JP2013030277A - Charged particle beam apparatus and measuring method - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a charged particle beam apparatus and a measuring method that enable the dimensions of a pattern bottom to be measured at high accuracy even with a high-aspect ratio structure.SOLUTION: In a method for measuring the dimensions of a pattern bottom of a sample on which a pattern with a high aspect ratio is formed, a comparison is made between a first scanned image and a second scanned image that are based on signal charged particles that are generated when a charged particle beam is scanned onto the same region of the sample in a first direction and in a second direction inverse to the first direction, the scanned image having a lower signal intensity is taken to be the true value, the scanned image is reconstructed, and the reconstructed scanned image is used to measure the dimensions of the pattern bottom formed on the sample. Also provided is a charged particle beam apparatus for implementing this method.

Description

  The present invention relates to a charged particle beam apparatus and a measurement method for measuring a dimension of a fine circuit pattern such as a semiconductor device or a liquid crystal.

  In semiconductor device manufacturing lines, circuit pattern dimension management is positioned as an indispensable technique for yield improvement and quality control. For this dimension management, a CD-SEM (Critical-Dimension Scanning Electron Microscope) using an electron microscope that realizes high spatial resolution is used. The CD-SEM uses the high spatial resolution and deep focal depth peculiar to the SEM to measure the dimensions of the circuit pattern in the lateral direction (in-plane direction of the circuit pattern), thereby realizing sub-nm measurement accuracy. For example, Patent Documents 1 and 2 disclose conventional electron beam type inspection apparatuses and inspection methods.

JP 2004-163420 A Japanese National Patent Publication No. 08-506665

  In order to realize further high integration of semiconductor devices, in recent years, in the field of lithography, a spot (hot spot) where an error is likely to occur at the time of transfer is observed with a CD-SEM, and an observed image and CAD (Computer Aided Design at design) ) A method for measuring a transfer error from a comparison with data has been proposed. For example, Patent Document 1 discloses a technique for acquiring an image by scanning an electron beam in a reciprocating manner to accurately measure the shape of a pattern, and using a signal of a pattern contour opposite to the penetration direction of the electron beam. Yes. Further, in Patent Document 2, focusing on the fact that the transferred pattern is an insulator, a technique is disclosed in which an electron beam is scanned in a reciprocating manner and the signals are added and averaged in order to eliminate image obstruction caused by charging. Yes. These are techniques for accurately measuring the shape of the transferred resist, and can be said to be very effective in managing the state of the exposure apparatus.

  However, in order to achieve high integration of semiconductor devices, it is indispensable to improve not only lithography but also etching for processing circuit patterns. In order to realize devices of recent 2 × nm nodes, processing with an aspect ratio of 10 or more is required, and it is said that it is indispensable to measure the bottoms of processed grooves and holes with a CD-SEM. FIG. 2 is a graph showing the relationship between the aspect ratio and the signal electrons emitted from the bottom of the groove and the hole. The vertical axis represents 100 as the signal electrons emitted from the upper surface of the groove. The decline is shown as a percentage. The solid line and the broken line in the figure indicate the signal electrons emitted from the center of the groove bottom and the signal electrons emitted from the corners, and the expected angle of the signal electrons emitted as the aspect ratio increases ( Since α) becomes smaller, the signal electrons decrease monotonously. The difference between the signal electrons emitted from the central portion and the corner portion of the groove indicates the accuracy required for detecting the signal electrons when measuring the bottom of the groove and the hole. In a general resist pattern measurement, since the aspect ratio is around 1, if the signal detected with an accuracy of about 4% is imaged, the bottom of the groove aligned holes can be measured. However, in a pattern that has been etched, it is necessary to image a signal detected with an accuracy of 0.05% at an aspect ratio of 10 and 0.01% at an aspect ratio of 60, which is extremely high for a signal detection system of a CD-SEM. It turns out that accuracy is required.

  On the other hand, FIG. 3A shows the response delay of each part generally used in the detection system of SEM, and FIG. 3B shows the signal waveform observed when the detection system of FIG. 3A is used. FIG. 3A shows a response delay of a phosphor that converts signal electrons into light as a representative example, and a photomultiplier tube that converts and amplifies the converted light again. In addition, since the rise time is sufficiently fast, the afterglow time and the fall time are expressed as response delay. In catalogs of each part, 90-10% of the signal intensity is often defined as afterglow time and fall time, and the afterglow time of the phosphor shown in FIG. 3A is less than 100 ns. Fall time is said to be less than 30ns. However, when discussing the accuracy of 0.01-0.05% described above using this detection system, it is necessary to consider a response delay of several μs for a phosphor and a little less than 100 ns for a photomultiplier tube. FIG. 3B shows a cross-sectional shape of the pattern and a signal waveform obtained when the electron beam is scanned from the left to the right of the drawing with respect to the pattern. When the electron beam enters the convex part of the pattern, the response of the phosphor and the photomultiplier tube is sufficiently fast, so that a signal waveform corresponding to the bottom corner of the groove is obtained, but the electron beam goes from the convex part to the concave part. When entering, the signal waveform corresponding to the bottom corner of the groove cannot be observed due to the response delay. In other words, it can be said that it is extremely difficult to measure the bottom of the high aspect structure by the conventional technique.

  In order to realize further higher integration of semiconductor devices in the future, not only the measurement of the pattern transferred by lithography, which was taken up in the prior art, but also the measurement of the high aspect structure completed by etching is indispensable. An object of the present invention is to provide a charged particle beam apparatus and a measurement method capable of measuring a dimension of a pattern bottom with high accuracy even in a high aspect structure.

  As an embodiment for achieving the above object, a deflector that scans a charged particle beam converged on a predetermined region of a sample on which a concavo-convex pattern is formed, and scans the predetermined region of the sample with the charged particle beam. In the charged particle beam apparatus comprising: a detector that detects charged particles generated from the sample, and a signal processing unit that processes a scanning image based on the charged particle signal detected by the detector. Is a first scanned image based on a charged particle signal from a predetermined region of the sample when the charged particle beam is scanned in the first direction, and a second direction opposite to the first direction. A memory that records a second scanned image based on a charged particle signal from the predetermined region of the sample, and a comparison operation circuit that compares the first scanned image and the second scanned image recorded in the memory; , The first scan In the second scan image and, it was the correct value the signal strength of the scanned image is small, it is a charged particle beam apparatus characterized in that and a signal processing circuit to reconstruct the scanned image.

  Further, the measurement includes a step of scanning the charged particle beam converged on the sample on which the concavo-convex pattern is formed, storing the generated signal charged particles as a scanned image, and measuring the size of the sample using the scanned image. In the method, the charged particle beam converged in the first scanning direction is scanned and the generated signal charged particles are stored as the first scanning image, and the same region of the sample is opposite to the first scanning direction. Scanning the charged particle beam converged in the second scanning direction and storing the generated signal charged particles as a second scanned image, and comparing the first scanned image and the second scanned image, and scanning signal A measurement including: a step of setting a smaller intensity as a true value and reconstructing a scanned image; and a step of measuring a dimension of a pattern bottom formed on the sample using the reconstructed scanned image. The method.

  According to the present invention, the above configuration can provide a charged particle beam apparatus and a measurement method capable of measuring the dimension of the pattern bottom with high accuracy even in a high aspect structure.

BRIEF DESCRIPTION OF THE DRAWINGS It is a schematic block diagram (partial cross section figure) of the charged particle beam apparatus (SEM type semiconductor measuring device) which concerns on a 1st Example. It is a figure which shows the relationship of the signal electron discharge | released from the bottom of the aspect ratio and a groove | channel and a hole. It is a figure which shows the responsiveness of each component generally used for the detection system of SEM. It is a figure which shows the signal waveform observed when the detection system of FIG. 3A is used. It is a flowchart for demonstrating an example of this Embodiment. It is a schematic block diagram of the signal processing part in the charged particle beam apparatus (SEM type semiconductor measuring device) which concerns on a 1st Example. It is a flowchart for demonstrating the process of the signal waveform in the charged particle beam apparatus (SEM type semiconductor measuring device) which concerns on a 1st Example. It is a figure for demonstrating the alignment process of the signal waveform in the charged particle beam apparatus (SEM type semiconductor measuring device) which concerns on a 1st Example. It is a figure for demonstrating the reconstruction process of the signal waveform in the charged particle beam apparatus (SEM type semiconductor measuring device) which concerns on a 1st Example. It is a flowchart for demonstrating the procedure of the recipe preparation in a charged particle beam apparatus (SEM type semiconductor measuring device). It is a flowchart for demonstrating the procedure of the recipe preparation in the charged particle beam apparatus (SEM type semiconductor measuring device) which concerns on a 1st Example. It is a schematic diagram of the GUI screen of recipe creation in the charged particle beam apparatus (SEM type semiconductor measuring device) according to the first embodiment. It is a schematic block diagram of the correction process of the detection system characteristic in the charged particle beam apparatus (SEM type semiconductor measuring device) which concerns on a 2nd Example. It is a schematic block diagram (partial cross section figure) of the probe current measuring device in the charged particle beam device (SEM type semiconductor measuring device) concerning the 2nd example. It is a flowchart of the correction process of the detection system characteristic in the charged particle beam apparatus (SEM type semiconductor measuring device) which concerns on a 2nd Example. It is a schematic diagram of the GUI screen of correction | amendment of the detection system characteristic in the charged particle beam apparatus (SEM type semiconductor measuring device) which concerns on a 2nd Example. It is a schematic block diagram of the signal processing part in the charged particle beam apparatus (SEM type semiconductor measuring device) which concerns on a 3rd Example. It is a flowchart for demonstrating the process of the signal waveform in the charged particle beam apparatus (SEM type semiconductor measuring device) which concerns on a 3rd Example. It is a schematic block diagram (partial cross section figure) of the charged particle beam apparatus (SEM type semiconductor measuring device) concerning the 4th example. It is a schematic block diagram (partial cross section figure) of the detection system in the charged particle beam apparatus (SEM type semiconductor measuring device) concerning the 4th example. It is a schematic block diagram of the signal processing part in the charged particle beam apparatus (SEM type semiconductor measuring device) which concerns on a 4th Example. It is a figure for demonstrating the alignment process of the signal waveform in the charged particle beam apparatus (SEM type semiconductor measuring device) which concerns on a 4th Example. It is a schematic diagram of the GUI screen of recipe creation in the charged particle beam apparatus (SEM type semiconductor measuring device) which concerns on a 4th Example. It is a schematic block diagram of the signal processing part in the charged particle beam apparatus (SEM type semiconductor measuring device) which concerns on a 5th Example. It is a figure for demonstrating the scanning image formation in the charged particle beam apparatus (SEM type semiconductor measuring device) which concerns on a 6th Example. It is a figure for demonstrating extraction of the edge in the charged particle beam apparatus (SEM type semiconductor measuring device) which concerns on a 6th Example. It is a figure for demonstrating the reconstruction process of the signal waveform in the charged particle beam apparatus (SEM type semiconductor measuring device) which concerns on a 6th Example.

