JP2010135679A - Semiconductor inspection device and inspection method - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a semiconductor inspection device capable of reducing an influence of unevenness of electrostatic charging from a detection signal in real time. <P>SOLUTION: A capacitor 114 is added in series to a transmission line for the detection signal so as to eliminate a low-frequency signal generated owing to unevenness of electrostatic charging from an actual signal. A preamplifier has an electromagnetic relay 116 connected through an electromagnetic relay 112, the signal line 113, and a signal path 115. The signal line 113 is formed of signal wires connected in parallel to each other, and capacitors 114 differing in electrostatic capacity value are connected to the respective signal wires. The electromagnetic relay 116 is connected to an AD converter through an amplifier 117, a high-frequency limiting filter 118, and an amplifier 119. Further, the AD converter is connected to an offset adjusting portion 120. <P>COPYRIGHT: (C)2010,JPO&INPIT

Description

  The present invention relates to a semiconductor inspection apparatus and inspection method for inspecting semiconductor defects.

  In the manufacturing process of semiconductor devices such as memories and microcomputers used in computers, the quality of lithography processing, etching processing, and other processing results, and the presence of defects such as the generation of foreign matter greatly affect the manufacturing yield of semiconductor devices. Effect.

  Therefore, in order to detect the occurrence of abnormalities and defects early or in advance, the inspection of the pattern on the semiconductor wafer is performed using a scanning electron microscope (SEM) type visual inspection device at the end of each manufacturing process. Has been.

  In order to perform high-throughput and high-precision inspection using the SEM type visual inspection apparatus, it is necessary to acquire an image having a high S / N ratio at a very high speed. Therefore, the number of electrons irradiated using a large current beam 1000 times or more (100 nA or more) of a normal SEM is secured, and a high SN ratio is maintained. Furthermore, high-speed and highly efficient detection of secondary electrons and reflected electrons generated from the substrate is essential. Such an inspection method is referred to as an electron beam inspection method.

  In the electron beam inspection method, an image having a higher resolution than that of the optical appearance inspection or the laser inspection can be obtained, so that it is possible to detect minute foreign matters and defects on a fine circuit pattern. In addition, due to the potential contrast in which the surface potential difference is reflected in the secondary electron emission efficiency due to the effect of charging by electron beam irradiation, circuit pattern conduction or non-conduction on the surface or lower layer, wiring or transistor short circuit, etc. It is also possible to detect electrical defects. Non-patent document 1 describes the potential contrast and the technology using this.

  On the other hand, in the case of the electron beam inspection method, the influence of charging may affect the inspection result. In other words, in the electron beam inspection, since secondary electrons or reflected electrons generated by irradiating an electron beam during inspection are detected and converted into a signal for inspection, the electron beam is continuously irradiated during inspection. . Therefore, when the local part of the surface of the wafer to be inspected is an insulating material and is easily affected by charging, when the structure to be floated from the substrate is formed on the wafer to be inspected and charging is likely to accumulate, or the electron dose irradiated to the sample If the sample is unstable, there may be uneven charging on the sample surface. When uneven charging occurs, not only the actual defect portion becomes difficult to see, but also a contrast abnormality such as a normal portion that appears bright or dark may occur and be erroneously recognized as a defect.

  Therefore, Patent Document 1 discloses a method for controlling charging by monitoring charging on the surface of a sample in real time, observing an absorption current flowing into the sample. Patent Document 2 discloses a method for measuring an electrification state with high sensitivity and acquiring an image.

JP 2006-234789 A JP 2004-146802 A “Electron, Ion Beam Handbook” (Nikkan Kogyo Shimbun) pp 622-623

  However, the methods described in Patent Documents 1 and 2 are techniques that are useful when the electron beam is scanned at a low speed because of the method for controlling the charged state of the entire measurement wafer. However, high throughput is required and high speed is required. In an electron beam inspection apparatus that scans an electron beam, it is very difficult to control charging of a local region during actual inspection.

  The reason for this is that while the inspection apparatus needs to scan the electron beam and detect the signal at high speed in order to increase the throughput as much as possible, the capacitance component (capacitance component) of the semiconductor wafer to be inspected is very large. This is because the response speed of the current value absorbed by the semiconductor wafer is very slow. That is, it takes a long time for the electrons or current irradiated to the sample to reach the ammeter.

