JP2008032582A - Foreign matter/flaw-inspecting device and foreign matter/flaw inspection method - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To solve the problem, wherein although it is necessary to enhance the resolving power of a θ encoder, in order to enhance the resolving power (accuracy) of flaw detection, there is a limit to the resolving power of the encoder that can be improved and it is difficult to improve the positional accuracy of flaw detection. <P>SOLUTION: This foreign matter/flaw-inspecting device is provided with a means, constituted so as to make the sampling time interval of A/D conversion for performing the flaw detection smaller than the time interval of angle data (θ encoder) obtained, when a wafer is rotated and transferred to measure the place where the peak position thereof exists, within one unit (to measure Δθ) from the sampled A/D conversion result. <P>COPYRIGHT: (C)2008,JPO&INPIT

Description

  The present invention relates to an inspection apparatus, and more particularly to an apparatus for inspecting an object to be inspected while rotating it.

  There are two types of semiconductor wafer inspection apparatuses: an inspection apparatus that inspects while rotating the wafer and an inspection method that scans the wafer in the XY directions. The method according to the present invention relates to a method for detecting a foreign substance or a defect on a wafer by using a scattered light reflected on the wafer by optically irradiating a beam while rotating the wafer.

  As a method for detecting the scattered light intensity of a wafer, a signal output from an angle detector (encoder) attached to a rotary stage is used to perform signal processing by A / D converting the detection signal.

  Conventionally, for example, there is a foreign matter inspection apparatus described in Patent Document 1 (Japanese Patent Laid-Open No. 6-242012). As a control method of A / D conversion, a detection signal is sampled for each scanning position as shown in FIG. Further, as described in paragraph (0015) of Patent Document 1, a linear encoder is provided in the straight traveling direction (R direction), and a rotary encoder is provided in the rotary table. Signals from the respective encoders Thus, the present position is detected, and the data acquisition for each scanning position shown in FIG. 3 is performed with reference to the present position.

  Conventionally, rotary encoders are used for position detection in the rotation direction, and A / D conversion is performed based on angle detection signals (rotary encoders). Therefore, in order to detect defect positions with high accuracy, It is necessary to improve the resolution. In order to increase the resolution of the angle detector (encoder), an expensive encoder is required. Furthermore, since the encoder has a structure having an optical slit in the disk, there is a limit in improving the resolution.

JP-A-6-242012

  In the method of measuring the scattered light intensity by measuring the intensity of scattered light using a signal (hereinafter referred to as the θ pulse or encoder pulse) according to the rotation angle per rotation of the wafer, the resolution of the θ pulse is a foreign particle / defect in the rotation direction. It affects the coordinate resolution.

  An object of the present invention is to detect foreign matter / defects with higher resolution than the angle detection resolution of an inspection object, and to improve the coordinate accuracy of foreign matter / defect detection.

  One feature of the present invention is that, in a foreign object / defect detection apparatus, an object moving stage in which main scanning is rotational movement at a constant rotational angular velocity and sub-scanning is translational movement, a laser light source, and the laser light source are emitted. Illumination means for irradiating laser light as an illumination spot of a predetermined size on the surface of the object to be inspected, and detecting light that is scattered, diffracted, or reflected at the illumination spot and converting it to an electrical signal Scattering / diffraction / reflected light detection means, A / D conversion means for converting the electrical signal into digital data, and angular coordinate information in the main scanning direction of the position where the illumination spot is irradiated on the surface of the object to be inspected Angular coordinate detection means for detecting, particle size calculation means for calculating the size of foreign matter and defects from the digital data, and Foreign matter / defect coordinate calculation means for calculating a position coordinate value on the surface of the object to be inspected, and the A / D conversion means samples the electrical signal at a constant sampling time interval, One unit time is smaller than the increment (one unit) time for updating the angle information from the angle coordinate detecting means.

  Another feature of the present invention is that the surface of the object to be inspected is irradiated with laser light, at least scattered light is detected and converted into an electric signal, the electric signal is converted into digital data, and the laser light is irradiated. Detecting the angle information of the inspected object at the position, calculating the size of the foreign object or defect from the digital data, and calculating the position of the foreign object or defect on the surface of the inspected object based on the angle information In the conversion to the digital data, the electric signal is sampled at a constant sampling time interval, and the foreign matter / defect detection method is such that the sampling time interval is smaller than the update interval of the angle information.