  In this embodiment, in order to measure the pattern size of the bottom part of the high aspect structure, a plurality of scanning images having different scanning directions are acquired for the same region on the pattern, and only the rising response of the detection system is used therefrom. Reconstruct the scanned image. FIG. 4 is a diagram showing an example of the dimension measurement process in the present embodiment, and is an example in which the signal waveform is reconstructed by applying this process to the pattern having the cross-sectional shape shown in FIG. 3B. As described in FIG. 3B, when the electron beam is scanned from left to right with respect to the paper surface, a signal waveform corresponding to the bottom corner of the groove is obtained on the left side of the convex portion, and a signal waveform corresponding to it on the right side is I cannot get it. On the other hand, when the electron beam is scanned from right to left with respect to the paper surface, a signal waveform corresponding to the bottom corner of the groove is obtained on the right side of the convex portion in the reverse manner as described above, and the signal waveform corresponding to it on the left side is I cannot get it. In order to reconstruct the signal waveform using only the rising response of the detection system from these two, in this process, the two signal waveforms are compared, and the signal waveform having the lower signal strength is set as a true value and the signal waveform is reconstructed. By reconstructing, a signal waveform corresponding to the corner of the bottom of the groove can be extracted, and the measurement of the dimension of the bottom of the groove becomes possible for the first time.

  In this process, since at least two scanned images are acquired at the same location, it takes twice or more time for imaging compared to the conventional method. However, it has been found that the use of two scanning images that are orthogonal to the pattern and differ in the scanning direction by about 180 degrees enables measurement of the bottom dimension of the high aspect structure, which has been difficult in the past. Although the present embodiment has been described with reference to dimension measurement as an example, this technique is also effective in inspection, and can be applied to various applications from development of semiconductor devices and monitoring of process variations in mass production. It is.

  Hereinafter, the embodiment will be described in detail.

  In the present embodiment, a configuration example of an SEM type semiconductor measuring device suitable for measuring the bottom of a high aspect structure will be described. In the embodiment, the case where an electron beam is used will be described, but the present invention can also be applied when an ion beam is used. The SEM type semiconductor measuring device, which is one of the charged particle beam devices described in the present embodiment, stores two scanned images in the memory, which are perpendicular to the pattern at the same location on the wafer and differ in scanning direction by about 180 degrees. To do. A comparison is made between the signal waveforms obtained from the respective scanned images, the signal waveform having a lower signal intensity is set to a true value, the signal waveform is reconstructed, and the dimension of the pattern is measured using the reconstructed waveform. .

FIG. 1 shows a configuration of an SEM type semiconductor measuring apparatus according to the present embodiment. Hereinafter, a basic configuration of the SEM type semiconductor measuring device of this embodiment will be described. The measuring apparatus is roughly divided into an SEM casing 1, a sample chamber 2, a casing control section 3, a signal processing section 4, a stage control section 5, a wafer transfer section 8, and a vacuum exhaust section 9, all of which are consoles. 6 can be controlled. The console 6 has a large-capacity storage medium 7 capable of storing recipes, measurement results, and acquired scanned images. Based on the data recorded on the storage medium 7, the console 6 operates and manages the data. Here, the vacuum is maintained in the SEM casing 1 and the sample chamber 2 by the exhaust pump included in the vacuum exhaust section 9, and the vacuum exhaust section 9 is controlled from the console 6 according to the purpose, and the SEM casing 1 and the sample chamber 2 are controlled. 2. The sample preparation chamber 10 can be exhausted or leaked. Hereinafter, the structure and the function performed by each part will be described in order. In addition, the same code | symbol shows the same component.
(SEM housing)
The SEM housing 1 includes an electron source 11, a first condenser lens 12, a diaphragm 13, a probe current measuring unit 14, a second condenser lens 15, a conversion plate 16, a deflector 17, an objective lens 18, and a height sensor 19. The The electron beam emitted from the electron source 11 is converged by the first condenser lens 12 as the primary electron beam 20 and then irradiated to the diaphragm 13 that limits the probe current. The primary electron beam 20 that has passed through the diaphragm 13 and whose probe current is limited is converged again by the second condenser lens 15 so as to pass through a hole provided in the center of the conversion plate 16. The primary electron beam 20 that has passed through the conversion plate 16 is converged and irradiated onto the sample 21 by the objective lens 18 after the trajectory is changed by the deflector 17 so as to scan a desired region on the sample 21 two-dimensionally. The Here, a voltage for decelerating the primary electron beam 20 is applied to the sample 21 installed in the sample chamber 2, and the signal electrons 22 emitted from the sample 21 have energy corresponding to the voltage applied to the sample 21. To be accelerated. The accelerated signal electrons 22 pass through the objective lens 18 and the deflector 17 and collide with the conversion plate 16, and then are supplemented by a detector 23 to which a high voltage is applied so as to draw the signal electrons 22. The above is a general configuration of the SEM casing, but in this embodiment, since the two scanning images differing in the scanning direction by about 180 degrees are compared, the scanning signal needs to be accurately inverted between the two scanning images. There is. When the frequency of the scanning signal is slow, an electromagnetic method may be used for the deflector 17. However, when the frequency becomes fast, an eddy current due to an alternating magnetic field is generated in the SEM casing, and it is difficult to accurately invert the scanning signal. Become. When the frequency of the scanning signal is slow, signal electrons per pixel increase, so that a high S / N scanning image can be obtained. However, on the other hand, since the irradiation time of the primary electrons becomes long, charging of the sample 21 may cause an image defect. In such a case, only by adopting the electrostatic system for the deflector 17 described above, scanning in a wide frequency band becomes possible, and image disturbance due to charging of the sample 21 can be reduced. As a result, the present embodiment can also be applied to an insulator sample in which image damage due to charging is likely to occur.
(Sample chamber, stage control unit, and wafer transfer unit)
The sample chamber 2 includes a stage 24, an insulator 25, a sample folder 26, and a mirror 27. The sample folder 26 and the grounded stage 24 are electrically insulated by an insulating material 25, and the sample 21 and the mirror 27 are electrically grounded to the sample folder 26. A high voltage can be applied from outside the chamber 2 through a feedthrough. By applying a high voltage from the retarding power source 28 to the sample folder 26, the high voltage is applied to the sample 21, and the energy of the primary electron beam 20 incident on the sample 21 can be arbitrarily adjusted.

  Further, the stage 24 is driven two-dimensionally in a direction perpendicular to the central axis of the SEM casing 1 by the stage driving device 30 in the stage controller 5, and the entire region of the sample folder 26 is moved to the central axis of the SEM housing 1. It can be moved directly below. The mirror 27 is attached to the sample folder 26 in order to measure the position of the sample 21, and the laser is transmitted from the laser measuring device 29 in the stage control unit 5 through a glass window that partitions the vacuum of the sample chamber 2. It becomes the composition which can be irradiated. By measuring the position of the sample 21 with the laser measuring device 29 in this way, a scanned image at a desired position can be obtained even with a semiconductor pattern in which fine patterns are integrated.

The wafer transfer unit 8 includes a transfer control unit 35 and a transfer robot 36. In the wafer transfer unit 8, the transfer control unit 35 controls the transfer robot 36 based on a control signal from the console 6, and transfers the sample 21 installed in the wafer transfer unit 8 to the sample preparation chamber 10. Here, the sample 21 is transferred stepwise from the wafer transfer unit 8 to the sample preparation chamber 10 and the sample chamber 2, and a valve 37 is provided between each unit. The console 6 controls the valve 37 and the vacuum exhaust unit 9 and can automatically transport the sample 21 so that the vacuum in the sample chamber 2 can be always maintained even during the transporting operation.
(Case control unit)
The housing control unit 3 operates the electron source 11 and various lenses included in the SEM housing 1 based on a control signal sent from the console 6. The housing control unit 3 includes an electron gun power supply 31, a housing control power supply 32, a probe current measuring device 33, and a retarding power supply 28. The electron gun power supply 31 applies a cathode voltage to the electron source 11 and stably emits electrons. Then, the electron source 11 is operated. The housing control power supply 32 can supply current to each of the first condenser lens 12, the second condenser lens 15, and the objective lens 18, and can set the current supplied to each lens based on a control signal from the console 6. The probe current measuring device 33 can measure the probe current of the primary electron beam 20 detected by the probe current measuring unit 14. The retarding power supply 28 can apply a high voltage to the sample folder 26 in accordance with the control signal from the console 6 as described above.
(Beam control circuit and signal processor)
The beam control circuit 34 and the signal processing unit 4 form a scanned image of the sample 21 based on a control signal sent from the console 6. In order to form a scanning image, the beam control circuit 34 sends a scanning signal to the deflector 17, and the signal processing unit 4 samples the signal electrons 22 taken in by the detector 23 in synchronization with the scanning signal. Reference numeral 39 denotes a level adjustment circuit, and reference numeral 41 denotes an AD conversion unit.

