so as to oblique against its optical axis or tilt sample stage to be scanned on the sample, measure spread of detected signal to line scanning direction of the charged particle radiation, compare the above spread with the one when the charged particle radiation has been scanned along the optical axis, then determine irregularity condition of the above scanning point based on variation in quantity of the spread. <P>COPYRIGHT: (C)2009,JPO&INPIT

Description

  The present invention relates to a method for determining unevenness of a pattern or obtaining three-dimensional information, and more particularly to a method and apparatus suitable for obtaining unevenness information of a line and space pattern formed on a semiconductor wafer.

  A charged particle beam apparatus such as a scanning electron microscope is an apparatus suitable for measuring and observing a pattern formed on a semiconductor wafer that is becoming finer. Conventionally, as a method for obtaining three-dimensional information of a sample with a charged particle beam apparatus, there is a stereo observation method as disclosed in JP-A-5-41195.

  In the stereo observation method, two tilted stereo images are taken, stereo matching is performed between the two images, and the corresponding points are calculated to calculate the height, thereby obtaining three-dimensional information. .

  Japanese Patent Application Laid-Open No. 5-17596 discloses a technique for measuring a pattern dimension by irradiating a beam obliquely to a pattern on a sample.

JP-A-5-41195 Japanese Patent Laid-Open No. 5-17596

  When measuring a line and space pattern on a sample with a scanning electron microscope, there is a problem that if the line and space widths are almost equal, it is difficult to discriminate the line and space pattern. In addition, the stereo matching method has a problem in that it is difficult to obtain a good three-dimensional image due to problems such as the S / N ratio of the image obtained by a scanning electron microscope, the resolution, and the structure of the sample. That is, when the S / N ratio and the resolution are low, it is difficult to find a corresponding point for matching two images, and as a result, there may be a blurred image that is not sufficiently matched. Furthermore, since the stereo matching method requires high-level image processing, there is also a problem that processing time is required.

  Further, the technique disclosed in Japanese Patent Laid-Open No. 5-17596 is not mentioned up to the line and space determination.

  An object of the present invention is to determine unevenness formed on a sample by a simple method, or to obtain three-dimensional information, and in particular, a determination method suitable for determining unevenness of a line & space pattern formed on a sample, And providing an apparatus.

  In the present invention, the charged particle beam is scanned on the sample so that the charged particle beam is inclined with respect to the optical axis of the charged particle beam, and the detection is performed based on the detection of the charged particle emitted from the scanning location. Measure the spread of the charged particle beam in the line scanning direction of the signal, compare with the spread when the charged particle beam is scanned along the optical axis, and based on the increase and decrease of the spread Determine the state.

  According to the configuration as described above, it is easy to determine the unevenness in the charged particle beam image, and in particular, it is easy to determine the uneven state of a pattern in which similar patterns such as line and space patterns are continuous. become.

  Other objects and other specific configurations of the present invention will be described in detail in the following embodiments of the present invention.

  According to the present invention, it is possible to easily realize unevenness determination on a sample.

  FIG. 3 is a block diagram showing a schematic configuration of an electron microscope apparatus which is an embodiment of the image processing apparatus of the present invention. Reference numeral 301 denotes a mirror body part of an electron microscope, and an electron beam 303 emitted from an electron gun 302 is converged by an electron lens not shown in the drawing, and is irradiated onto a sample 305. By the electron beam irradiation, the intensity of secondary electrons or reflected electrons generated from the sample surface is detected by the electron detector 306 and amplified by the amplifier 307.

  A deflector 304 moves the position of the electron beam, and causes the electron beam to be raster scanned on the sample surface by a control signal 308 of a control computer 310 (also referred to as a control processor or a control system in the present invention). . The signal output from the amplifier 307 is AD converted in the image processor 309 to create digital image data. Further, a profile is created from the digital image data by projection processing. Hereinafter, the digital signal waveform is also called a profile.

  Reference numeral 311 denotes a display device that displays the image data. Reference numeral 309 denotes an image memory for storing digital image data and an image processing processor for performing various image processing, and has a display control circuit for performing display control. Input means 312 such as a keyboard and a mouse is connected to the control computer 310.

  An electron microscope apparatus is used to measure the line width of a fine pattern drawn on a wafer when creating a semiconductor device. Here, when the line width measurement part on the wafer is a line or space, if the width of the line and the space is the same, the distinction cannot be made and it is necessary to distinguish from the three-dimensional information. Has been. The present embodiment relates to a scanning electron microscope that can obtain three-dimensional information of a line and space sample by a simple method, but of course is not limited thereto, and other charged particles such as a focused ion beam apparatus. The present invention can also be applied to a wire device.