  As a result, the measured value of the absorption current measured during the inspection is the result of superimposing the current values flowing from the local region, so it is very difficult to measure the charged state of each nano-order local region. .

  An object of the present invention is to realize a semiconductor inspection apparatus and inspection method capable of reducing the influence of charging unevenness from a detection signal in real time.

  In order to achieve the above object, the present invention is configured as follows.

  (1) The semiconductor inspection apparatus of the present invention includes a detector that detects secondary electrons or reflected electrons generated from a semiconductor wafer, and a detection signal formed from the secondary electrons or reflected electrons detected by the detector. A pass frequency adjusting unit that removes a signal having a frequency equal to or lower than a certain value; an image control unit that forms an image of the semiconductor wafer based on the detection signal output from the pass frequency adjusting unit; And a control unit that controls the operation of the detector and supplies a command signal to the pass frequency adjusting unit so that the pass frequency adjusting unit removes a signal having a frequency equal to or lower than a predetermined value.

  (2) The semiconductor inspection method of the present invention detects secondary electrons or backscattered electrons generated from a semiconductor wafer, and results from uneven charging of the semiconductor wafer from a detection signal formed from the detected secondary or backscattered electrons. In order to remove the detection signal, a signal having a frequency equal to or lower than a predetermined value is removed, and an image of the semiconductor wafer is formed based on the signal from which the signal having a frequency lower than the predetermined value is removed. judge.

  ADVANTAGE OF THE INVENTION According to this invention, the semiconductor inspection apparatus and inspection method which can reduce the influence of charging nonuniformity from a detection signal in real time are realizable.

  Embodiments of the present invention will be described below with reference to the accompanying drawings.

  FIG. 1 is a schematic configuration diagram of an SEM appearance inspection apparatus which is a semiconductor inspection apparatus to which an embodiment of the present invention is applied.

  In FIG. 1, an SEM type visual inspection apparatus includes an electron gun 1 having an electron source 20, a deflector 2, an objective lens 3, a sample stage 12, a detector 5, an image processing unit 6, and a computer 7. A monitor 8 and a controller 9 for controlling various electromagnetic lenses are provided.

  The primary electrons emitted from the electron gun 1 are deflected by the deflector 2, focused by the objective lens 3, and applied to the sample 4 disposed on the sample stage 12.

  Secondary electrons or reflected electrons emitted from the sample 4 reach the detector 5 and are then amplified by the preamplifier 10. The signal amplified by the preamplifier 10 is supplied to the AD converter 11 via a pass frequency adjusting unit 1000 described later, and is subjected to analog / digital conversion.

The signal converted into a digital signal by the AD converter 11 is sent to the image processing unit 6, where an image is constructed, and the presence or absence of a defect and the type of defect are determined from the image. The The monitor 8 is supplied with the defect position, defect type, number of defects, and the like determined by the image processing unit 6 via the computer 7 and displays them.
(Spatial frequency control method)
Hereinafter, the relationship between the pattern size and pixel size of the sample to be inspected and the principle of deriving the spatial frequency of the detection signal of secondary electrons or reflected electrons emitted from the sample will be described with reference to FIG.

  FIG. 2A shows a wiring sample 101 having a width of 100 nm and an interval of 100 nm as an inspection target sample. FIG. 2B shows an electronic signal waveform 60 emitted from the sample when the inspection object is irradiated with an electron beam with a 20 nm inspection pixel size. FIG. 2C shows an electronic signal waveform 62 emitted from the sample when the inspection object is irradiated with an electron beam at an inspection pixel size of 80 nm / pix.

  The electronic signal waveforms 60 and 62 in FIG. 2 are actual signal waveforms of the electron beam actually output from the sample to be inspected, and points 61 and 63 are points sampled by the AD converter 11.