  Hereinafter, the signal processing unit 4 which is a feature of the present embodiment will be described in detail. FIG. 5 is a schematic configuration diagram of the signal processing unit 4. The signal processing unit 4 includes a switching circuit A 42, a memory A 43, a memory B 44, a switching circuit B 45, an image processing circuit 46, a comparison operation circuit 47, and a signal processing. The circuit 48 is configured. The signal processing unit 4 is connected to the detector 23 via the level adjustment circuit 39 and the AD conversion unit 41 and to the console 6 by electrical wiring. Based on the control signal from the console 6, the signal from the detector 23 is connected. Are processed and output to the console 6. In the control from the console 6, the level adjustment circuit 39 adjusts the amplification and offset of the signal sent from the detector 23, and the AD converter 41 converts it into a digital signal in synchronization with the scanning signal. Then, the digital signal is stored as a scanned image in the memory A 43 or the memory B 44 via the switching circuit A 42, and the scanned image stored in each memory is selected via the switching circuit B 45 to perform image processing. Output to the circuit 46. Here, the image processing circuit 46 has a filter function such as smoothing and sharpening, a level adjustment function such as linearity of lightness and nonlinear conversion, and a data compression function with respect to the input scanned image. These functions can be freely selected by an operator who controls the console 6. Normally, as described above, since the signal electrons emitted from the bottom of the high aspect structure are very few, the brightness of the scanned image has a dynamic range of about 4 digits (= 0.01% accuracy) for measurement. Therefore, the data size of the scanned image becomes very large. However, the signal waveform to be measured shown in this embodiment and the scanned image to be displayed are independent data, and the scanned image has a function that can be adjusted to an optimum image quality by visual observation, so that the measurement accuracy is not a concern. The data size of the scanned image can be compressed. As a result, a scanned image that requires a four-digit dynamic range (14 bits) can be compressed to that required for a normal grayscale image (8 bits).

  Next, signal waveform processing in the signal processing unit 4 will be described. 6 extracts a signal waveform from the scanned images stored in the memory A 43 and the memory B 44 in FIG. 5 through the comparison operation circuit 47 and reconstructs the signal waveform through the signal processing circuit 48. FIG. 4 schematically shows a flow until measurement. The memory A 43 and the memory B 44 store scanning images that are reversed in left and right directions because the scanning directions are different by about 180 degrees. First, the comparison operation circuit 47 inverts one of the images horizontally, and performs alignment processing between the scanned images. Here, the alignment process forms a difference image from the gradation difference for each pixel of the scanned image, and shifts one of the scanned images so that the standard deviation of the difference image is minimized while shifting one of the scanned images. Adjust the position. In the alignment process, the scanned image itself may be used, or an image obtained by extracting only phase information by Fourier transform may be used. Next, based on the alignment information, a cursor for extracting a waveform is aligned with each scanned image, and a signal waveform is extracted. Here, the setting of the cursor for extracting the waveform will be described in detail later (recipe creation). Then, the two extracted signal waveforms are compared with each other, aligned with higher accuracy, and then the signal waveform is reconstructed and measured by the signal processing circuit 48. As described above, the comparison operation circuit 47 aligns the position in two stages of the scanned image and the signal waveform. The alignment with respect to the scan image is necessary to prevent the above-described cursor for extracting the signal waveform from being shifted between two scan images having different scan directions. In addition, even if the cursor is positioned and the signal waveform in the same region is extracted, the height and position of the signal waveform do not completely match due to contamination of the sample or noise superimposed on the scanning signal. This is necessary. As described above, by aligning the positions in two steps, the present embodiment can be applied even when the fields of view of the two scanned images are slightly shifted due to charging of the sample 21, contamination of the sample, damage, or the like.

  FIG. 7 is a diagram for explaining an example of signal waveform alignment processing, and shows two-stage processing of level correction and alignment for each signal waveform extracted from the memory A 43 and the memory B 44. The level correction is a process for adjusting the height of the signal waveform. As shown in FIG. 7, the average gradation (S1) of the low gradation area and the average gradation (S2) of the high gradation area are defined. The comparison operation circuit 47 expands and contracts the gradation of each signal waveform so that S1 and S2 are the same in the signal waveform. Here, in FIG. 7, the peak of the signal waveform is not S2. This is because when the level is corrected, the peak height of the signal waveform varies and correction is not performed erroneously due to the error of S2. If the peak height of the signal waveform is substantially constant, the average gradation of the peak There is no problem even if S2 is set. The alignment processing after level correction is a process for reconstructing a continuous signal waveform. As shown in FIG. 7, the pixels of each signal waveform are set so that the intermediate positions between the peaks of the signal waveform match. Move or stretch. Here, FIG. 7 shows an example in which there are two intermediate positions between peaks. However, the more intermediate positions to be aligned, the more accurate alignment can be performed. The operator can specify the number of intermediate positions between the peaks S1 and S2 and the peaks automatically extracted by the comparison operation circuit 47 at the stage of creating the recipe and match the position of the signal waveform.

  FIG. 8 is a diagram for explaining an example of a signal waveform reconstruction process, showing two signal waveforms input from the comparison operation circuit 47 to the signal processing circuit 48, and signal waveform reconstruction and smoothing using them. ing. Reconstruction can be realized by comparing two signal waveforms in the same manner as in FIG. 4 and reconstructing the signal waveform with a lower signal strength as a true value. However, even with the above-described alignment, discontinuous points 49 exist in the reconstructed signal waveform, which causes measurement errors. Therefore, the signal processing circuit 48 smoothes the reconstructed signal waveform to reduce the measurement error. By adding a smoothing function to the reconstruction process in this way, a measurement apparatus that is stable and has few measurement errors can be realized. In addition, although the data size of the signal waveform is not specified in the above, high-accuracy measurement can be performed with the minimum necessary data size by setting the scan image displayed on the console 6 to 8 bits and the signal waveform used for measurement to 14 bits or more. Can be realized. By reducing the data size, a large amount of measurement results can be stored in the storage medium 7 and the data processing can be speeded up, which is advantageous from the viewpoint of the processing speed of the measuring device. By performing the alignment process and the reconstruction process described above, it is possible to realize stable dimension measurement for the bottom of the high aspect structure.

In this embodiment, the switching circuit A 42 subsequent to the AD conversion unit 41, the image processing circuit 46, and the signal processing circuit 48 are described as parts independent of the console 6. However, these functions are included in the console 6. It can also be done with a calculator. When the above processing is performed by a computer included in the console 6, since the parts subsequent to the AD conversion unit 41 can be concentrated on the console, it is possible to expect a reduction in the installation area and manufacturing cost of the measuring device, and the effects of this embodiment are as follows. Not damaged.
(Recipe creation procedure)
A procedure for creating a recipe in the SEM type semiconductor measuring apparatus according to the present embodiment will be described with reference to FIG. The operator who measures the dimension of the pattern inputs the basic information of the sample to be measured in step 1. For example, when the sample is a semiconductor wafer, the type of wafer and the name of the manufacturing process correspond to the basic information described above, and these are used to classify and manage a plurality of existing recipes. Next, the optical conditions used for measurement in step 2 are selected. The parameters of the optical conditions are the probe current incident on the sample and the incident energy. These parameters are obtained by scanning image acquisition, such as "image quality deteriorates by multiple image acquisitions", "brightness unevenness that causes adverse effects during measurement, etc." Is determined so as not to occur. In this operation, the operator may arbitrarily select the optical conditions, or the recommended conditions determined by the manufacturer at the time of shipping the apparatus may be used as they are. In a sample on which a pattern such as a semiconductor wafer is formed, it is necessary to accurately measure the positional relationship between the coordinates of the stage on which the sample is moved and the coordinates of the pattern formed on the sample. In this embodiment, the process of measuring this positional relationship is referred to as an alignment process (process 5). In step 3, an image of the pattern on the sample that can be recognized on the optical image and the scanned image is registered in the console 6 as a template. Two types of optical images and scanned images can be registered in this template. The optical image template is used in the first alignment step, and the SEM image template is used in the second alignment step. Usually, after the first alignment step with low accuracy, the second alignment step with high accuracy is performed. For example, the registration operation is executed by the user storing the optical image and the scanned image displayed on the monitor of the console in the storage medium 7. (Step 3) In order to accurately correct the positional relationship between the coordinates of the stage and the coordinates of the pattern formed on the sample, it is necessary to perform an alignment step at least in two places. In step 4, the location for alignment is registered. The registration is executed, for example, when the user designates an appropriate position on the scanned image displayed on the monitor via the console. In step 5, the positional relationship between the coordinates of the stage and the coordinates of the sample pattern from the comparison between the template and the optical image captured at the location registered above (first alignment step) and the comparison of the scanned image (second alignment step). Measure. Then, the measurement result is stored in the storage medium 7.

  Next, a position search template for searching for a location to be measured is registered in the vicinity of the pattern to be measured in step 6, and a template for the location to be measured in step 7 is registered in the console. The information registered as a template here is a low-magnification scan image and stage coordinates in step 6, and a scan image and stage coordinates of the magnification to be measured in step 7. Here, the work performed at the time of registration is the same as the work of registering the alignment template. In step 8, a signal waveform is extracted based on the scanned image registered in step 7, measurement is performed, and the result is stored in the storage medium 7. If there are a plurality of measurement positions in the wafer, the operator repeats Step 6 to Step 8, and after all the measurement positions are registered, the recipe is stored in the storage medium 7 and the process ends. In automatic measurement using a recipe, the operator inputs the basic information of the sample to the console 6 and starts automatic measurement. The console 6 reads the recipe information stored in the storage medium 7 together with the start command, automatically measures the wafer based on the information, and when the measurement is completed for all the measurement points registered in the recipe, the measurement result is displayed. Save to the storage medium 7.