An address signal corresponding to the memory location of the image memory is generated in the control computer 31 and converted into an analog signal, and then supplied to the deflector 304 via a scanning coil control power supply (not shown). For example, when the image memory has 512 × 512 pixels, the X-direction address signal is a digital signal that repeats 0 to 512, and the Y-direction address signal is positive when the X-direction address signal reaches 0 to 512. 1 is a digital signal repeated from 0 to 512. This is converted into an analog signal.

  Since the address of the image memory corresponds to the address of the deflection signal for scanning the electron beam, a two-dimensional image of the deflection region of the electron beam by the deflector 304 is recorded in the image memory. Signals in the image memory can be sequentially read out in time series by a read address generation circuit (not shown) synchronized with a read clock. The signal read corresponding to the address is converted into an analog signal and becomes a luminance modulation signal of the display device 311.

  The image memory has a function of storing (combining) images (image data) in an overlapping manner for improving the S / N ratio. For example, a single completed image is formed by storing images obtained by eight two-dimensional scans in an overlapping manner. That is, a final image is formed by combining images formed in one or more XY scanning units. The number of images (frame integration number) for forming one completed image can be arbitrarily set, and an appropriate value is set in consideration of conditions such as secondary electron generation efficiency. Further, an image desired to be finally acquired can be formed by further overlapping a plurality of images formed by integrating a plurality of sheets. When the desired number of images is stored, or after that, blanking of the primary electron beam may be executed to interrupt information input to the image memory.

  The sample 305 is disposed on a stage (not shown) and can move in two directions (X direction and Y direction) in a plane perpendicular to the sample 305 electron beam.

  The apparatus according to the present invention has a function of forming a line profile based on detected secondary electrons or reflected electrons. The line profile is formed based on the amount of detected electrons when the primary electron beam is scanned one-dimensionally or two-dimensionally or the luminance information of the sample image. The obtained line profile is, for example, on a semiconductor wafer. It is used for measuring the dimension of the formed pattern.

  In the description of FIG. 3, the control computer is described as being integrated with or equivalent to the scanning electron microscope. However, the control computer is of course not limited thereto, and will be described below with a control processor provided separately from the scanning electron microscope body. Such processing may be performed. In this case, a detection signal detected by the secondary signal detector 13 is transmitted to the control processor, or a signal is transmitted from the control processor to a lens or a deflector of the scanning electron microscope, and via the transmission medium. An input / output terminal for inputting / outputting a transmitted signal is required.

  Alternatively, a program for performing the processing described below may be registered in a storage medium, and the program may be executed by a control processor that has an image memory and supplies necessary signals to the scanning electron microscope.

[Example 1]
FIG. 1 shows a change in a peak portion of an edge profile at the time of tilt measurement in a line image. In addition, although the Example described below demonstrates the example which inclines an electron beam using the deflector 304 with respect to the optical axis of an electron beam, it is not restricted to it, The sample irradiated with an electron beam is demonstrated. By tilting, scanning may be performed while tilting the electron beam with respect to the normal to the sample surface direction. In this case, a sample stage having a tilt function may be used, or the electron beam column itself may be tilted. Further, the electron beam may be inclined using a deflector different from the scanning deflector.

  As shown in FIG. 1, the apparent width W when the tilt amount θ is changed can be obtained by one formula from the width T and the step H of the tilted region. If attention is paid only to the amount of change in the inclined part during the inclination, it depends only on the step H.

W = T + H * tanθ (1)
FIG. 2 shows a profile change when the convex portion is photographed from the direction of + θ or −θ. Here, the half width of the + θ direction projection is HP1 and HP2, and the half width of the −θ direction projection is HM1 and HM2. When shooting from the -θ direction, the half-value width (HM1) of the signal at the left edge portion increases, and when shooting from the + θ direction, the half-value width (HP1) of the signal at the right edge portion increases. In the present embodiment, the half width of the peak is used as a reference. However, the present invention is not limited to this, and any position where the length increases or decreases depending on the inclination of the electron beam may be used at any position of the peak.

  As described above, since the half-value width of the edge signal in the photographing direction is increased, the use of this phenomenon makes it possible to determine the unevenness of the line and space based on the increase or decrease of the half-value width.