  Still another feature of the present invention is that the surface of the object to be inspected is irradiated with laser light, at least scattered light is detected and converted into an electric signal, the electric signal is converted into digital data, and the laser light is irradiated. The angle information of the object to be inspected at the position being detected is detected, the size of the foreign matter or defect is calculated from the digital data, and the position of the foreign matter or defect on the surface of the inspection object is calculated based on the angle information. In the conversion to the digital data, the electrical signal is sampled at a substantially constant sampling time interval, and the angle information from the angular coordinate detection means is updated for one unit time of the sampling time interval. It was configured to be smaller than one unit time, and sampled at every unit time when the angle information was updated based on the sampled data The maximum value of the digital data, in the foreign matter-defect detection method for measuring the time of sampling the maximum value.

  The above and other features of the present invention will be further explained by the following description.

  According to the present invention, foreign matter / defects can be detected with higher resolution than the angle detection resolution of the inspection object, and the coordinate accuracy of foreign matter / defect detection can be improved.

  Hereinafter, embodiments of the present invention will be described with reference to the drawings. However, the apparatus and method of the present invention are not limited to the configurations shown in the drawings, and various modifications are possible within the scope of the technical idea. Is possible.

  First, a configuration example of a semiconductor wafer foreign matter / defect detection apparatus as an example of an inspection object will be described. FIG. 12 is an embodiment showing the entire configuration of a foreign matter / defect inspection apparatus using the foreign matter / defect detection method of the present invention. The wafer 101 that is an object to be inspected is vacuum-sucked by a chuck 1204, and this chuck 1204 is mounted on an object-to-be-inspected moving stage 1205 and a Z stage 1208 including a rotary stage 1206 and a translation stage 1207. An illumination / detection optical system 1201 disposed above the wafer 101 is an illumination and detection optical system. A laser light source is used as the light source 1202 for the illumination light, and the irradiation beam emitted from the light source 1202 passes through the irradiation lens to form a spot-shaped beam, which is irradiated onto the wafer 101 and scattered by the foreign matter or defect 1200. Is detected by the photodetector 1203 and converted into an electrical signal, which is then input to the amplifier circuit 1211.

  The inspected object moving stage 1205 changes the rotational movement θ that is the main scanning and the translation movement R that is the sub-scanning in combination with time, so that the illumination spot is spirally scanned over the entire surface of the wafer 101.

  While the rotary stage makes one rotation, the sub-scan moves by Δr. In this embodiment, the illumination spot is scanned from the inner periphery to the outer periphery of the wafer 101, but it may be reversed. In this embodiment, the rotation stage 1206 is driven at a substantially constant angular velocity and the translation stage 1207 is driven at a substantially constant linear velocity in the entire region from the inner periphery to the outer periphery of the wafer 101. As a result, the relative linear velocity of the illumination spot with respect to the surface of the wafer 101 is higher at the outer periphery than at the inner periphery. An inspection coordinate detection mechanism 1214 is attached to the inspection object moving stage 1205 in order to detect the main scanning coordinate position θ and the sub-scanning coordinate position r being inspected. In this embodiment, an optical reading rotary encoder is used to detect the main scanning coordinate position θ, and an optical reading linear encoder is used to detect the sub scanning coordinate position r. Any other detection principle may be used as long as it can be detected. In this configuration, the foreign matter or defect 1200 passes through the illumination spot, and a light scattered light signal is obtained from the photodetector 1203. In this embodiment, a photomultiplier tube is used as the light detector 1203. However, a light detector of another detection principle may be used as long as it can detect scattered light from a foreign substance with high sensitivity. As described above, in this embodiment, the rotary stage 1206 is driven at a substantially constant angular velocity in the entire area from the inner periphery to the outer periphery of the wafer 101, and the relative linear velocity of the illumination spot relative to the surface of the wafer 101 is The outer circumference is faster than the inner circumference.

  FIG. 1 is a diagram showing a rotation angle of a semiconductor wafer in a unit time. FIG. 2 is a diagram illustrating waveforms of defect detection signals in the outer peripheral portion and the inner peripheral portion.