  The basic spatial frequency fs of the electronic signal waveform output from the wiring sample 101 is expressed by the following equation (1) where f is the sampling speed of the AD converter 11, p is the inspection pixel size, and x is the thin line width and interval of the inspection target. ).

fs = p · f / (2x) (1)
If the inspection pixel size is larger than the size of the inspection object, there is a possibility that a defective part may be missed.
x> or = p (2)
Substituting equation (2) into equation (1) gives equation (3) below.

fs <or = f / 2 (3)
The above expression (3) is the condition of the Nyquist sampling theorem that “the frequency of the actual signal sampled by the A / D converter must be ½ or less of the A / D sampling frequency”. From the point of view, it is necessary to determine the inspection pixel size p so as to satisfy the expression (2).

  By changing the inspection pixel size p according to the above equations (1) and (2), the basic spatial frequency fs of the electronic signal waveform can be freely controlled. For example, the frequency of a signal to be detected can be increased by increasing the inspection pixel size p.

Specifically, assuming that the sampling rate of the detection system is 100 MHz, the sample is irradiated with an electron beam for 10 nsec per pixel, so the waveform (pixel size: 20 nm / pixel) shown in FIG. The basic spatial frequency of pix) is 10 MHz. The spatial frequency of the waveform (pixel size: 80 nm / pix) shown in (c) of FIG. 2 is 40 MHz.
(Measurement sample size estimation method)
Next, a method for actually estimating the width and interval x of the inspection object using the above equation (1) will be described.

  The sampling speed f and the inspection pixel size p of the detection system are known because they are parameters set by the apparatus. The fundamental spatial frequency fs of the output signal from the sample can be measured. For example, the basic spatial frequency fs can be obtained by measuring the gradation period of the detection signal when line scanning is performed, or by Fourier transform processing of the detection signal.

  As a result, the unknown in equation (1) is only x. That is, by using the following equation (4) obtained by modifying the equation (1), the width and the interval x of the inspection object can be measured.

x = pf / (2fs) (4)
Specifically, if the sampling rate f of the detection system is 100 MHz, the inspection pixel size p is 50 nm / pix and the electron beam is scanned, and the gradation of the detected signal is 4 pixel cycles, it is 10 nsec per pixel. The spatial frequency fs is 1 / (4 × 10 nsec) = 25 MHz.

  Next, when these values are substituted into the above equation (4), the following equation (5) is obtained, and it can be seen that the interval and the width x of the target sample were 100 nm.

x = 50 [nm] × 100 [MHz] / (2 × 25 [MHz]) = 100 nm (5)
(Low frequency component removal method and configuration)
Next, the principle of removing a signal waveform caused by uneven charging will be described with reference to FIG. The sample to be inspected here is also a thin wire sample 101 with equal intervals, as in the sample shown in FIG.

  In FIG. 3, it is assumed that uneven charging 103 occurs on the sample due to non-uniformity of the surface potential on the sample. For the sake of simplicity, it is assumed that the signal amount is large in the charging unevenness 103 and the defective portion 107 as well. In fact, when such a part is inspected, if the signal 108 from the part 103 where the uneven charging occurs exceeds the inspection threshold value 102, not only the signal 105 from the defective part 107 but also uneven charging occurs. There arises a problem that the signal 108 from the existing portion 103 is erroneously recognized as a defect.

  On the other hand, the frequency of the signal 108 due to the charging unevenness 103 shown in FIG. 3 is lower than the spatial frequency of the electronic signal 104 from the thin line portion. Therefore, if the signal 106 from which the low-frequency component is removed can be acquired, erroneous recognition can be reduced by the charging unevenness 103.

  Next, a method and configuration for removing the low-frequency signal component 108 caused by uneven charging from the actual signal will be described with reference to FIG. FIG. 4 includes the pass frequency adjusting unit 1000 shown in FIG.

  In FIG. 4, a capacitor 114 is added in series with the transmission path of the detection signal in order to eliminate the low-frequency signal 108 generated due to uneven charging or the like from the actual signal. In this way, since the frequency of the signal that can pass through can be limited, only the spatial frequency of the signal or a frequency in the vicinity thereof can be acquired from the sample by the value of the capacitor 114. That is, an image (high-frequency image) formed only with a high frequency can be created.

  As shown in FIG. 4, the preamplifier 10 is connected to an electromagnetic relay 116 via an electromagnetic relay 112, a signal path 113, and a signal path 115. The signal path 113 includes signal lines connected in parallel to each other, and capacitors 114 having different capacitance values are connected to each signal line.