  That is, in the scanning direction of the electron beam, when the width of the profile obtained when the electron beam is tilted with respect to the optical axis of the electron beam becomes large, the portion corresponding to the profile is in the direction of the electron beam tilt. It can be determined that there is a portion where a side surface (edge) is formed. In other words, when the spread of the detection signal in the line scanning direction (X direction in the case of two-dimensional scanning) is measured and the electron beam is irradiated along the optical axis, It can be determined that there is an edge surface on the inclination direction side of the line.

  On the other hand, when the width of the profile is reduced, it can be determined that the portion corresponds to the back surface (opposite side) when viewed from the tilt direction of the electron beam. In other words, it can be seen that when one of the two peak widths appearing when scanning is performed with the electron beam tilted and the other becomes smaller, there is a concave or convex portion between them. Also, in this state, when there is a peak whose width increases on the side where the electron beam is inclined, a convex portion exists, and when a peak whose width decreases on the side where the electron beam is inclined, It can be determined that there is a recess.

  According to the above configuration, the unevenness determination on the sample can be performed without employing a complicated image processing technique. In addition, since it is possible to perform unevenness determination by a simple method compared to image processing of forming a line profile, it is easy to incorporate in automatic measurement of a semiconductor inspection apparatus, and also to improve the throughput of the inspection process. effective.

  In particular, it becomes easy to determine the uneven state of a pattern in which similar patterns such as line & space patterns are continuous. In addition, accurate unevenness determination can be performed regardless of the composition forming the line and space. Even when there is no difference in image quality between the line pattern and the space, the unevenness determination can be performed accurately.

[Example 2]
In the present embodiment, an example of performing scanning not only from one direction but also from both directions of ± θ will be described as a method for further improving the accuracy of unevenness determination.

  Depending on the electron beam, it may be difficult to judge the increase / decrease in the half-value width in the inclined scanning only in one direction, such as when the half-value width of the left and right edge signals is originally different. In comparison, the increase / decrease in the half-value width is reversed between the left and right peaks, so that the determination can be made easily.

  FIG. 4 is a diagram for explaining an unevenness determination method using a profile photographed from ± θ directions. When HM1 (first peak) has a width equal to or greater than HP1 (third peak) and HP2 (fourth peak) is determined to have a width equal to or greater than HM2 (second peak) If the pattern is determined to be a convex pattern, and if HM1 is smaller than HP1 and HP2 is determined to be smaller than HM2, the pattern is determined to be a concave pattern as described above. It is possible to determine the unevenness of the line and space pattern with high accuracy. In the present embodiment, an example is described in which unevenness estimation between two peaks is performed when HP1 ≦ HM1 and HP2 ≧ HM2, and when HP1> HM1 and HP2 <HM2, but this is not the only example. For example, it may be determined that “HP1 <HM1, HP2> HM2 → convex portion” and “HP1 ≧ HM1, HP2 ≦ HM2 → concave portion”. This is the same in the embodiments described later.

  Further, depending on the pattern, it is possible to make a determination using the left and right edge heights ((TM1 or TM2) and (TP1 or TP2)) of the profile. In addition, it is also possible to use a value (HM1 * TM1) or an area obtained by multiplying the half width and the height.

Example 3
Next, a more preferable unevenness determination method when there are a plurality of line & space patterns at equal intervals will be described with reference to FIG.

  (1301) is an image obtained by scanning an electron beam from directly above in such an image that the contrast of the scanning electron microscope image is obtained only by the edge effect. The pattern that requires the unevenness determination is such a pattern, and the measurement position (feature part) that should come to the center of the image due to the control error of the apparatus or the pattern shape difference is, for example, the measurement position (1312) of the pattern cross-sectional view (1302). In such a case, it is difficult to recognize the measurement position (1312) from the image (1301).

  Therefore, it is determined whether each edge is the left or right edge of the line pattern from the two profiles (1303) obtained by tilting the electron beam in the vertical direction ± by the above-described method using the half width and the peak height.

  For example, assume that the right / left determination result of seven edges is as shown in (1304). Here, the symbol R indicates the right edge, and the symbol L indicates the left edge. symbol? Indicates a case where determination could not be made because there was no difference in half-width or peak height. Symbol R * in (1304) is an example in which an edge is erroneously determined due to variations in pattern shape or noise.

  Next, two templates (1305) and (1307) are created for seven edges using the premise that the pattern to be recognized next is a line & space pattern. Each template is compared with the left and right edge determination result (1304), and the template with the most correct answers (here, (1305)) is selected as correct. That is, the characteristic part on the sample is specified by a template having a high accuracy rate of the conditions defined by the template (in this embodiment, the arrangement of lines and spaces). A symbol that could not be determined using the correct template at this time? It is possible to correct an edge or an erroneously determined edge. According to this method, since the unevenness is determined using a plurality of edges, the reliability of the unevenness determination can be improved.