  Since the wafer 101 rotates at a substantially constant angular velocity, when the illumination spot scans the outer peripheral portion, the outer peripheral portion linear velocity 102 becomes high, and when the illumination spot scans the inner peripheral portion. , The inner peripheral linear velocity 103 becomes low. Accordingly, the time for which the foreign matter on the wafer 101 crosses the illumination spot is shorter when the previous foreign matter is on the outer peripheral portion of the wafer 101 than on the inner peripheral portion, and thus the light detection shown in FIG. As shown in FIG. 2, the time-varying waveform of the scattered light signal obtained from the amplifier 1203 via the amplifier 1211 is generally detected as the outer peripheral portion, that is, the position where the foreign matter is located at a larger radial position in the sub-scanning direction. As shown by the waveform 202, the half width of the signal peak is reduced. On the other hand, the half-width of the signal peak increases as the inner peripheral portion, that is, the place where the foreign matter is smaller in the radial position in the sub-scanning direction, as the inner peripheral foreign matter detection waveform 203.

  Next, signal processing in the inspection apparatus of this embodiment will be described with reference to FIG. The scattered light signal from the photodetector 1203 is amplified by an amplifier 1211, then sampled at a predetermined sampling interval ΔT by an A / D converter 1212 and converted into digital data. The sampling interval ΔT determines the period of the A / D conversion sampling signal 204 so that the outer peripheral foreign matter detection waveform 202 and the inner peripheral foreign matter detection waveform 203 which are signal waveforms shown in FIG. 2 can be sampled with sufficient time resolution. By this sampling, a time series digital data group corresponding to the detection signal is obtained. By the way, this time-series digital data group includes signal components such as noise components in addition to the intensity information of scattered light corresponding to the size of the detected foreign matter / defect that is originally required. In general, a noise signal component does not take a constant value because it changes due to fluctuations of an electric signal or mechanical vibrations of an optical system. Therefore, in order to correctly calculate the size of the foreign matter / defect, it is necessary to remove the influence of this noise component. Therefore, in this embodiment, data processing such as filtering processing is performed on the signal of the A / D converter 1212.

  The scattered light intensity value obtained as a result of the data processing is compared with a predetermined detection threshold value by the foreign object / defect determination mechanism 1213. If the scattered light intensity value is equal to or greater than the threshold value, the foreign object The defect determination mechanism 1213 generates foreign matter / defect determination information. When the foreign matter / defect determination information is generated, the foreign matter / defect coordinate detection mechanism 1215 calculates the coordinate position of the detected foreign matter / defect based on the information from the inspection coordinate detection mechanism 1214. The particle size calculation mechanism 1210 calculates the size of the detected foreign matter / defect from the scattered light intensity value.

  The foreign object / defect coordinate detection mechanism 1215 generates an R coordinate and generates a rotation angle θ coordinate of the rotating wafer.

FIG. 3.1 is a diagram showing a signal sampling method of the present embodiment on the inner circumference. FIG. 3.2 is a diagram showing a signal sampling method of the present embodiment on the outer periphery. The inner peripheral foreign matter detection waveform 203 and the outer peripheral foreign matter detection waveform 202, which are scattered detection signals, are A / D converted by the A / D conversion sampling signal 301, and a converted digital signal corresponding to the timing of each sampling signal is generated. At this time, the A / D conversion sampling signal 301 generates A / D conversion timing more finely than the encoder signal 304 corresponding to the rotation angle of the wafer. As shown in FIGS. 3.1 and 3.2, the A / D conversion means samples the electrical signal at a substantially constant sampling time interval, and the sampling time interval is rotated by the main scanning. It is desirable that at least two or more points are sampled within the time required for one point on the object to be inspected to pass through the illumination spot in the outer periphery as well as the inner periphery on the object. . The reason for using two or more points is to provide a resolution that is at least twice the encoder signal interval, and the higher the number of points, the better the resolution. For example, if the frequency of the encoder signal is about 10 MHz, the AD conversion frequency is about 80 MHz to 200 MHz. The upper limit of the AD conversion frequency is determined by the performance of the AD conversion element to be used.

  FIG. 4 is a diagram showing an outline of one embodiment of the Δθ calculation method of the present invention. The inner peripheral foreign matter detection waveform 203 which is a scattered light detection signal is A / D converted at each timing of the A / D conversion sampling signal 301 and is included in the θ encoder signal 304 at each A / D conversion timing. The signal strength 401 having the maximum signal strength is obtained as a peak value from the detected signal strengths 411, 412, 401, 414, and 415, and the time 403 from the θ encoder to the peak detection position is measured. Here, the period of the encoder signal 304 that is the θ encoder pulse and the sampling signal 301 for A / D conversion is set to be ½ or less of the period of the encoder signal 304 for the A / D conversion sampling signal 301. Δθ, that is, the θ angle up to the peak position of the scattered light within the θ encoder pulse width, is a time measurement until the scattered light intensity is maximized based on the measurement result 402 of the pulse interval of the encoder signal 304 and the θ encoder pulse. It is obtained as data represented by the ratio of the peak detection position time 403 from the θ encoder indicating the result. In the case of FIG. 4, Δθ is approximately 0.5.