The above is an example of the recipe creation procedure and automatic measurement in the measuring apparatus, but in order to apply the present embodiment, the above-described signals for the steps 7 and 8 in the recipe creation procedure shown in FIG. It is necessary to make settings to automatically execute processing. FIG. 10 illustrates a recipe creation procedure for automatically performing the signal processing shown in the present embodiment, and only the steps 7 and 8 in FIG. 9 related to the present embodiment are extracted and illustrated. . First, in step 7-1, a scan image including a position to be measured is registered as a template, and the method of measurement in step 7-2 is selected. This measuring method indicates a pattern pitch, a space between lines, or a line width in the case of a repetitive pattern such as a line and space. This method varies depending on the object to be measured. For the line and space pattern shown in FIG. 10, the above measurement method is used. However, in the case of an isolated pattern without contact holes or periodicity, the contact diameter or A method of measuring the distance between patterns is used. The operator who creates the recipe can select the measurement method described above according to the pattern to be measured. When the operator selects a measurement method, a cursor for specifying the measurement location is displayed on the registered scanned image. In step 7-3, the operator adjusts the position and size of the cursor, and moves the cursor to the measurement location. In addition, register the measurement position. (A) in FIG. 10 shows an example of measuring a space between lines of line and space. In this case, two rectangular cursors are displayed with bold lines, and the edges of the lines at both ends of the space where the cursor is measured are displayed. Space can be measured by putting each cursor. Next, the position registered in step 7-3 is measured. In this embodiment, since it is necessary to acquire two scanned images with different scanning directions by about 180 degrees, in step 8-1, the operator selects step 7- Two scanning images having different scanning directions are acquired at the same position as the scanning image registered in 1. In step 8-2, parameters for alignment processing for the two acquired scanned images are set. Here, the parameters of the alignment process indicate an algorithm for alignment and an effective area for determining alignment in the scanned image, and the operator may freely set them. The parameters set by the manufacturer of the measuring device at the time of shipment may be used as they are. FIG. 10B shows a case of an algorithm in which a difference image is formed from the two scanned images described above, and alignment is performed so that the standard deviation of the difference image is minimized. At this time, the difference image is displayed on the monitor of the console 6, and the operator arbitrarily adjusts the position and size of the rectangular area selection cursor indicating the effective area for the alignment, thereby adjusting the position. Parameters can be adjusted. After setting the alignment process, in step 8-3, signal waveforms are extracted from the two scanned images for which the alignment process has been completed based on the measurement positions registered in step 7-3 ((c) in FIG. 10). In step 8-4, the operator sets signal waveform alignment parameters according to the procedure described above with reference to FIG. 7, and in step 8-5, the signal waveform is reconstructed. At this time, the reconstructed signal waveform is displayed on the monitor of the console 6, and the operator sets the conditions for measuring the subsequent step 8-6 distance while viewing the displayed signal waveform, step 8 Measure the -7 distance. In setting the conditions for measuring the distance in step 8-6, the operator selects an algorithm to be measured while viewing the signal waveform. Here, the algorithm indicates an algorithm for measuring a distance from a signal waveform, and the effect of the present embodiment can be exhibited by an algorithm generally used in a measuring apparatus.
(GUI screen)
FIG. 11 is a schematic diagram of a GUI screen displayed on the monitor of the console 6 when the operator operates the recipe creation procedure shown in FIG. The GUI screen includes a general screen 52 for instructing the overall operation of the measuring apparatus from the console 6, an image display screen 53 for displaying a scanned image, a gain adjustment screen 54 for adjusting the level of the detected signal, and the signal waveform shown in this embodiment. And a length measurement condition adjustment screen 55 for selecting a process and a measurement algorithm. This GUI screen is automatically displayed on the monitor on the console 6 after the template for measurement position search in step 6 is registered in the recipe creation procedure shown in FIG. 9, and the operator registers the template for the measurement position as it is. (Step 7) and subsequent steps can be set.

  The general screen 52 at the bottom of the GUI screen has a plurality of function buttons 56 indicating the operation state of the apparatus, and the operator can operate the function displayed on the button by clicking the button on the screen. it can. Here, in the general screen 52, representative functions for explaining the present embodiment, a load button for transporting a wafer to the sample chamber, an image display button for displaying a scanning image, a recipe creation button for creating a recipe, and automatic measurement are performed. Although an automatic length measurement button, a maintenance button used for various maintenance of the SEM type semiconductor measuring device, and an end button for ending the operation of the SEM type semiconductor measuring device are shown, the effect of this embodiment can be achieved even if functions other than these are displayed. Will not be harmed. Since FIG. 11 shows a state in which a recipe is being created while observing a scanned image, the image display and recipe creation function buttons 56 are in an active state (inverted black state).

  The image display screen 53 is a screen for displaying a scanned image or an optical image. As function buttons 56, a length measurement condition adjustment button, an optical adjustment (optical condition) button, a gain adjustment button, an image condition button, and an imaging condition button are provided. It has been. In FIG. 11, since the gain adjustment screen 54 and the length measurement condition adjustment screen 55 are raised, these function buttons 56 are in an active state, and the optical adjustment buttons are not in an active state. The optical adjustment button is a function button used when adjusting the optical axis of the SEM casing 1 and is not active because it is not necessary to adjust the optical axis of the SEM casing 1 at the recipe creation stage. Further, the image condition button is a button for adjusting the energy of the primary electron beam 20 incident on the sample 21 and the probe current. Since the setting has already been completed in the step 2 in FIG. 9, the image condition button is set in FIG. Energy (acceleration voltage) and probe current are displayed. The imaging condition button is a button for setting the scanning speed of the scanned image displayed on the screen and the number of times of integration. The imaging condition button can be activated at any step of recipe creation, but FIG. 11 shows a state where the setting has already been completed. In this case, the current setting is displayed on the screen with the scanning speed of the scanned image as the scanning condition and the number of times of scanning image integration as the number of times of integration. The image display screen 53 is provided with two pull-down buttons 57. One is an image switching button for switching an image to be displayed between a scanned image and an optical image, and the other is an arbitrary magnification of the scanned image. This is a magnification selection button to be selected. These can be operated at any stage as long as the image display screen 53 is activated regardless of the recipe creation. Reference numeral 51 denotes a cursor.

  The gain adjustment screen 54 is a screen for adjusting the level of the detection system of the measuring apparatus. FIG. 11 shows two types of brightness, brightness for changing the offset of the signal waveform and contrast for changing the amplitude of the signal waveform. These can be adjusted by the scroll bar 58, and the set values are displayed in the window 59 on the right side of the scroll bar 58. In addition, the gain adjustment screen 54 has a pull-down button 57 for selecting a mode for adjusting the level. By selecting the automatic or manual operation with this button, the measurement device automatically adjusts the level or the operator manually You can choose to adjust the level.

  The length measurement condition adjustment screen 55 is a screen on which the operator performs the procedure shown in FIG. 10. The length measurement condition adjustment screen 55 performs an area for registering the measurement position template in step 7 and the length measurement in step 8. There is an area to do. In the area where the template is registered, there is a pull-down button 57 for selecting a length measurement method, and the operator can select a length measurement method according to the length measurement target with the pull-down button 57. At this time, a cursor corresponding to the length measurement method selected here is displayed on the scanned image of the image display screen 53, and the operator can adjust the position and size of the cursor on the screen. Further, in the area for registering templates, there are command buttons 60 that are used when a registered or registered template is deleted. The operator can register or delete a template by clicking the command button 60. . In the length measurement condition adjustment screen 55, the operator adjusts the length measurement condition in the procedure of step 8 in FIG. In step 8, the operator can select whether to adjust the procedure from steps 8-1 to 8-5 with the check button 61 in the screen. The check button 61 in the high aspect length measurement mode is a button for determining whether or not to acquire an image having a different scanning direction shown in step 8-1, and by enabling the check button 61, the check buttons after that are checked. 61 operation becomes possible. The image position correction check button 61 is a button for determining whether or not to align the scanned image shown in step 8-2. When this button is enabled, input to the degeneration window 59 becomes possible, and image display is performed. The area selection cursor described above is displayed on the scanned image of the screen 53. Here, degeneration is a parameter that determines how many pixels from the outer periphery of the scanned image are to be ignored during the alignment process, and the operator may freely adjust or use the initial value determined by the device manufacturer as it is. May be. The waveform position correction check button 61 is a button for determining whether or not the signal waveform is aligned as shown in step 8-4. When this button is enabled, a pull-down button 57 for determining the alignment method is enabled. The operator can select the alignment method and make the adjustment shown in FIG. The smoothing check button 61 is a button for determining whether or not to smooth the signal waveform shown in FIG. 8 when the signal waveform shown in Step 8-5 is reconstructed. Then, the input to the level window 59 becomes possible. The operator can determine the degree of smoothing by inputting a number that determines the degree of smoothing in this window 59. Subsequently, in step 8-6 set by the operator, by clicking the length measurement method function button 56, a screen for setting the length measurement method can be newly launched. Although the screen for setting the length measurement method is not shown in FIG. 11, various parameters relating to the length measurement method can be set in accordance with the steps 8-6 and 8-7 in FIG. . The above is an operation in which the operator performs the procedure shown in Step 8-1 to Step 8-7 in FIG. 10 on the GUI. In performing these operations, acquisition and reconstruction of the signal waveform and measurement of the dimensions are performed. And the operation of saving the set condition is necessary. These can be operated by clicking the command button 60 at the bottom of the length measurement condition adjustment screen 55, and the signal waveform displayed in the screen is updated as needed according to the operation. The adjustment procedure when the high aspect length measurement mode is enabled and the operation using the GUI have been described so far. However, when the high aspect length measurement mode is disabled, the operator performs steps 8-1 to 8- 5 is not performed, the process proceeds to Step 8-6, and the length measurement conditions can be set in the same manner as in the conventional measuring apparatus.

  When the pattern dimension of the bottom part of the NAND gate was measured using the SEM type semiconductor measuring apparatus according to this example, a good result was obtained. As a result, it is possible to measure a high aspect structure after etching, which has been difficult in the past, and to cope with various dimension measurements required in the manufacturing process of a semiconductor device.

  As described above, according to the present embodiment, the position of both ends of the bottom portion is obtained using the rising response (detection signal obtained by scanning charged particles from the bottom portion to the upper portion). However, it is possible to provide a charged particle beam apparatus and a measurement method capable of measuring the dimension of the pattern bottom with high accuracy.