  In this embodiment, an example of using a line & space pattern template in which lines and spaces are alternately repeated has been described. However, the present invention is not limited to this. It is possible to confirm the unevenness determination on the sample. In this embodiment, the example of two templates has been described, but three or more templates may be used.

Example 4
In the third embodiment, the method for estimating the information of the edge that cannot be determined by referring to the template has been described. There may be a problem with the optical conditions of the scanning electron microscope, or there may be a problem with the line and space pattern itself. In the technology for determining the accuracy rate of the unevenness using the template described in the third embodiment, in principle, either one of the accuracy rates should be high and the other accuracy rate should be low. That is, if any accuracy rate is high or low, there may be a problem in the optical conditions of the scanning electron microscope and the line & space pattern itself.

  In such a case, the left and right edge determination results are stored in association with at least one of a line profile, a scanning electron microscope image, an optical condition of the scanning electron microscope, a semiconductor wafer manufacturing history, and the like, and can be read thereafter. If it is assembled, the cause of the determination error can be easily pursued.

  If the unevenness determination is not successful, it is often difficult to perform further measurements. Therefore, an error message may be generated to alert the operator. Furthermore, in the case of a sample having a large number of measurement points or observation points on a single wafer, such as a semiconductor wafer, the measurement or observation of the portion where the error has occurred is skipped, and the next measurement or observation is performed. You may make it do.

  In the case of a semiconductor wafer, if there are 2 or more points that cannot be determined, there may be some problem with the scanning electron microscope, or there may be a problem with the entire semiconductor wafer. Measurement or observation by an electron microscope may be interrupted, or measurement or observation of the semiconductor wafer may be interrupted.

  FIG. 15 is a process flow of unevenness determination using a template.

  In steps 1501 to 1504, the electron beam is scanned so that a line profile in a direction perpendicular to the edge direction of the line & space pattern can be formed. In the present embodiment, the unevenness determination by electron beam scanning from ± θ described in the second embodiment is performed. In step 1505, the correct answer rate is calculated using template 1. In step 1506, the correct answer rate is calculated using the template 2.

  At this time, when the correct answer rate T1 of the template 1 is equal to or greater than the predetermined value n1, and the correct answer rate T2 of the template 2 is smaller than the predetermined value n2 (the lower table (2) in FIG. 15), the pattern is a pattern estimated by the template 1 (Step 1508). Further, when the correct answer rate T1 of the template 1 is smaller than the predetermined value n1 and the correct answer rate T2 of the template 2 is equal to or larger than the predetermined value n2 (the lower table (3) in FIG. 15), the pattern is a pattern estimated by the template 2. Is determined (step 1509). If the above two conditions are not met (tables (1) and (4) in FIG. 15), it is determined that measurement is impossible and error processing is performed (step 1509).

  In the above embodiment, one template (for example, template 1) corresponds to the pattern, and the other template (for example, template 2) defines a shape different from the pattern. That is, the accuracy of determination can be further improved by selecting information (line profile, image data, etc.) using templates that define different shapes or conditions (image data, etc.).

  When it is determined as (1) and (4) in the lower table of FIG. 15, as described above, the left and right edge determination results include the line profile, the scanning electron microscope image, the optical conditions of the scanning electron microscope, or the semiconductor wafer. If a sequence that can be stored in association with each other's manufacturing history is stored and then read out, it becomes easy to identify the cause of pattern estimation, and subsequent measurement or feedback to the manufacturing process is performed. It becomes easy.

  In particular, when the line pattern is continuously formed to be inclined in the same direction, for example, there may be an error in the apparatus conditions of the optical exposure apparatus in the semiconductor production line. Therefore, if the fact is stored as incidental information, the cause can be identified more easily. The state in which the pattern is tilted in a specific direction is discovered based on the phenomenon that the half width of the profile formed by the pattern edge in the tilt direction of the electron beam becomes thick only when the electron beam is irradiated from one tilt direction, for example. be able to.

Example 5
FIG. 5 is an example in which the obtained concavo-convex profile is displayed superimposed on the photographed image. By displaying in this way, it is possible to easily determine the line or space on the sample image on which the unevenness determination on the scanning electron microscope image is difficult. FIG. 6 is a processing flow for determining unevenness of a line & space image. A projection waveform photographed from each inclination direction is input at 601, and a half-value width of each peak is calculated at 602. In 603, the processing from 601 to 602 is performed on the projection waveform in the ± θ direction. In 604, the half-value width change of the corresponding peak in the ± θ direction is calculated, and the unevenness is determined from the change by the determination method shown in FIG. In step 605, a profile representing unevenness is output.