  The electromagnetic relay 116 is connected to the AD converter 11 via an amplifier 117, a high-frequency limiting filter 118, and an amplifier 119. The AD converter 11 is connected to an offset adjustment unit 120.

  The electromagnetic relays 112 and 116, the signal path 113, the capacitor 114, the signal path 115, and the amplifiers 117, 119, and 120 constitute a frequency component adjustment unit 1000. The operation of the electromagnetic relays 112 and 116 is controlled by a command signal from the computer 7.

  As shown in FIG. 4, when signals are passed through several types of signal paths 113 to which capacitors 114 having different capacitance values are connected using electromagnetic relays 112, 116, etc., the frequency components of the signals to be passed are obtained. It is also possible to select a large number.

  On the other hand, when an appearance inspection apparatus using a scanning electron microscope is used, it may be desired to view a normal SEM image in order to adjust the focus and the like. Therefore, it is necessary to prepare means for displaying a normal microscope image. In that case, the signal path 115 may be selected by the electromagnetic relays 112 and 113.

  Further, when the actual signal passes through the capacitor 114, the DC component signal is also cut off, so that the image gradation may be extremely dark. Therefore, it is desirable to add a function for offsetting the brightness. For example, when the full scale of the image gradation is 256 gradations and the signal is not input, and the voltage signal from the offset adjustment unit 120 in FIG. When this is done, the brightness is an image formed with 100 gradations plus or minus high-frequency signal amplitude.

  The circuit shown in FIG. 5 is a modification of the pass frequency adjusting unit 1000 shown in FIG. In FIG. 5, it is possible to reduce the number of signal paths by connecting a controllable element such as a variable capacitor 124 that can be controlled from the control computer 7 and means having a function to the signal path 113. A control signal from the computer 7 is supplied to the variable capacitor 124 via the DA converter 121 and the amplifier 122.

  If the signal passes through the signal path 113 and then passes through the high band limiting filter 118 immediately before the AD converter 11, the frequency band as the circuit system is as shown in FIG. In FIG. 6, the frequency determined by the capacitor 114 or the variable capacitor 124 is a passing frequency 130. Reference numeral 131 denotes a frequency band when the low frequency is cut off, and 133 denotes a frequency band when the low frequency is not cut off.

  Further, a frequency determined by the high band limiting filter 118 is set as a cutoff frequency 132. That is, the circuit system is a band-pass filter (band-pass filter) that passes only the frequency from the pass frequency 130 to the cutoff frequency 132.

In the example shown in FIGS. 4 and 5, the capacitor 114 is inserted in series with the signal line. However, it is also possible to arrange an inductor between the signal line and the ground. It is also possible to cut the low frequency signal by differentiating the detected signal. A method for generating a differential image or a waveform by image processing is described in, for example, a publicly known example (Japanese Patent Laid-Open No. 2005-195361).
(High frequency image)
Next, an image (high frequency image) from which low frequency components have been cut will be described with reference to FIG. A normal SEM image is shown on the left side in FIG. 7, and a high frequency image is shown on the right side in FIG. If the inspection pixel size condition is p << x, the normal SEM image 200 is an image in which the edge of the pattern can be clearly observed, but the high-frequency image is an image 201 in which only the edge has contrast.

On the other hand, when the inspection pixel size condition is p <x or p≈x, both the normal SEM image 202 and the high-frequency image 203 are images with blurred pattern edges, and the normal SEM image 202 and the high-frequency image 203 are almost the same. Equivalent image.
(Pass frequency determination method)
FIG. 8 is a graph showing the relationship between the pass frequency 130 (horizontal axis) and the number of detected defects (vertical axis). Here, for example, the spatial frequency of the actual signal from the sample is 10 MHz. That is, FIG. 8 is a graph when the sampling rate f of the detection system is 100 MHz, the pixel size p is 20 nm / pix, and the fine line width x is 100 nm from the above equation (1).