  This will be described using a specific example. A rotating stage on which a wafer is placed is rotating at 4000 rpm, and an encoder (θ encoder) that generates approximately 226,800 pulses / pulse of one round is attached to the rotating body. In this case, the period of the θ pulse (the pulse interval measurement result 402 of the θ encoder signal in the figure) is about 66 ns (15 MHz). On the other hand, A / D conversion is performed at a sampling frequency of the A / D conversion circuit (A / D conversion circuit sampling frequency 404 in the figure) at 80 MHz. In order to measure (calculate) the maximum value and the maximum position of the input signal from the A / D converted signal at 80 MHz, compared to the conventional case of measuring in units of θ encoder pulses, the θ direction Resolution is improved. In the present embodiment, the accuracy in the θ direction at the outermost periphery (300Φ) is about 4 μm in the resolution of the θ encoder, and according to this circuit configuration with Δθ, is about 4 μm / 5, and the detection position accuracy of 0.8 μm. It becomes possible.

  FIG. 5 is a diagram showing another embodiment of the Δθ calculation method of the present invention. As in FIG. 4, the positions of the maximum scattered light intensity values 501 and 502 are detected using the encoder signal 304 and the A / D conversion sampling signal 301. For example, in the case of the scattered light intensity signal 511, the Δθ time measurement result of the maximum value 501 in the scattered light intensity is an arrow 503 with respect to the pulse interval measurement result 402 of the θ encoder signal, and in the case of the scattered light intensity signal 512. , The time measurement result of Δθ is an arrow 504 with respect to the pulse interval measurement result 402 of the θ encoder signal.

FIG. 6 is a waveform diagram in which the scattered light intensity signal is saturated. When the scattered light intensity signal is large, that is, for the light scattered from a large defect, the inner peripheral foreign matter detection waveform 203 that is the scattered light intensity signal may be saturated. When it is assumed that the light is not saturated, the scattered light has a waveform indicated by a dotted line and shows a peak value 602. However, the peak position and peak value 602 cannot be detected as shown by the saturation signal waveform 604 by passing through a photomultiplier or an analog circuit for signal processing. Although FIGS. 4 to 6 have been described using the inner peripheral foreign matter detection waveform 203 as an example of the scattering detection signal, it is needless to say that the outer peripheral foreign matter detection waveform 202 corresponds to the outer peripheral portion inspection. Also, as shown in FIG. 4, the sampling time interval is such that one point on the object to be inspected is the illumination spot not only on the inner periphery on the object to be inspected rotated in the main scanning but also on the outer periphery. It is preferable that at least two or more points are sampled within the time passing through, and it is more preferable that the sampling is performed at five or more points. The reason for sampling at 5 points or more is that in the case of a saturated signal, fitting processing is performed using ± 2 points of unsaturated signals at least before and after the center of the signal.

  FIG. 7 shows points when the saturation signal is A / D converted by the A / D conversion sampling signal. A / D conversion is performed by an A / D sampling signal, and a plurality of sampling signals 701, 702, 703-1 to 703-n are detected.

  The present embodiment uses the data of the sampling signals 701, 702, 703-1 to 703-n, and uses the position of the maximum value and the position of the maximum value of the scattered light peak value 602 when not saturated as shown in FIG. The size is calculated.

  FIG. 8 shows an embodiment of the Δθ detection circuit of the present invention. The scattered light signal 801 is input to the A / D conversion circuit 806 via the amplifier circuit 805, A / D converted by the A / D conversion sampling signal generated by the sampling signal generation circuit 808, and the maximum position detected as a digital signal. Input to the circuit 809. The maximum position detection circuit 809 receives the θ clock signal 810, the sampling clock 812, and a plurality of data 813 in the A / D converted θ clock signal as inputs, and the maximum value 401 and the maximum of the scattered light signal shown in FIG. The peak detection position time 403 from the θ encoder, which is the value of Δθ up to the value, is calculated, and the maximum scattered light value 802 and the maximum value Δθ 803 within the θ pulse period are output. By calculating the maximum value and Δθ up to the maximum value using data sampled at a frequency earlier than the θ pulse by the method shown in FIG. 8, it is possible to calculate and detect the peak position with higher accuracy than the θ pulse interval. Become.