Example 6
FIG. 7 is a processing flow when the present invention is used for specifying the position of a line or space when measuring a length using a line & space image. A measurement condition of a scanning electron microscope (SEM) is set in 701, and a beam irradiation angle or a stage angle is tilted by + θ in 702. Shooting is performed at 703. Since the projection waveform is used for the unevenness determination, in this case, the entire frame image may be captured, or data for several lines may be measured. In 704, the data measured in 703 is transferred to the image processor. In 705, the beam irradiation angle or stage angle is tilted by −θ in the same manner as in 702, the image is taken in 706, and the data is transferred in 707. At 708, the slope at 705 is restored. In 709, the unevenness determination is performed by the image processor from the data transferred in 704 and 707.

  Finally, at 710, the coordinates of the convex portion are output if it is a line portion, and the coordinates of the concave portion are output if it is a space portion. Here, when a plurality of uneven portions are detected, the coordinates closest to the center may be output.

  According to the above configuration, visual field alignment, length measurement, and inspection position alignment can be easily performed accurately without using complicated image processing techniques such as pattern matching.

Example 7
FIG. 8 shows a profile change when the convex portion is photographed from directly above and from the + θ direction or the −θ direction. Here, the full width at half maximum of the profile directly above is HE1, HE2, the full width at half maximum of the + θ direction profile is HP1, HP2, and the full width at half maximum of the −θ direction profile is HM1, HM2.

  As described with reference to FIG. 2, it is possible to determine unevenness that is difficult from one direction by comparing the changes in the half-value width taken from the two directions of the directly upward direction and the + θ direction or the −θ direction. In the case of the present embodiment, the unevenness determination can be performed with a single inclination, so that the processing speed can be improved as compared with the case where the electron beam is inclined to ± θ.

  FIG. 9 shows an unevenness determination method using a profile photographed from one of the directly upward direction and the positive / negative direction.

  FIG. 10 and FIG. 10 show the processing flow when the present invention is used for specifying the position of a line or space when length measurement is performed by the unevenness determination method using a profile photographed from one of the directly upward direction and the positive / negative direction. 11 will be used for explanation.

The observation conditions of the scanning electron microscope (SEM) are set in (1001), the beam is scanned from directly above in (1002), and the profile or image is transferred to the image processor 309 as data 0 (1003). At this time, reference is made to the conditions None (1103), One Side (1105), and Double Side (1104) of the setting item Line & Space Detection (1102) on the pattern detection condition setting screen (1101). However, this condition is set in advance.

If it is None (1103) in the unevenness determination execution of (1004), the position detection (1013) is executed using data 0 without executing the unevenness determination. If it is One Side (1105) or DoubleSide (1104), after tilt setting 1 is performed in (1005), the beam is scanned in (1006) and the profile or image is transferred to image processor 309 as data 1 ( 1007).

  In (1008), the unevenness determination is executed using data 0 and data 1. The determination method at this time is the method described in FIG. If the unevenness determination is possible in (1008), position detection (1013) is executed using the unevenness profile obtained here. If not, beam tilt setting 2 is performed in the direction opposite to tilt setting 1 (1005) (1009). After the beam tilting, the beam is similarly scanned and transferred to the image processor 309 as data 2 (1010) (1011). In (1012), the unevenness determination is executed using data 1 and data 2, and if the determination is possible, the position detection (1013) is executed using the unevenness profile.

If not, position detection is performed using data 0, data 1, data 2, or synthesized data. After the position detection, a length measurement value is calculated (1014). By processing as described above, the data 0 for calculating the length measurement value can suppress the beam irradiation to the minimum, leading to improvement in length measurement reproducibility. In addition, the determination that the line width and the space width are almost the same and the line and the space are difficult to distinguish can be set by the user at the time of registration, and pattern damage due to beam irradiation for unevenness determination can be suppressed.

Example 8
Next, a method for detecting a pattern using a beam tilted image will be described with reference to FIG. The observation condition of the scanning electron microscope (SEM) is set in (1201), and the beam is scanned in the direct upward direction in (1202), and the image 0 is transferred to the image processor 309 (1203). The image processor 309 executes pattern detection using a search algorithm such as normalized correlation with reference to a template registered in advance (1204).