  If a band restriction that prevents a signal having a frequency of 10 MHz from passing though the spatial frequency is 10 MHz is provided, the signal amplitude of the high-frequency component is reduced. As a result, as shown by the C region in FIG. 8, the number of defects becomes small, and the defects cannot be recognized. For this reason, the number of detected defects is reduced, and a defect count plot 142 is obtained.

  Next, assuming that the band limit is set to 100 kHz as shown in the area A in FIG. 8, when the low frequency component 108 due to the charging unevenness is 100 kHz or less, the number of defects as shown in FIG. A plot 141 can be obtained.

  On the other hand, when the low frequency component due to the charging unevenness is 100 kHz or more, there is a possibility that it is erroneously recognized as a defect. Therefore, the number of defects detected as shown in FIG.

  That is, it can be determined that the number of defects fluctuates due to a change in frequency, and the increased portion is due to uneven charging, and the decreased portion is the frequency at which the defect detection accuracy decreases. Therefore, in this way, the relationship between the spatial frequency and passing frequency of the detection signal and the number of defects is examined, the optimum condition in the range B in FIG. 8 is found, and the electromagnetic relay is selected so as to select the capacitor 114 having an appropriate value. 112 and 116 can be set by a command from the computer 7.

  FIG. 9 is an operation flowchart in one embodiment of the present invention.

  In FIG. 9, first, the inspection pixel size p and the detection sampling frequency f are set (step S1). Next, it is determined whether or not the pattern size or the basic period x of the sample to be inspected is known (step S2). If the pattern size or the basic period x is not known in step S2, the process proceeds to step S3, the electron beam scanning is performed, and the inspection object derived from the spatial frequency fs of the electronic signal obtained from the detected signal and the equation (4) The pattern size or basic period x of the sample is calculated (step S4). Then, the process proceeds to step S5. Since the relationship between the pattern size x obtained from the detected signal and the inspection pixel size p needs to satisfy the expression (2), it is possible to limit the range in which the inspection pixel size p can be set.

  In step S2, if the pattern size or the basic period x2 is known m, the process proceeds to step S5.

  Next, in order to display a normal SEM image, a normal image display mode is set, and an electron optical system such as a focus is adjusted (step S5). If already adjusted, this step S5 is not necessary.

  Next, the signal path (or passing frequency) is switched to switch to a high frequency image (step S6).

  Next, a test inspection is performed, and a graph (or table) showing the relationship between the signal path (or passing frequency) and the number of detected defects is created. The same test inspection is performed by switching the signal path (or passing frequency), and the state is displayed or recorded as needed (steps S6 to S9).

  The optimum signal path (or passing frequency) is selected and determined from the results of the test inspection and the data analysis, and the final conditions for this inspection are determined (step S10).

  Further, since the relationship between the signal path (or passing frequency) and the number of detected defects changes when the inspection threshold 102 is changed, it is necessary to perform the above flow again while changing the inspection threshold 102. That is, when the inspection threshold is lowered, noise is also recognized as a defect, and the number of detected defects increases. On the other hand, if the inspection threshold value is increased, noise will not be recognized as a defect, but an actual defect may be missed. Therefore, the optimum inspection condition is found by changing the inspection threshold and the passing frequency.

  By the above procedure, an optimal inspection operation can be performed in which low-frequency components generated due to uneven charging are removed.

Further, when the spatial frequency of the detection signal is changed, the inspection time throughput and the image S / N are also changed. For example, if the spatial frequency of the detection signal is increased, the throughput of the inspection time increases and the image S / N decreases. Generally, inspection time throughput and image S / N are in a trade-off relationship. Therefore, it is only necessary to find the optimum conditions for the throughput of the inspection time and the image S / N in the region B in FIG.
(Screen operation)
An example of an input screen displayed on the monitor 8 in order to determine and inspect the pass frequency 130 will be described with reference to FIG.

  In FIG. 9, in the child window 150, a part 151 for setting the inspection pixel size p, a part 152 for setting the detection sampling frequency f, a button 153 for starting scanning or image acquisition with an electron beam, or a button 154 for stopping. A part 155 for displaying the spatial frequency fs of the electronic signal obtained from the detected signal, a part 156 for displaying the calculation result of the pattern size x to be inspected using the equation (4), and the signal shown in FIG. And a portion 157 for selecting a path (or passing frequency 130). Further, a display unit of a detected signal image 170 and a waveform 171 and a graph 160 (or table 161) showing the relationship between the signal path (or passing frequency 130) and the number of detected defects is provided.