  FIG. 9 shows an example of a saturation signal fitting circuit. The scattered light signal 801 is input to the A / D conversion circuit 806 via an amplifier circuit 805 (not shown), and whether or not it is saturated is detected by the saturation detection circuit 901. The number of sampling points 912 saturated by the detection circuit 902 is counted. On the other hand, the A / D converted digital signal obtains two points (A point signal 913 and B point signal 914) of the unsaturated signal corresponding to 701 and 702 shown in FIG. 7 from the unsaturated signal. The saturation point number and the non-saturation points A and B calculated by the above circuit are input to the maximum position detection circuit 904, and the maximum value and maximum position of the saturation signal are fitted (for example, extrapolated by Gaussian fitting). Process). The fitting process of the maximum position detection circuit 904 may be either a method using a calculation formula such as Gaussian fit or a method using a lookup table.

  FIG. 10 is a diagram showing another embodiment of the present invention. The scattered light signal is superimposed with high-frequency noise (noise) by passing through a photomultiplier or an analog circuit, and this noise is a factor that degrades the accuracy of defect detection. FIG. 10 shows a Δθ detection circuit having a filter circuit. The output of the A / D conversion circuit 806 is input to the maximum position detection circuit 809 via the digital filter circuit 1001. Since the A / D conversion circuit 806 is sampled at a frequency several times faster than the θ clock 810, and the scattered light signal converted at the fast frequency can be removed by passing the digital filter circuit 1001, the maximum value is obtained. In addition, the maximum position detection accuracy is improved. The digital filter circuit 1001 uses, for example, a convolution filter. Since the detected signal is a signal on a pulse having a pulse width of several tens to several hundreds ns, a moving average process using a convolution filter constitutes a region pass filter, and noise superimposed on the pulse is reduced. There is an effect to remove.

FIG. 11 is a diagram showing another embodiment of the Δθ detection circuit. FIG. 11 shows a method in which an addition circuit 1002 is provided between the A / D conversion circuit 806 and the maximum position detection circuit 809. The adder circuit calculates a Σ (total) value 1011 of A / D conversion in the θ pulse for the digital data A / D converted by the A / D conversion sampling signal generated from the sampling signal generation circuit 808. Operate. The adding circuit 1002 does not calculate the peak point of the scattered light signal, but calculates the area of the scattered light signal in the θ pulse. Data corresponding to this area is input to the maximum position detection circuit 809 and output as a maximum value 802. On the other hand, Δθ803 of the maximum scattered light value within the θ pulse period indicating the maximum position signal is obtained by inputting the A / D conversion signal 1012 as in FIG.

  FIG. 13 is a diagram illustrating a configuration of a signal processing unit of the semiconductor inspection apparatus according to the present embodiment. The light output from the laser light source 1301 for inspecting the semiconductor is irradiated onto the wafer 101, and the scattered light is received by a plurality of photos 1311, 1312, and 1313. The wafer 101 is mounted on a rotary moving table and moves as indicated by an arrow 1302 while rotating. If a defect such as a scratch or a foreign substance adheres to the wafer, the intensity change of the scattered light occurs, and this light intensity change is converted into an electrical signal by a plurality of (in the embodiment, three) photomultipliers, and each amplifier circuit 1321 , 1322, 1323 through a plurality of A / D conversion circuits 1331, 1332, 1333, individually converted into digital signals, and further, signal processing circuits 1341, 1342, 1343 for identifying the positions and shapes of defects Are individually processed, and a higher-level signal processing device 1351 further identifies defects from the results of a plurality of channels. The circuits for detecting Δθ according to the present embodiment are incorporated in the A / D conversion circuits 1331, 1332, and 1333 shown in FIG.

FIG. 14 is a diagram illustrating an example of a positional relationship among a wafer, a light source, and a light receiving element. The laser beam 1400 is irradiated from the lower side of the figure, for example, obliquely upward with respect to the inspection surface of the wafer 101 which is a semiconductor wafer. The light reflected and scattered from the foreign matter 1401 existing on the wafer 101 is scattered not only in the front direction but also in the lateral direction. Light 1402 scattered forward with respect to incident light is received by a light receiving element (photomal) 1411, and light 1403 and 1404 scattered on the side surfaces are detected by light receiving elements 1412 and 1413, respectively.