  Success of pattern detection is determined based on a preset threshold value, and if successful, the process ends. If it is determined to be unsuccessful, the beam in the direction set at the time of registration or the direction obtained by calculation is tilted (1205) and scanned to obtain image 1 (1206). This is transferred to the image processor 309 (1207) and synthesized with the image 0 which is not inclined. For composition, image 0 and image 1 may be added together, or only image 1 may be used.

  Pattern detection is executed in the same manner as (1204) using the composite image (1210). In this method, it is also possible to detect a pattern using only an inclined image without executing (1202) to (1204). As described above, the amount of information at the edge of the image obtained by tilting the beam and performing pattern detection increases, enabling pattern detection with a small number of frames and improving throughput. Further, the amount of beam irradiation can be reduced, and damage caused by beam irradiation can be reduced.

  When determining the tilt direction by calculation at the time of registration, the template image is differentiated into four directions including the up / down / left / right direction and the diagonal direction to obtain the sum of absolute values, and the angle with the maximum direction is 90 degrees. Choose the direction. This is effective when the edge is biased in a specific direction.

Example 9
Another method using electron beam inclination for pattern detection will be described with reference to FIG. Usually, when image recognition is performed using a template image for positioning or the like, a template is registered in advance and matching with the template is performed at the time of inspection.

  In template registration, an area (1402) to be a template is selected using the input unit 312 on the image (1402) scanned from directly above. The selected partial image is transferred to the image processor 309, and a direction image is calculated using an operator for obtaining the direction (for example, Sobel filter).

  Next, a distribution map of the obtained direction image is created, and an angle (a plurality of angles) equal to or greater than a predetermined threshold value is obtained. Since these angles indicate the distribution of pattern edges, if the electron beam is inclined in this direction, an image with a large amount of information can be obtained due to the edge effect. (1403) is an image obtained by tilting the electron beam from the vertical direction to the right, and the right edge of the pattern (1404) is emphasized. That is, the pattern shape can be specified by extracting the portion where the edge is emphasized.

  Similar processing can be performed for a plurality of directions including vertical and horizontal directions and diagonal directions. Further, if the same processing is applied to the line pattern that penetrates up and down the image, it can be determined that only the left and right inclinations are necessary. The tilt image obtained as described above is registered, and template matching is performed at the time of inspection. This method makes it possible to recognize a low step pattern with high reliability. If the shape of the pattern is known in advance, if the slope is selectively tilted in the direction in which the edge can be emphasized (ideally, the direction perpendicular to the longitudinal direction of the edge), the wasteful slope is reduced. Pattern edges can be made clear without doing so.

Example 10
In the embodiments so far, the technology for determining the unevenness on the sample based on the increase / decrease in the peak width of the line profile obtained as a result of scanning with the electron beam tilted has been described. An example will be described in which signal measurement is performed while changing the focal position of the image, and the unevenness determination of the image composed of the line and the space is performed by comparing the focus evaluation values of the line and the space area in each image.

  FIG. 16 is a block diagram of a schematic configuration of a scanning electron microscope apparatus which is an embodiment of the image processing apparatus of the present invention. This scanning electron microscope incorporates an automatic focus control function. In FIG. 16, reference numeral 1601 denotes a sample stage, 1602 denotes a sample to be photographed on the sample stage, 1604 denotes a cathode, 1605 denotes a scanning coil, 1606 denotes an electronic lens, 1608 denotes a scanning coil control circuit, and 1609 denotes a lens control circuit. The electron beam 1614 is scanned on the sample 1602 by the scanning coil 1605, and the electrons emitted from the sample 1602 are detected by the detector 1603. A signal S1 from the detector 1603 is input to the AD converter 1607 and converted into a digital signal S2. The digital signal of S2 is input to the image processor 1610, image processing and feature amount extraction are performed, and the result is sent to the control computer 1611. The processed image is sent to the display device 1612 and displayed. The focus control signal S3 from the control computer 1611 is input to the lens control circuit 1609, and the focus control can be performed by adjusting the excitation current of the lens 1606. Reference numeral 1613 denotes input means connected to the control computer 1611.

  The automatic focus control in the scanning electron microscope configured as described above is a control for automatically setting the focus condition of the electron lens to an optimum value, and the method can be performed while changing the condition of the electron lens. A focus evaluation value is calculated and evaluated from detection signals of secondary electrons and reflected electrons obtained by scanning one frame, and an optimum value is set as a condition of the electron lens.

  An electron microscope apparatus is used to measure the line width of a fine pattern drawn on a wafer when creating a semiconductor device. Here, when the line width measurement part on the wafer is a line or space, if the width of the line and the space is the same, the distinction cannot be made and it is necessary to distinguish from the three-dimensional information. Is done. Since the present invention relates to a charged particle beam apparatus that can obtain unevenness information of a line and a space sample by a simple technique, it can be applied to the scanning electron microscope apparatus of FIG.