A second embodiment will be described. Note that the items described in the first embodiment and not described in the present embodiment can be applied to the present embodiment unless there are special circumstances.
In the first embodiment, one configuration example of the SEM type semiconductor measuring device suitable for the dimension measurement of the bottom of the high aspect structure has been described. In a semiconductor device manufacturing factory, a plurality of SEM type semiconductor measuring devices are often installed in one line, and automatic measurement using the same recipe is performed by a plurality of measuring devices. Thus, when using the SEM type semiconductor measuring device, individual differences (machine differences) between the measuring devices become a problem, and calibration of the machine differences is indispensable. In the second embodiment, calibration of machine differences in an SEM type semiconductor measuring device suitable for measuring the size of the bottom of a high aspect structure will be described. As described in the first embodiment, the measurement of the bottom of the high aspect structure requires a dynamic range and accuracy of four digits (14 bits) or more in the signal electron detection system. The present embodiment is characterized in that the SEM type semiconductor measuring device has a function of converting the detected signal electron intensity into a signal current in order to calibrate the individual difference of the detection system. Here, the signal current is a physical quantity capable of quantifying the amount of the signal electrons 22 emitted from the sample 21. In this embodiment, the brightness of the sample observed with a known probe current is shown as an example of the signal current. Even if the probe current is the same, the signal current varies depending on the sample, but the probe current and brightness characteristic data obtained by changing the probe current for the same sample shows the characteristics of the measured detection system. The intensity of the signal electrons including the individual differences of the system can be converted into the amount of the signal electrons 22.
(Signal waveform calibration processing)
FIG. 12 shows an outline of the calibration method of the detection system in the present embodiment, and the signal processing unit 4, the console 6 and the storage medium 7 related to the calibration processing in FIG. 1 are extracted. In order to convert the detected signal electron intensity into a signal current, a data sheet 62 that is an aggregate of characteristic data 63 obtained by measuring the characteristics of the detection system is required. This data sheet 62 is stored in the storage medium 7. ing. The console 6 reads the characteristic data 63 corresponding to the adjustment value instructed to the level adjustment circuit 39 in FIG. 1 from the data sheet 62, and uses the characteristic data 63 to determine the intensity of the signal waveform sent from the signal processing unit 4 as the signal current. Convert to. Then, the console 6 stores the signal waveform converted into the signal current and the result measured from the waveform in the storage medium 7. Here, the data sheet 62, which is an aggregate of the characteristic data 63, is created by the manufacturer or operator of the measuring device at the time of shipment of the measuring device or during periodic maintenance of the device. The data 63 and the data sheet 62 can be updated. The individual difference in the detection system upstream of the AD conversion unit 41 can be calibrated by the method described above. The procedure for creating the characteristic data 63 and the data sheet 62 will be described later (data sheet creation procedure).
(Probe current measuring device)
In order to accurately acquire the probe current and brightness characteristic data, the SEM type semiconductor measurement device needs to have a function of measuring the probe current of the primary electron beam. FIG. 13 shows the configuration of a device that measures the probe current, and shows a part of the SEM casing 1 in FIG. 1, the probe current measuring device 33, and the console 6. The probe current measuring unit 14 included in the SEM casing 1 includes a blanking device 64 that deflects the primary electron beam 20, a Faraday cup 65 that detects the probe current, and the Faraday cup 65 and the SEM casing 1 are electrically insulated. It is comprised with the insulator 25 which carries out. The probe current measuring device 33 also includes a constant voltage circuit 66 that deflects the primary electron beam 20, an amplifier 38 that converts and amplifies the probe current detected by the Faraday cup 65, and a digital signal of the amplified voltage signal. It comprises an AD converter 41 that converts the signal. Here, the blanking device 64 and the constant voltage circuit 66, as well as the Faraday cup 65 and the amplifier 38 are electrically connected by a feedthrough terminal that maintains a vacuum of the SEM casing 1 that is not shown in the drawing. Therefore, the constant voltage circuit 66 is configured to apply a constant voltage to the blanking device 64 based on a control signal from the console 6 so that the probe current can be detected by the Faraday cup 65. The Faraday cup 65 is provided with a hole through which the primary electron beam 20 passes, and when the primary electron beam 20 is deflected by the blanking device 64, the primary electron beam 20 is captured by the Faraday cup 65 for the first time. The probe current can be detected.
(Datasheet creation procedure)
To create the data sheet 62 of the detection system, the circuit constant of the level adjustment circuit 39 of the detection system is fixed, the brightness of the sample is measured with a known probe current, and the relationship between the probe current (signal current) and the image brightness is determined. Need to ask. FIG. 14 shows a procedure for creating the data sheet 62, which is roughly divided into two steps. One is to measure the brightness of the image while changing the circuit constant of the level adjustment circuit 39 stepwise without irradiating the specimen 21 with the primary electron beam 20, and the other is to measure the probe current. This is a measurement of the detection system characteristic for obtaining the relationship between the signal current and the image brightness while changing the circuit constant of the level adjustment circuit 39 step by step. Hereafter, each process is demonstrated along a procedure. First, the dark level is measured by setting the probe current to zero so that the sample 21 is not irradiated with the primary electron beam 20 in the first step (beam off). In this setting, the electron gun power supply 31 may be controlled from the console 6 so that electrons are not emitted from the electron source 11, or when there is a valve in the SEM casing 1, the valve may only be closed. In the next second step, circuit constants of the detection system are set, and in the third step, the brightness of the scanned image is measured. The dark level can be measured by repeating the second to fourth steps while changing the circuit constant of the detection system stepwise.

Subsequently, in the measurement of the detection system characteristic, the sample 21 is irradiated with the primary electron beam 20 in the fifth step (beam-on). In this setting, the electron gun power supply 31 may be controlled from the console 6 so that electrons are emitted from the electron source 11, or when there is a valve in the SEM casing 1, only the closed valve is opened. But it ’s okay. In the next sixth step, the probe current of one electron beam is set, but as a means for changing the probe current, the electron gun power supply 31 is controlled from the console 6 and the operating conditions of the electron source 11 (the extraction voltage for emitting electrons). Etc.), or the trajectory of the primary electron beam 20 may be adjusted by the first condenser lens 12 disposed on the diaphragm 13. Any means for changing the probe current may be used, but in order to exert the effect of the present embodiment, the probe current range needs to be four digits or more. This is because the detection system requires a 4-digit dynamic range in the measurement of the bottom of the high aspect structure realized by this embodiment. After setting the probe current in the sixth step, the circuit constant of the detection system is set in the next seventh step, and the brightness of the scanned image is measured in the eighth step. By repeating the sixth to eighth steps, the characteristics of the detection system can be measured. It should be noted here that in the case of detection system characteristics, unlike the case of the dark level, the sample is irradiated with an electron beam. The brightness of the image changes. Therefore, if it is determined that there is the next measurement in the ninth step, it is necessary to devise so that the brightness of the scanned image can be measured at a new location by moving the sample. The data sheet 62 can be created by measuring the dark level and the detection system characteristics for each circuit constant of the detection system according to the procedure described above.
(GUI screen)
Next, a GUI screen displayed on the monitor of the console 6 when the data sheet 62 is created by the procedure shown in FIG. 14 will be described. FIG. 15 is a schematic diagram of a GUI screen displayed on the monitor of the console 6 when measuring the characteristics of the detection system. The GUI screen is maintained with a general screen 52 for instructing the operation of the measuring apparatus from the console 6 and maintenance. A maintenance screen 67 for selecting items and a detection characteristic measurement screen 68 for automatically measuring the characteristics of the detection system are configured.

  The general screen 52 at the bottom of the GUI screen has a plurality of function buttons 56 indicating the operation state of the apparatus, and the operator can operate the function displayed on the button by clicking the button on the screen. it can. Here, in the general screen 52, representative functions for explaining the present embodiment, a load button for transporting a wafer to the sample chamber, an image display button for displaying a scanning image, a recipe creation button for creating a recipe, and automatic measurement are performed. Although an automatic length measurement button, a maintenance button used for various maintenance of the SEM type semiconductor measuring device, and an end button for ending the operation of the SEM type semiconductor measuring device are shown, the effect of this embodiment can be achieved even if functions other than these are displayed. Will not be harmed. Since FIG. 15 shows a state in which the measuring apparatus is being maintained, the maintenance function button 56 is in an active state (a state in which it is blackened).

  The maintenance screen 67 has a function button 56 for each maintenance item, and the latest date and time when the maintenance is performed is displayed beside the function button 56. By displaying the latest date and time when calibration was performed for each maintenance item, the operator can check whether the SEM type semiconductor measuring device is regularly maintained. In FIG. 15, since the detection system characteristic is being calibrated, the detection system characteristic function button 56 is activated (inverted black).

  The detection characteristic measurement screen 68 is divided into an area for displaying an image and an area for setting measurement conditions. The area for displaying an image has the same configuration as that shown in FIG. It is possible to search for a location to be measured on the sample while arbitrarily switching the scanning image and confirming the image on the monitor. The area for setting the measurement conditions includes a check button 61 for each item of the dark level and the detection system characteristic, and the operator can select an item to be measured by checking or not. By providing a function for selecting items to be measured, unnecessary measurement can be reduced and time required for maintenance can be reduced. In addition, the probe current, which is a parameter during measurement, and the circuit constants (brightness and contrast) of the detection system can be entered numerically with the measurement range and measurement interval. By inputting a desired number, the measurement range and interval can be arbitrarily adjusted and set. A series of operations required for measurement, such as erasure of measurement results, start of measurement, and registration of measurement results, can be performed by the measurement device by the operator instructing command buttons 60 for each operation. it can.

  As in this embodiment, the SEM type semiconductor measurement device has a function of calibrating the characteristics of the detection system, so that individual differences (measurement differences) between the measurement devices can be reduced and the bottom of the high aspect structure can be reduced. When measuring, the required 4-digit dynamic range can be realized with high accuracy.

When the pattern dimension of the bottom part of the NAND gate was measured using the SEM type semiconductor measuring apparatus according to this example, a good result was obtained.
According to the present embodiment, the same effect as in the first embodiment can be obtained. Moreover, the individual difference of the detection system of a some charged particle beam apparatus can be calibrated by providing the function which converts the intensity | strength of the detected signal charged particle into a signal current.

A third embodiment will be described with reference to FIGS. 16 and 17. Note that the matters described in the first or second embodiment and not described in the present embodiment can be applied to the present embodiment unless there are special circumstances.
In the first embodiment, a configuration example of an SEM type semiconductor measuring apparatus suitable for measuring the bottom of a high aspect structure will be described, and a signal waveform used for measurement and a scanning image displayed on the monitor of the console 6 are converted into independent data. It was shown that high-precision measurement can be realized with the minimum necessary data size. In the third embodiment, one configuration example of an SEM type semiconductor measuring device that can positively utilize this advantage will be described. In this embodiment, the schematic configuration of the SEM type semiconductor measuring apparatus is the same as that in FIG. 1, but the configuration of the signal processing unit is greatly different. Hereinafter, the signal processing unit 4 which is a feature of the present embodiment will be described in detail.
(Signal processing part)
FIG. 16 is a schematic configuration diagram of the signal processing unit 4. The signal processing unit 4 includes a switching circuit A 42, a memory A 43, a memory B 44, a memory C 69, an image processing circuit 46, a comparison operation circuit 47, and a signal processing circuit. 48. The signal processing unit 4 is connected to the detector 23 via the level adjustment circuit 39 and the AD conversion unit 41 and to the console 6 by electrical wiring. Based on the control signal from the console 6, the signal from the detector 23 is connected. Are processed and output to the console 6. In the control from the console 6, the level adjustment circuit 39 adjusts the amplification and offset of the signal sent from the detector 23, and the AD converter 41 converts it into a digital signal in synchronization with the scanning signal. Then, the digital signal is stored in the memory A 43 or the memory B 44 and the scanning image is stored in the memory C 69 via the switching circuit A 42. The operations of the comparison operation circuit 47 and the signal processing circuit 48 that process the signal waveform that follows and the image processing circuit 46 that processes the scanned image are the same as those in FIG.