  Hereinafter, an example of a procedure for determining an optimum signal path (or pass frequency 130) using the screen displayed on the monitor 8 will be described.

  First, the operator sets the inspection pixel size p and the detection sampling frequency f to arbitrary values using the screen parts 151 and 152 in order to obtain the pattern size x or the basic period of the sample to be inspected.

  Next, a button 153 for starting electron beam scanning or image acquisition is pressed, and the spatial frequency fs of the electronic signal obtained from the detected signal and the pattern size or basic period x of the sample to be inspected derived from the equation (4) are determined. 155 and 156 are automatically displayed.

  At this time, as shown in FIG. 10, an image 170 or a line signal 171 may be displayed in the child window. If the pattern size or basic period x of the sample to be inspected is known, the procedure for deriving the pattern size or basic period x of the sample to be inspected is not necessary.

  Since it is desirable that the relationship between the pattern size x obtained from the detected signal and the inspection pixel size p satisfies the expression (2), the value displayed on the setting portion 151 is manually or automatically set here. It is also possible to change to.

  Next, in order to display a normal SEM image, the signal path (or passing frequency) selection unit 157 sets the normal image display mode. Here, an electron optical system such as a focus is adjusted. Next, the part 157 for selecting the signal path (or the passing frequency) is switched to switch to the high frequency image.

  Subsequently, the test inspection start button 158 is pressed to perform a test inspection. A graph 160 (or table 161) indicating the relationship between the signal path (or passing frequency 130) and the number of detected defects is displayed or recorded as a test result. In order to stop, the stop button 159 is operated.

  The same test inspection is performed by switching the portion 157 for selecting the signal path (or the passing frequency 130), and the state is displayed or recorded as needed.

  These results are displayed on a graph 160 (or Table 161) showing the relationship between the signal path (or passing frequency 130) and the number of detected defects.

  From the results of the test inspection and the data analysis result, the optimum signal path (or pass frequency 130) can be selected and determined by the check box 162 or the like, and the final conditions of the main inspection are determined.

  By the above procedure, an optimal inspection operation can be performed in which low-frequency components generated due to uneven charging are removed.

  Then, the start and stop of this inspection can be instructed with buttons 163 and 164.

  Also, as shown in FIG. 11, it is possible to provide a part 172 for setting an inspection threshold value, display the inspection threshold value 173, and summarize the number of defects when the passing frequency is a parameter in one table. .

  As described above, according to an embodiment of the present invention, a frequency that causes erroneous recognition of the number of defects due to charging unevenness is detected in advance, and the frequency component is removed from the detection signal. The filter circuit constant (capacitor value that cuts low frequency) is set, the set frequency component is removed from the detection signal from the actual semiconductor wafer, and the number of defects is inspected. It is possible to suppress erroneous detection of defects due to unevenness.

  Therefore, it is possible to realize a semiconductor inspection apparatus and inspection method that can reduce the influence of charging unevenness in real time from the detection signal.

  The above-described example is an example when the present invention is applied to an SEM visual inspection apparatus, but the present invention can be applied not only to an SEM general inspection apparatus but also to other electron beam inspection apparatuses. .

It is a schematic block diagram of the SEM type external appearance inspection apparatus with which one Embodiment of this invention is applied. It is a derivation principle explanatory drawing of the spatial frequency of the detection signal of a secondary electron or a backscattered electron in the present invention. It is a principle explanatory drawing which removes the signal waveform which arises by charging unevenness in the present invention. It is a block diagram of the pass frequency adjustment part for controlling the pass frequency in one Embodiment of this invention. It is a figure which shows the modification of the pass frequency adjustment part shown in FIG. It is a figure explaining the frequency band of the frequency-controlled detection signal in the present invention. It is a figure explaining the pixel size dependence of the normal image and high frequency image in this invention. It is a graph explaining the relationship between a passage frequency and the number of detected defects in the present invention. It is an operation | movement flowchart of one Embodiment of this invention. It is an example of the input screen of the monitor of the SEM type external appearance inspection apparatus in one Embodiment of this invention, and is a figure which shows an example of a pass frequency determination screen. It is an example of the input screen of the monitor of the SEM type external appearance inspection apparatus in one Embodiment of this invention, and is a figure which shows the other example of a pass frequency determination screen.