  FIG. 17 shows a case where the line and the uneven portion of the pattern sample are focused. F1 is a case where the line part (convex part) is in focus, and Fn is a case where the space part (concave part) is in focus. As described above, since the focal position is different in the uneven portion, the unevenness is determined by using the information.

  FIG. 24 is a diagram showing changes in the focus evaluation value when the excitation current of the electron lens is changed. Here, as the focus evaluation value, a differential value between pixels, a secondary differential value, a Sobel value, a Laplacian value, or the like is used. When the focus evaluation value is calculated for each image taken while changing the condition of the electronic lens, the focus evaluation value becomes the maximum value under the condition where the electronic lens is in focus. In FIG. 24, when the excitation current value of the electron lens is Fi, the focus evaluation value becomes the maximum value, so Fi is a focused condition.

  FIG. 25 shows the relationship between the excitation current and the focus evaluation value in the line portion and the space portion. As described with reference to FIG. 24, since the focus evaluation value is maximized under the focused condition, when the line portion (convex portion) is in focus, the focus evaluation value of the line portion (convex portion) is maximized. However, the focus evaluation value is small because the space portion (the concave portion) is not focused. The opposite is also true when the space portion (concave portion) is in focus.

  FIG. 26 shows the relationship between the excitation current and the focal length. Since the focal length is determined by the excitation current as shown in FIG. 26, if the excitation current at the time of in-focus is known, the focal length can be estimated from the value of the excitation current. That is, if the excitation current at the time of focusing is determined for two different areas, the level of the two areas can be determined.

  FIG. 18 shows a schematic diagram until the focus evaluation value is calculated. In order to change the focal position, the excitation current is changed from F1 to Fn, and the corresponding images G1 to Gn are acquired. Next, a focus evaluation filter (differential, second-order differential, Sobel, Laplacian, etc.) is applied to each of the G1 to Gn images to create focus evaluation images Gf1 to Gf2, and focus evaluation values FE1 to FEn are calculated. . Here, as the focus evaluation value, a sum of all pixel values of the focus evaluation value image, an average value thereof, a variance value, or the like can be used. The steps so far are usually executed as autofocus, and the excitation current value at the time of maximum of FE1 to FEn is set as the excitation current at the time of focusing.

  FIG. 19 shows projection waveforms in line and space images. In the electron microscope image, the edge portion of the line is imaged with high brightness by the edge effect. Therefore, since the projection waveform shows a peak at the boundary between the line and the space as shown in FIG. 19, it can be seen that the portion sandwiched between the peaks is a line or a space.

  FIG. 20 shows a method for calculating the focus evaluation values of the line portion and the space portion. As shown in FIG. 19, a projection waveform is obtained from the focused image in autofocus, and regions (1) to (5) for calculating focus evaluation values are determined. Focus evaluation values FE11 to FEn5 are calculated in the regions (1) to (5) of the focus evaluation value images Gf1 to Gfn. Next, for the region (1), the maximum value (FEi1) of FE11 to FEn1 is obtained, and the excitation current value (Fi) for focusing in the region (1) is obtained from the corresponding excitation current. Similarly, the steps (2) to (5) are performed, and the focus matching excitation current in each region is obtained. The focal length of the area is obtained from the magnitude of the focus current value for focusing in (1) to (5), the height of the area is judged, and the unevenness is judged.

  FIG. 23 shows a processing flow according to an embodiment of the present invention. Reference numeral 2301 denotes a part for obtaining an image (Gi) and a focus evaluation image (Gfi) with different focus described with reference to FIG. 18, and this part can be executed simultaneously with autofocus. Reference numeral 2302 denotes a portion for calculating the focus evaluation value described with reference to FIG. 19, obtaining a focus matching excitation current, and obtaining a projection waveform from the corresponding image. 2303 is a part for obtaining the focus evaluation values (FEmn) of the peak-to-peak areas (1) to (5) described with reference to FIG. Reference numeral 2304 denotes a portion for obtaining the excitation current value for focusing in the peak-to-peak areas (1) to (5) described with reference to FIG. Reference numeral 2305 denotes a portion for determining the level of (1) to (5) from the exciting current value obtained in 2304 and determining the unevenness.