  Here, the difference between the schematic configuration diagrams of the signal processing unit 4 shown in FIG. 16 and FIG. 5 is that the memory for temporarily storing the signal waveform and the scanning image is a memory for the signal waveform (memory A 43, memory B 44). ) And a scanned image memory (memory C 69).

So far, this embodiment has shown that the accuracy of 0.01% is required for the signal waveform in order to measure the bottom of the high aspect structure. In order to realize this accuracy, not only a device for increasing the accuracy of the detection system but also the accuracy of the detected signal electrons themselves is required. When the accuracy of the signal electrons is dominated by shot noise, in order to improve the conventional accuracy of several percent to 0.01%, it is necessary to integrate the signal waveform by about 10,000 times. In the configuration of the first embodiment, a desired region is selected from the scanned image stored in the memory, and a signal waveform is formed from the brightness of the region. Therefore, it is necessary to increase the number of integrations of the scanned image itself by about 10,000 times. It takes a lot of time to acquire a scanned image. However, in this embodiment, since the signal waveform and the memory of the scanned image are independent, it is not necessary to increase the number of integration of the scanned image itself more than necessary, and the measurement can be performed in the minimum necessary time. For example, if the accuracy of the scanned image is 8 bits, the accumulated number of scanned images may be as many as 256 times at most, and the signal waveform selectively scans only the region to be measured in the scanned image. Then, it may be integrated approximately 10,000 times more than the scanned image.
(Recipe creation procedure)
FIG. 17 illustrates a recipe creation procedure for automatically performing the signal processing shown in the present embodiment, and shows only the steps 7 and 8 in FIG. 9 shown in the first embodiment. In step 7, the position to be measured is registered as a template. Unlike the first embodiment, the scanned image is stored in a dedicated memory C 69 in the signal processing unit 4. Other than that, a recipe is created in the same procedure as described in FIG. 10 from step 7-1 to step 7-3. Next, in step 8, the pattern registered as a template is measured, but only the signal waveform acquisition condition setting in step 8-1 is different from that in FIG. 10, and thereafter, the recipe is created in the same procedure as described in FIG. . The operator determines how many scanning lines are arranged in the direction (scanning feed direction) perpendicular to the scanning direction of the cursor range corresponding to the position measured in Step 8-1, and then determines the number of times scanning lines are integrated. . FIG. 17B is a diagram for explaining signal waveform acquisition conditions. In the figure, N scanning lines are arranged in the range of the cursor. Here, when the number of scan line integrations is M, signal waveform acquisition conditions are set so that the product of the number N of scan lines and the number of integrations M is approximately 10,000 times the number of scan image integrations. Thereafter, the operator acquires the signal waveforms by storing two signal waveforms having different scanning directions in the memory A 43 and the memory B 44 in step 8-2 in accordance with the signal waveform acquisition conditions determined in step 8-1. be able to.

When the pattern dimension of the bottom part of the NAND gate was measured using the SEM type semiconductor measuring apparatus according to this example, a good result was obtained.
According to the present embodiment, the same effect as the above embodiment can be obtained. Further, by dividing the signal waveform memory and the scan image memory, it is not necessary to increase the number of times of integration of the scan image itself more than necessary, and measurement can be performed in the minimum necessary time.

A fourth embodiment will be described. Note that items described in any of Examples 1 to 3 and not described in this example can be applied to this example as long as there are no special circumstances.
In the first to third embodiments, the example of the SEM type semiconductor measuring device suitable for the dimension measurement of the bottom of the high aspect structure has been shown. Up to now, the single detector 23 provided in the SEM casing 1 is used. Signal electrons were detected to form a scanning image and a signal waveform. However, in order to ensure a 4-digit dynamic range with one detection system, the level adjustment circuit 39 that adjusts the level of the detected signal requires a dynamic range of 5 digits or more, which may be difficult to realize. In the fourth embodiment, a configuration example of an SEM type semiconductor measuring device that improves this problem will be described. In the present embodiment, a plurality of parts from the detector to the memory for storing data are provided, and the detection system is selectively used according to the amount of the signal electrons 22. By using this configuration, the dynamic range required for one detection system can be narrowed, and the design of the level adjustment circuit 39 is facilitated. Hereinafter, the present embodiment will be described in detail with reference to the drawings.

FIG. 18 shows the configuration of the SEM type semiconductor measuring apparatus according to the present embodiment. As in FIG. 1, the measurement apparatus is roughly divided into a SEM casing 1, a sample chamber 2, a casing control section 3, a signal processing section 4, a stage control section 5, a wafer transfer section 8, and a vacuum exhaust section 9. All of these can be controlled by the console 6. The console 6 has a large-capacity storage medium 7 capable of storing recipes, measurement results, and acquired scanned images. Based on the data recorded in the storage medium 7, the console 6 operates and manages the data. Unlike the previous embodiments, this embodiment shows a case where there are two detection systems as a representative example in that a plurality of detection systems are provided. Hereinafter, among the parts shown in this embodiment, the configuration of the detection system different from that shown in FIG.
(SEM housing)
Unlike the first embodiment, the SEM housing 1 captures the signal electrons 22 generated in the sample 21 with two detectors (detector A 70 and detector B 71). In FIG. A configuration in which the signal electrons 22 are simultaneously captured by the detector B 71 is shown. When a plurality of detectors are used, the signal electrons 22 may be simultaneously captured by a plurality of detectors as shown in FIG. 18, but the trajectory of the signal electrons 22 is changed, and the signal electrons 22 are captured individually by each detector. Also good.

FIG. 19 shows a schematic configuration of a measuring apparatus in the case where signal electrons 22 are individually captured in a configuration that realizes the present embodiment in a form different from that in FIG. FIG. 19 shows a part of the SEM casing 1 related to the capture of the signal electrons 22, two detection systems, and the casing control power supply 32 of the casing control unit 3. In addition to the conversion plate 16, the detector A 70, and the detector B 71, the SEM housing 1 is provided with an ExB deflector 72 that deflects the trajectory of the signal electrons 22 toward the detectors. The ExB deflector 72 does not change the trajectory of the primary electron beam 20 traveling in the direction of the sample 21 from the electron source 11 and electromagnetic deflector 75 and electromagnetic deflection so as to deflect the trajectory of the signal electron 22 traveling in the opposite direction. The deflection field of each device 76 is adjusted. The two detection systems are configured by level adjustment circuits 39, 39 ′ AD conversion units 41, 41 ′ and a signal processing unit 4 as in the first embodiment. The casing control power supply 32 of the casing control unit 3 is a constant current circuit 73 and a constant voltage circuit 66 for operating the ExB deflector 72 required in the description of FIG. 19, and a relay circuit 74 that switches an output from the constant voltage circuit 66. It is configured. FIG. 19 shows an example in which the detector A 70 and the detector B 71 are arranged to face each other. The ExB deflector 72 deflects the signal electrons 22 so that the signal electrons 22 are captured by either one of the detectors. Reverse the direction. This can be realized by inverting the output of the constant voltage circuit 66 that outputs both positive and negative polarities with the relay circuit 74 and inverting the polarity of the output of the constant current circuit 73. As described above, even when the signal electrons 22 are individually captured by a plurality of detectors, the signal electrons 22 can be efficiently guided to the detector by using the ExB deflector 72.
(Signal processing part)
FIG. 20 is a schematic configuration diagram of the signal processing unit 4 of FIG. 18 and FIG. 19 in the present embodiment. Since there are two detectors, there are two systems from the level adjustment circuit to the memory after that, and each of them. A synthesizing unit 77 for synthesizing the detector signals is newly provided. The signal processing unit 4 is electrically connected to the two detectors 70 and 71 and the console 6, processes the signal from the detector based on the control signal from the console 6, and outputs it to the console 6. In the control from the console 6, the level adjustment circuits 39 and 39 ′ adjust the amplification and offset of the signal sent from the detector, and the AD conversion units 41 and 41 ′ synchronize with the scanning signal and convert it into a digital signal. The digital signal is stored as a scanned image in each of the memories 43, 44, 69 and 80 via the switching circuits 42 and 45. The signal processing from the detector to the memory is the same as in FIG. 5, but the amplification and offset adjustment of the level adjustment circuits 39 and 39 ′ are changed in each system according to the amount of the signal electrons 22. For this reason, the brightness of the scan image stored in the memory is different for each system, and the newly provided combining unit 77 combines the scan image stored in each memory and the signal waveform extracted from the scan image. Play a role.

  FIG. 21 schematically shows a procedure for extracting signal waveforms from the scan images stored in the memories 43, 44, 69, and 80 and synthesizing the extracted signal waveforms in the signal processing in the combining unit 77. Since the scanning directions differ by about 180 degrees, the memory of each system stores a scanning image in which the left and right are reversed. First, the synthesizing unit 77 inverts one of the left and right images, and performs alignment processing between the scanned images. Here, the alignment process may be the same as the alignment described with reference to FIG. 6, and a figure corresponding to this process is omitted in FIG. 21. Next, based on the alignment information, a cursor for extracting a waveform is aligned with each scanned image, and a signal waveform is extracted. The setting of the cursor has already been described in (Recipe creation) in the first embodiment, and FIG. 21 shows the extracted signal waveform. The signal waveform extracted here differs in intensity and shape of the signal waveform because the brightness of the scanned image differs in each system.

  The center of FIG. 21 shows an example in which one system is adjusted to give accuracy to the signal at the top of the high aspect structure, and the other system is adjusted to give accuracy to the signal at the bottom of the high aspect structure. In this case, the gray level of the region corresponding to the bottom of the high aspect structure is 0 in the former signal waveform, and the gray level of the region corresponding to the upper portion of the high aspect structure is maximum in the latter signal waveform. In these regions, since the amplitude of the signal waveform exceeds the dynamic range of the detection system, the gradation is fixed to 0 or the maximum, and the signal intensity is meaningless. Therefore, it is necessary to synthesize the signal waveform by discarding the area where the gradation is 0 or the maximum from these two signal waveforms.