Explanation of symbols

  DESCRIPTION OF SYMBOLS 1 ... Electron gun, 2 ... Deflector, 3 ... Objective lens, 4 ... Measurement sample, 5 ... Detector, 6 ... Image processing control part, 7 ... Computer, DESCRIPTION OF SYMBOLS 8 ... Monitor, 9 ... Electron gun control part, 10 ... Preamplifier, 11 ... AD converter, 12 ... Sample stage, 20 ... Electron source, 112, 116 ... Electromagnetic Relay 113, signal path, 114, capacitor, 115, signal path without capacitor, 117, 119, 122 ... amplifier, 118 ... high band limiting filter, 120 ... Offset adjustment unit, 121 ... DA converter, 124 ... variable capacitor

Claims (20)

In a semiconductor inspection apparatus for irradiating a semiconductor wafer with an electron beam to acquire an image of the semiconductor wafer and detecting defects in the semiconductor wafer,
A detector for detecting secondary electrons or reflected electrons generated from the semiconductor wafer;
A passing frequency adjusting unit for removing a signal having a frequency equal to or lower than a predetermined value from a detection signal formed from secondary electrons or reflected electrons detected by the detector;
Based on the detection signal output from the pass frequency adjusting unit, an image control unit that forms an image of the semiconductor wafer and determines defects of the semiconductor wafer;
A control unit for controlling the operation of the detector and for supplying a command signal to the pass frequency adjusting unit so that the pass frequency adjusting unit removes a signal having a frequency equal to or lower than a predetermined value;
A semiconductor inspection apparatus comprising:   2. The semiconductor inspection apparatus according to claim 1, wherein the pass frequency adjusting unit includes means for removing a signal having a frequency equal to or less than an integral value from a detection signal formed from secondary electrons or reflected electrons detected by the detector. A semiconductor inspection apparatus characterized in that any one of them is selected by the control unit.   3. The semiconductor inspection apparatus according to claim 2, wherein the pass frequency adjusting unit includes a plurality of capacitors having different capacitance values, and one of the plurality of capacitors is selected according to a command signal from the control unit. A semiconductor inspection apparatus, wherein a detection signal from the detector is passed through a selected capacitor.   4. The semiconductor inspection apparatus according to claim 3, wherein the pass frequency adjusting unit includes a variable capacitor capable of changing a capacitance value, and a capacitance value of the variable capacitor is set according to a command signal from the control unit. In addition, a semiconductor inspection apparatus in which a detection signal from the detector is passed.   The semiconductor inspection apparatus according to claim 3, wherein the control unit changes a frequency of a detection signal detected by the detector, calculates a relationship between the frequency of the detection signal and the determined number of defects of the semiconductor wafer, Based on the relationship between the calculated frequency and the number of defects, the frequency of the signal generated by the uneven charging of the semiconductor wafer is determined, and a command signal for removing a detection signal having a frequency equal to or lower than the determined frequency is supplied to the passing frequency adjusting unit. A semiconductor inspection apparatus.   3. The semiconductor inspection apparatus according to claim 1, wherein the semiconductor inspection apparatus irradiates the semiconductor wafer with an electron beam using a scanning electron microscope to detect secondary electrons or reflected electrons generated from the semiconductor wafer. Semiconductor inspection equipment. In a semiconductor inspection method for irradiating a semiconductor wafer with an electron beam to acquire an image of the semiconductor wafer and detecting defects in the semiconductor wafer,
Detect secondary electrons or backscattered electrons generated from the semiconductor wafer,
In order to remove the detection signal generated from uneven charging of the semiconductor wafer from the detection signal formed from the detected secondary electrons or reflected electrons, a signal having a frequency equal to or lower than a certain value is removed,
Based on the signal from which the signal having a frequency equal to or lower than the predetermined value is removed, an image of the semiconductor wafer is formed, and defects of the semiconductor wafer are determined.