  FIG. 21 is a processing flow in the case where the present invention is used for specifying the position of a line or space when length measurement is performed using a line and space image. A scanning electron microscope (SEM) measurement condition is set in 2101, autofocus is performed in 2102, and imaging is performed in 2103. In 2104, unevenness determination is performed according to the processing flow shown in FIG. Finally, in 2105, if it is a line portion, the coordinates of the convex portion are output, and if it is a space portion, the coordinates of the concave portion are output. Here, when a plurality of uneven portions are detected, the coordinates closest to the center may be output.

  FIG. 22 is an example in which the obtained concavo-convex profile is displayed superimposed on the photographed image.

  In this embodiment, the unevenness determination of the line and space is performed by calculating a score evaluation value of a portion corresponding to the line and space from an image acquired at the time of autofocus, obtaining a focal length from an excitation current at the time of focusing, and from the value In order to obtain the unevenness information of the image, the unevenness information can be obtained by a simple method without using complicated image processing such as matching processing. Moreover, since the obtained uneven | corrugated information is used for position determination, the specific error of the length measurement point in a line & space image can be reduced. In addition, since necessary information can be collected at the autofocus timing, it is not necessary to provide a new process for obtaining unevenness information at another timing, which can contribute to an improvement in throughput.

The figure which shows the change of the edge at the time of the inclination measurement in a line profile. The figure which shows the profile change at the time of image | photographing a convex part from the direction of +/- theta. BRIEF DESCRIPTION OF THE DRAWINGS Schematic of the scanning electron microscope which is one Example of this invention. The figure explaining the unevenness | corrugation determination method using the profile image | photographed from +/- theta direction. The figure which shows the example which displayed and superimposed the obtained uneven | corrugated profile on the image | photographed image. Processing flow for unevenness determination of line & space images. A processing flow when the present invention is used to specify the position of a line or space. The figure which shows the profile change at the time of image | photographing a convex part from a directly upward direction and + (theta) direction or-(theta) direction. The figure which shows the unevenness | corrugation determination method using the profile image | photographed from any one of the directly upward direction or the positive / negative direction. A processing flow when the present invention is used to specify the position of a line or space. Pattern detection condition setting screen. A processing flow in the case of pattern detection using an image obtained by tilting a beam. The figure explaining the unevenness | corrugation determination method using a template. The figure explaining the example which emphasizes the edge of a pattern by the beam inclination of multiple directions. Processing flow of unevenness determination by template. 1 is a schematic configuration diagram of a scanning electron microscope apparatus according to an embodiment of the present invention. The figure showing the case where a line and an uneven part of a pattern sample are focused. Schematic until calculating a focus evaluation value. The figure which shows the projection waveform in a line and a space image. The figure which shows the calculation method of the focus evaluation value of a line part and a space part. The figure which shows the processing method at the time of using this invention for the position specification of a line or a space. The figure which shows the example which displayed and superimposed the obtained uneven | corrugated profile on the image | photographed image. The figure which shows the processing flow which is one Example of this invention. The figure which showed the change of the focus evaluation value at the time of changing the excitation current of an electronic lens. The figure which shows the relationship between the excitation current and focus evaluation value of a line part and a space part. The figure which shows the relationship between an exciting current and a focal distance.

Explanation of symbols

  DESCRIPTION OF SYMBOLS 301 ... Body part of an electron microscope, 302 ... Electron gun, 303 ... Electron beam, 304 ... Deflector, 305 ... Sample, 306 ... Electron detector, 307 ... Amplifier, 308 ... Control signal, 309 ... Image processor, 310 ... control computer, 311 ... display device, 312 ... input means.

Claims (3)

Obtained at a charged particle source, a stage system on which an observation sample is placed and moved, a scanning deflection system that scans a charged particle beam emitted from the charged particle source on the sample, and a charged particle beam irradiation location of the sample In a charged particle beam apparatus equipped with a detection system for detecting a signal,
The width of the peak of the line profile relating to one edge portion obtained when the line and space pattern on the sample is irradiated from the first irradiation direction that is inclined from the optical axis of the charged particle beam is When it is larger than the peak width of the line profile relating to the first edge portion obtained when the first irradiation direction is irradiated from the second irradiation direction on the side opposite to the tilt direction of the first irradiation direction. The charged particle beam device characterized in that it is determined that a space portion exists on the side of the tilt direction with respect to the edge portion of 1, and a line portion exists on the side opposite to the tilt direction. In claim 1,
A charged particle beam apparatus comprising: means for displaying the unevenness information of the obtained line and space image superimposed on an image captured as a profile waveform. In claim 1,
A charged particle beam apparatus characterized in that the unevenness information of the obtained line and space image is used for specifying the position.

Images