  The right side of FIG. 21 shows the synthesized signal waveforms, and the signal waveforms having the same scanning direction are selected and synthesized from the respective systems. Here, in the process of combining, signal intensity level correction and alignment are necessary so that signal waveforms are continuously connected. The signal intensity level is corrected by first determining the two gradations (S1 and S2) and offsetting them to the intensity of each signal waveform so that the top gradation of the high aspect structure is S2 and the bottom gradation is S1. give. At this time, since it is meaningless to match the area of 0 or the maximum gradation in each signal waveform to S1 and S2, the signal waveform obtained by giving accuracy to the upper signal of the high aspect structure is changed to S2. The offset is adjusted so that the signal waveform obtained by adding accuracy to the signal at the bottom of the structure matches S1. Next, the positions are aligned so that the two signal waveforms are continuously connected. This processing may be performed by the method shown in FIG. 7 or may be performed at the rise of the signal waveform shown in FIG. Finally, the gradation of each signal waveform is expanded and contracted so that the rising portions of the two signal waveforms are continuously connected to be combined into one signal waveform. This synthesis is a process of extracting an appropriate signal for each region from two signal waveforms. Signals are alternately extracted at the intersection of two signal waveforms as a boundary, and the amplitude of the waveform of the two extracted signal waveforms is The larger one is the correct result. With the above procedure, two signal waveforms with different intensities and shapes acquired by the two systems can be combined into one signal waveform. When combining the scanned images, first, the signal waveforms are combined by the above procedure, and the level correction value and the position correction value used at that time are applied to each of the scanned images. It can be combined with the scanned image.

In the present embodiment, the switching circuit subsequent to the AD conversion unit 41, the image processing circuit 46, and the signal processing circuit 48 are described as parts independent of the console 6. However, these functions are also included in the computer included in the console 6. It can be carried out. When the above processing is performed by a computer included in the console 6, since the parts subsequent to the AD conversion unit 41 can be concentrated on the console, it is possible to expect a reduction in the installation area and manufacturing cost of the measuring device, and the effects of this embodiment are as follows. Not damaged.
(GUI screen)
FIG. 22 is a schematic diagram of a GUI screen displayed on the monitor of the console 6 when the operator synthesizes a signal waveform in recipe creation. The GUI screen includes a general screen 52 for instructing the overall operation of the measuring apparatus from the console 6, an image display screen 53 for displaying a scanned image, a gain adjustment screen 54 for adjusting the level of the detected signal, and the synthesis processing shown in this embodiment. The composition processing adjustment screen 78 is used to adjust the parameters. This GUI screen is automatically displayed on the monitor on the console 6 after registering the measurement position template in step 7 in the recipe creation procedure shown in FIG.

  In the present embodiment, since there are two detection systems, the gain adjustment screen 54 is provided with brightness and contrast capable of adjusting the level for each detection system. In addition, the gain adjustment screen 54 is newly provided with a pull-down button 57 for selecting a scanning image to be displayed on the image display screen 53, and the operator can select a detector by operating this button.

  The synthesis processing adjustment screen 78 is a screen on which an operator adjusts parameters according to the procedure shown in FIG. 21, and is divided into a region where signal waveforms are synthesized and a region where scanned images are synthesized. In the area for synthesizing the signal waveform, the signal waveform obtained by each detection system and the synthesized signal waveform are displayed. The number of bits of the signal waveform after synthesis by the pull-down button 57 and the method of alignment are set by the operator. Can be selected. Further, the two gradations (S1 and S2) to be determined when combining can be determined by manually adjusting the level cursor displayed on the signal waveform. As in the case of the signal waveform, the area where the scan image is combined is configured such that the operator can select the number of bits of the scan image after combining with the pull-down button 57 and the alignment method. The extraction of the signal waveform, the synthesis of the signal waveform, and the saving of the adjustment result necessary for proceeding with the procedure of FIG. 21 are performed by the operator clicking the command button 60 arranged at the bottom of the synthesis processing adjustment screen 78. Can be implemented. If the operator manually performs all the procedures described so far, trial and error are required to adjust the parameters to be synthesized. Therefore, when the operator determines only the number of bits, the alignment method, and the two gradations (S1 and S2) determined when combining, the console 6 automatically clicks the combining command button 60 and each console 6 automatically It is preferable that the level of the detection system is adjusted, the signal waveform is continuous, and the scanned image has a function that can be optimized for natural contrast.

  The above is an example of the SEM type semiconductor measuring device, which uses a plurality of detectors depending on the amount of signal electrons emitted from the sample. By using this embodiment, a dynamic range required for one detection system is used. The level adjustment circuit 39 can be easily designed. In addition, since the signal waveform reconstruction shown in the first embodiment can be used as it is as the means for measuring using the synthesized signal waveform, the bottom of the high aspect structure can be measured even with the configuration of the present embodiment. In addition, since a several detection system is used in a present Example, the influence of the individual difference of a detection system cannot be denied. However, by combining the present embodiment with the method for calibrating the characteristics of the detection system shown in Embodiment 2, it is possible to synthesize the signal waveform and the scanned image with high accuracy.

When the pattern dimension of the bottom part of the NAND gate was measured using the SEM type semiconductor measuring apparatus according to this example, a good result was obtained.
According to the present embodiment, the same effect as the above embodiment can be obtained. Further, by providing a plurality of detectors and using a plurality of detectors in accordance with the amount of signal electrons emitted from the sample, the dynamic range required for one detection system can be narrowed.

A fifth embodiment will be described. Note that items described in any of Examples 1 to 4 and not described in this example can be applied to this example as long as there are no special circumstances.
In the fourth embodiment, an example in which a plurality of detectors are selectively used in accordance with the amount of signal electrons emitted from the sample in the SEM type semiconductor measuring device has been shown. However, the dynamic range of the level adjustment circuit 39 can be set in multiple stages. In this case, the same function as that of the fourth embodiment can be realized with one detector. In the present embodiment, the configuration of an SEM type semiconductor measuring device including a level adjustment circuit 39 capable of setting a dynamic range in multiple stages will be described. In this case, since the number of detection systems is reduced as compared with the fourth embodiment, the manufacturing cost of the measuring device is reduced, and the functions described so far can be realized at low cost. Hereinafter, the present embodiment will be described in detail with reference to the drawings.
(Signal processing part)
FIG. 23 is a schematic configuration diagram of the signal processing unit 4 in the present embodiment, and includes a level adjustment circuit 39 capable of setting a dynamic range in multiple stages. Therefore, a plurality of memories 43, 44, 69,. ..90 is provided. The signal processing unit 4 is connected to the detector 23 via the level adjustment circuit 39 and the AD conversion unit 41 and to the console 6 by electrical wiring, and based on a control signal from the console 6, a signal from the detector 23 is connected. Are processed and output to the console 6. In the control from the console 6, the level adjustment circuit 39 adjusts the amplification and offset of the signal sent from the detector 23, and the AD converter 41 converts it into a digital signal in synchronization with the scanning signal. The digital signal is stored as a scanning image in each memory via the switching circuit A42. The signal processing from the detector to the memory is the same as in FIG. 5, but the amplification factor and offset of the level adjustment circuit 39 are set in multiple stages based on the control signal from the console 6, and different memories 43, 44 for each setting. , 69... 90 is different from FIG. In each memory, a pair of scanned images that are reversed left and right are stored for each setting of the level adjustment circuit 39, and using these, the synthesizing unit 77 synthesizes one signal waveform and the scanned image. Here, the synthesizing unit 77 extracts the signal waveform from the scanned image of each memory and synthesizes the signal waveform so as to be continuous. However, in this embodiment, processing that is as difficult as in the fourth embodiment is not necessary. This is because the present embodiment realizes a multi-stage dynamic range only by setting the level adjustment circuit 39, so that a continuous signal waveform can be synthesized only by applying the method shown in the second embodiment. Here, the synthesis of signal waveforms has been taken up, but a scanned image can also be synthesized by a substantially similar procedure. Since the combined scan image is used for display purposes and not used for measurement, the combining unit 77 expands / contracts the gradation of the scan image stored in each memory so that the combined scan image has a natural contrast. Just do it.

  The above is an example of an SEM type semiconductor measuring device provided with a level adjustment circuit capable of setting a dynamic range in multiple stages, and the function described in the fourth embodiment can be realized at low cost. In addition, since the signal waveform reconstruction shown in the first embodiment can be used as it is as the means for measuring using the synthesized signal waveform, the bottom of the high aspect structure can be measured even with the configuration of the present embodiment.

When the pattern dimension of the bottom part of the NAND gate was measured using the SEM type semiconductor measuring apparatus according to this example, a good result was obtained.
According to the present embodiment, the same effect as the above embodiment can be obtained. In addition, since the number of detection systems is reduced by providing a level adjustment circuit that can set the dynamic range in multiple stages, the manufacturing cost of the measuring device can be reduced.

A sixth embodiment will be described. Note that items described in any of Examples 1 to 5 and not described in this example can be applied to this example as long as there is no particular circumstance.
Until now, signal processing focused on two scanned images differing in scanning direction by about 180 degrees has been taken up by taking a line and space pattern as an example of a high aspect structure. The effect of the embodiment can be explained by this signal processing because the edge of the pattern is only in one direction in the line and space. For the contact hole whose edge is in all directions, It is insufficient. In the present embodiment, an SEM type semiconductor measurement device that improves the measurement accuracy of the bottom of a high aspect structure for a contact hole having edges in all directions will be described. The configuration of the SEM type semiconductor measurement apparatus is the same as that of the previous embodiments, but the signal waveform and the process for reconstructing the scanned image are different from those of the past. Hereinafter, the signal waveform and the process for reconstructing the scanned image, which are the features of this embodiment, will be described.