A semiconductor inspection method.   8. The semiconductor inspection method according to claim 7, wherein it is possible to select whether or not to remove a signal having a frequency equal to or lower than an integral value from the detection signal formed from the detected secondary electrons or reflected electrons. Method.   9. The semiconductor inspection method according to claim 8, wherein one of a plurality of capacitors having different capacitance values is selected, and the detection signal is passed through the selected capacitor.   9. The semiconductor inspection method according to claim 8, wherein the detection signal is passed through a variable capacitor whose capacitance value can be changed, and the capacitance value of the variable capacitor is set to remove a signal having a frequency equal to or less than an integral value. A semiconductor inspection method.   9. The semiconductor inspection method according to claim 8, wherein the frequency of the detection signal is changed, the relationship between the frequency of the detection signal and the determined number of defects in the semiconductor wafer is calculated, and the relationship between the calculated frequency and the number of defects is calculated. A semiconductor inspection method comprising: determining a frequency of a signal generated due to uneven charging of the semiconductor wafer, and removing a detection signal having a frequency equal to or lower than the determined frequency.   8. A semiconductor inspection method according to claim 1, wherein a secondary electron or a reflected electron generated from the semiconductor wafer is detected by irradiating the semiconductor wafer with an electron beam using a scanning electron microscope. In a semiconductor inspection apparatus that acquires an image of a circuit pattern formed on a semiconductor wafer and determines a defect,
Means for setting the inspection pixel size;
Means for setting the sampling speed or frequency of the image detection signal of the semiconductor wafer;
At least two different signal paths with different frequency characteristics;
Means for selecting at least one of the signal paths;
A semiconductor inspection apparatus for inspecting a semiconductor wafer using a detection signal having a specific frequency component determined from the inspection pixel size, the sampling speed or frequency, and the signal path. In a semiconductor inspection apparatus that acquires an image of a circuit pattern formed on a semiconductor wafer and determines a defect,
Means for setting the inspection pixel size;
Means for setting the sampling rate or frequency of the image signal of the semiconductor wafer;
Means for digital signal processing for controlling the frequency component of the detection signal;
And a semiconductor inspection apparatus for inspecting the semiconductor wafer using a detection signal having a specific frequency component determined from the inspection pixel size, the sampling speed or frequency, and the digital signal processing.   14. The semiconductor inspection apparatus according to claim 13, further comprising a display unit that displays an image composed of high-frequency components from which low-frequency components have been removed.   14. The semiconductor inspection apparatus according to claim 13, wherein a means for setting a pixel size, a means for setting a speed at which a detection signal is sampled, a means for setting a frequency at which the detection signal passes or is cut off, and a test inspection are performed. And a means for displaying the test result, and a screen for determining a frequency for passing or blocking the detection signal based on the test result is displayed on the display means. A featured semiconductor inspection device.   17. The semiconductor inspection apparatus according to claim 16, wherein the size or interval of the sample pattern to be inspected is calculated using a spatial frequency of a detection signal calculated using the pixel size and the sampling speed. Inspection device.   18. The semiconductor inspection apparatus according to claim 17, wherein the pixel size smaller than the calculated size or interval of the inspection target sample pattern is set. In a semiconductor inspection method for determining a defect by acquiring an image of a circuit pattern formed on a semiconductor wafer,
Set the inspection pixel size,
Set the sampling speed or frequency of the image detection signal of the semiconductor wafer,
Selecting at least one of signal paths having at least two different frequency characteristics;
A semiconductor inspection method, comprising: inspecting the semiconductor wafer using a detection signal having a specific frequency component determined from the inspection pixel size, the sampling speed or frequency, and the signal path. In a semiconductor inspection method for determining a defect by acquiring an image of a circuit pattern formed on a semiconductor wafer,
Set the inspection pixel size,
Set the sampling speed or frequency of the image detection signal of the semiconductor wafer,
Digital signal processing to control the frequency component of the detection signal,
A semiconductor inspection method for inspecting the semiconductor wafer using a detection signal having a specific frequency component determined from the inspection pixel size, the sampling speed or frequency, and the digital signal processing.

Images