  FIG. 24 shows the signal waveform and processing for reconstructing the scanned image in this embodiment, and shows scanning of the primary electron beam, capturing of signal electrons, extraction of edges, and reconstruction of the scanned image from the left side of the figure. ing. Extraction and reconstruction of signal waveform edges in FIG. 24 will be described in detail with reference to FIGS. 25 and 26. FIG. First, since there are edges in all directions in the contact hole, it is necessary to scan the primary electron beam in all directions (360 degrees). The left side of FIG. 24 shows an example in which the scanning direction is changed to azimuth angles 0, 45, and 90 degrees. The scanning region is 1.4 in the field of view so that the entire field of view is scanned in all scanning directions. It is necessary to double or more. Since the signal electrons emitted from the sample are captured in a coordinate system depending on the scanning direction, each point in the field of view is sampled with different coordinates in each scanned image. In the example of FIG. 24, when the scanning direction is 0 degree, the lower left corner of the field of view is sampled at (3, 4) pixels, 45 degrees is (1, 10), and 90 degrees is sampled at (3, 16). It will be. In this embodiment, scanning images are acquired in all directions, but the edge of the contact hole can be captured by the rising response of the detection system in the direction normal to the tangent of the edge and passes through the contact. Only scan lines. Therefore, in FIG. 24 where there is one contact hole in the center of the field of view, there are about 1 to several scanning lines in each scanning image in which the edge is captured by the rising response of the detection system. In order to extract this edge from the scanned image, bright coordinates are extracted from the image differentiated in the scanning direction, and only the gradation corresponding to the coordinates is extracted from the scanned image as valid data. When valid data is extracted in all scanning directions, a scanning image in which an edge is captured by the rising response of the detection system can be reconstructed by averaging all of them.

  Next, referring to FIG. 25, a method for extracting a scanning line from a scanning image and forming a signal waveform will be described. The left side of FIG. 25 shows a signal waveform obtained when the scanning image is divided in the scanning line feeding direction. In the case of a contact hole, the edge interval widens as it advances in the scanning line feeding direction, and passes through the maximum value. Reduce from. Here, the scanning line that is normal to the edge tangent line and passes through the contact point is a scanning line in which the edge interval is maximized, and the scanning line in which the edge interval is maximized for each scanned image is By extracting, it is possible to extract a signal waveform in which the edge is captured by the rising response of the detection system. Note that the scanning lines may be extracted from the maximum value of the distance between the edges as described above. However, when the shape of the contact hole is flat, the distance between the edges is measured in order for each scanning line. A scanning line that minimizes the change in the distance may be extracted. At this stage, since each signal waveform includes a signal of a falling response of the detection system, it is necessary to extract a rising response region of the detection system from the signal waveform as before. Although the same method as in the first embodiment may be used, in FIG. 26, the differential waveform of the signal is utilized to extract the rising response region of the detection system. The center of FIG. 26 shows the signal waveform and its differential waveform. In the differential waveform, a positive peak is seen in the rising response region of the detection system. By extracting a signal in a region where the differential waveform has a positive peak from the signal waveform, it is possible to extract a rising response region of the detection system. The right side of FIG. 26 shows the reconstruction of the signal waveform. The procedure is to select a pair of signal waveforms whose directions are different by about 180 degrees from the signal waveforms in all directions, and position them so that the centers between the edges overlap. Add together a pair of signal waveforms. Next, this processing is performed on the signal waveforms in all directions, and finally all signal waveforms are added and averaged with the center between the edges as a reference. Thereby, the signal waveform of the contact hole using only the rising response of the detection system can be reconstructed.

  The above is the signal processing of the SEM type semiconductor measurement device that improves the measurement accuracy of the bottom of the high aspect structure with respect to the contact hole. In the present embodiment, the description is focused on the contact hole, but the bottom of the high aspect structure can be measured by a process similar to the above even for a complicated pattern having edges in a plurality of directions.

When the pattern dimension at the bottom of the contact hole was measured using the SEM type semiconductor measuring apparatus according to this example, a good result was obtained.
According to the present embodiment, the same effect as the above embodiment can be obtained. Further, by scanning the charged particle beam in a plurality of directions with respect to the sample, it is possible to measure a complicated pattern having edges in a plurality of directions with high accuracy.

  In addition, this invention is not limited to an above-described Example, Various modifications are included. For example, the above-described embodiments have been described in detail for easy understanding of the present invention, and are not necessarily limited to those having all the configurations described. Further, a part of the configuration of a certain embodiment can be replaced with the configuration of another embodiment, and the configuration of another embodiment can be added to the configuration of a certain embodiment. Further, it is possible to add, delete, and replace other configurations for a part of the configuration of each embodiment.

DESCRIPTION OF SYMBOLS 1 ... SEM housing | casing, 2 ... Sample chamber, 3 ... Housing control part, 4 ... Signal processing part, 5 ... Stage control part, 6 ... Console, 7 ... Storage medium, 8 ... Wafer conveyance part, 9 ... Vacuum exhaust part DESCRIPTION OF SYMBOLS 10 ... Sample preparation room, 11 ... Electron source, 12 ... 1st condenser lens, 13 ... Aperture, 14 ... Probe electric current measurement part, 15 ... 2nd condenser lens, 16 ... Conversion board, 17 ... Deflector, 18 ... Objective Lens, 19 ... Height sensor, 20 ... Primary electron beam, 21 ... Sample, 22 ... Signal electron, 23 ... Detector, 24 ... Stage, 25 ... Insulator, 26 ... Sample folder, 27 ... Mirror, 28 ... Retarding Power source, 29 ... laser measuring device, 30 ... stage driving device, 31 ... electron gun power source, 32 ... housing control power source, 33 ... probe current measuring device, 34 ... beam control circuit, 35 ... conveyance control unit, 36 ... conveying robot 3 ... Valve, 38 ... Amplifier, 39,39 '... Level adjustment circuit, 41,41' ... AD converter, 42 ... Switching circuit A, 43 ... Memory A, 44 ... Memory B, 45 ... Switching circuit B, 46 ... Image Processing circuit 47... Comparison operation circuit 48. Signal processing circuit 49. Discontinuous point 51. Cursor 52. General screen 53 53 image display screen 54 gain adjustment screen 55 length measurement condition adjustment screen 56 ... Function button, 57 ... Pull-down button, 58 ... Scroll bar, 59 ... Window, 60 ... Command button, 61 ... Check button, 62 ... Data sheet, 63 ... Characteristic data, 64 ... Blanking device, 65 ... Faraday cup, 66 ... Constant voltage circuit, 67 ... Maintenance screen, 68 ... Detection characteristic measurement screen, 69 ... Memory C, 70 ... Detector A, 71 ... Detector B, 72 ... E B deflector 73 ... constant current circuit, 74 ... relay circuit, 75 ... electrostatic deflector, 76 ... electromagnetic deflector, 77 ... combining unit, 78 ... synthetic process adjustment screen, 80, 90 ... memory.

Claims (14)

A deflector that scans a charged particle beam that has converged on a predetermined region of the sample on which the concavo-convex pattern is formed, and charged particles generated from the sample are detected by scanning the charged particle beam on the predetermined region of the sample. In a charged particle beam apparatus comprising a detector and a signal processing unit that processes a scanning image based on a charged particle signal detected by the detector,
The signal processing unit
A first scanning image based on a charged particle signal from a predetermined region of the sample when the charged particle beam is scanned in a first direction, and the second scanning direction when scanning in a second direction opposite to the first direction; A memory for recording a second scanned image based on a charged particle signal from the predetermined region of the sample;
A comparison operation circuit for comparing the first scanned image and the second scanned image recorded in the memory;
A charged particle beam comprising: a signal processing circuit that reconstructs a scanned image by setting a lower value of the signal intensity of the scanned image to a correct value in the first scanned image and the second scanned image. apparatus. The charged particle beam apparatus according to claim 1,
The charged particle beam apparatus, wherein the scanned image reconstructed in the signal processing circuit is used for measuring a size of a bottom of a concavo-convex pattern formed in a predetermined region of the sample. The charged particle beam apparatus according to claim 1,
The charged particle beam apparatus to be scanned is an electron beam. The charged particle beam apparatus according to claim 1,
The charged particle beam apparatus, wherein the first scanning direction and the second scanning direction are different by about 180 degrees. The charged particle beam apparatus according to claim 1,
The comparison operation circuit compares the first scanned image and the second scanned image, aligns the positions of the scanned images, and compares the signal waveforms extracted from the first scanned image and the second scanned image. Then, the charged particle beam apparatus is characterized in that the signal waveforms are aligned with each other. The charged particle beam apparatus according to claim 1,
A charged particle beam apparatus further comprising a function of calibrating a signal intensity of a scanning image based on a charged particle signal detected by the detector from characteristics of a detection system including the detector. The charged particle beam apparatus according to claim 6.
A charged particle beam device characterized in that the characteristic of the detection system is a relationship between a multiplication factor of the detection system and a signal intensity. The charged particle beam apparatus according to claim 1,
A charged particle beam apparatus characterized in that a plurality of detectors are provided. The charged particle beam device according to claim 8.
The charged particle beam apparatus, wherein the signal processing circuit reconstructs one scanned image from each scanned image based on the charged particle signals detected by the plurality of detectors. The charged particle beam device according to claim 8.
The charged particle beam device, wherein the plurality of detectors have different detection sensitivities. In a measurement method including a step of scanning a charged particle beam converged on a sample on which a concavo-convex pattern is formed, storing the generated signal charged particles as a scanned image, and a step of measuring a dimension of the sample using the scanned image ,
Scanning the charged particle beam converged in the first scanning direction and storing the generated signal charged particles as a first scanned image;
Scanning the same region of the sample with a charged particle beam converged in a second scanning direction opposite to the first scanning direction, and storing the generated signal charged particles as a second scanned image;
Comparing the first scanned image with the second scanned image, setting the smaller signal intensity of the scanned image to a true value, and reconstructing the scanned image;
A measurement method comprising a step of measuring a dimension of a pattern bottom formed on the sample using the reconstructed scanned image. The measurement method according to claim 11,
The step of comparing the first scanned image and the second scanned image, setting the smaller signal intensity of the scanned image to a true value, and reconstructing the scanned image,
Comparing the first scanned image and the second scanned image;
A step of aligning the positions of the scanned images based on the comparison result;
Comparing each signal waveform extracted from the first scan image and the second scan image;
A step of aligning the positions of signal waveforms based on the result of comparison,
A method of comparing the signal intensities of the aligned signal waveforms, setting the smaller signal intensity as a true value, and reconstructing the signal waveform. The measurement method according to claim 11,
From the characteristics of the detection system for detecting the generated signal charged particles, further comprising the step of calibrating the signal intensity of the scanning image based on the signal charged particles,
The measuring method of measuring a dimension of the sample uses a scanned image in which the signal intensity is calibrated. The measurement method according to claim 11,
The step of storing the generated signal charged particles as a scanning image is a step of storing each scanning image acquired by a plurality of detectors,
The step of reconstructing the signal waveform is a step of reconstructing one scan image from the plurality of stored scan images.

Images