FIG. 15 is a diagram illustrating a data format of defect detection data of the inspection apparatus according to the embodiment. The defect data of the embodiment has a defect R coordinate 1501, a defect θ coordinate 1502, a defect Δθ coordinate 1503, and defect size data 1504. Here, the defect R coordinate 1501 is position information in the radial direction of the wafer 101 and is a value generated from a position detection encoder (linear scale) of the radial direction feed motor of the wafer transfer mechanism. Further, the defect θ coordinate 1502 is information representing the circumferential position of the rotary stage (spindle) on which the wafer 101 is mounted, and is generally generated from a rotary encoder attached to the spindle motor shaft. The signals generated from this rotary encoder are the θ encoder pulse 304 and the θ clock 810 used in the description of the embodiment. The defect size data 1504 and the defect Δθ coordinate 1503 are the maximum value data and the maximum position coordinate Δθ of the method described in the embodiment.

  In an embodiment of the present invention, a rotating stage for rotating a semiconductor wafer, a moving stage for moving the semiconductor wafer in a radial direction, a laser light source for irradiating a laser beam on the wafer, and scattering / diffraction / reflection due to foreign matters or defects on the wafer. A light detection circuit that detects the converted light and converts it into an electrical signal, an A / D conversion circuit that converts the detection signal into digital data, an angle coordinate detection circuit that detects the position of the rotary stage, that is, angle coordinate information, A A foreign substance / defect coordinate calculation circuit that calculates the size of a foreign substance or a defect from digital data that has been / D-converted and calculates a position coordinate value based on information from an angle coordinate detection circuit, and has a constant rotational angular velocity of the wafer In the foreign matter / defect detection method for detecting foreign matter or defects existing on the wafer surface or in the vicinity of the surface while scanning by the above-mentioned method, the A / D conversion circuit is The electrical signal is sampled at a sampling time interval, is one unit time of the sampling time interval, it is desirable to configure such that the angle information from the angle coordinate detection circuit is smaller than the increment (one unit) time to be updated.

  The A / D converter circuit samples the electrical signal at a constant sampling time interval, and updates the angle information from the angle coordinate detection circuit based on the sampled data every unit (one unit) time. It is desirable to provide a circuit for obtaining the maximum value of the sampled A / D conversion data.

  Furthermore, a circuit for measuring and recording the time when the angle information from the angle coordinate detection circuit is updated, and a circuit for measuring and recording the time when the A / D conversion circuit samples the maximum value of the electrical signal are provided. Is desirable.

  The maximum value of the electric signal is sampled after the angle information is updated by the time when the angle information is updated and the time when the A / D conversion means samples the maximum value of the electric signal. It is desirable to provide a circuit for calculating the time until the time.

  Further, the A / D conversion means samples the electric signal, the time when the angle information from the angle coordinate detection means is updated, and the time when the A / D conversion means samples the maximum value of the electric signal. It is desirable that the measurement is performed based on the sampling time interval.

  Also, the position coordinate value of the foreign object or defect on the surface of the object to be inspected, the latest angular coordinate information from the angular coordinate detection means at the time when the foreign object or defect is detected, and the generation of the latest angular coordinate information The digital data corresponding to the time when the foreign matter or defect is detected is calculated using both the elapsed time information until the time when the maximum data is sampled in the A / D conversion means. Therefore, it is desirable that the position coordinate value of the foreign matter or the defect is obtained with a good angular resolution.

  Further, an inspection object moving stage in which main scanning is rotational movement and sub-scanning is translational movement, a laser light source, and an illumination spot having a predetermined size on the surface of the inspection object is emitted from the laser light source. Irradiating light, a scattering / diffraction / reflected light detection circuit for detecting light that is scattered, diffracted and reflected at the illumination spot and converting it into an electrical signal, and converting the electrical signal into digital data An A / D conversion circuit, an angle coordinate detection circuit for detecting angle coordinate information in the main scanning direction of the position where the illumination spot is irradiated on the surface of the object to be inspected, and the size of the foreign matter or defect from the digital data Particle size calculation circuit that calculates the position coordinate value of the foreign object or defect on the surface of the object to be inspected based on information from the particle size calculation circuit to calculate and the angle coordinate detection means A foreign matter / defect detection method for detecting foreign matter or defects existing on or near the surface of the inspected object while main-scanning the inspected object at a substantially constant rotational angular velocity. The converting means samples the electrical signal at a substantially constant sampling time interval, and the sampling time interval is such that one point on the object to be inspected is also at the outer periphery on the object to be inspected rotated by the main scanning. It is desirable that at least two or more samplings be performed within a time passing through the illumination spot.

  More preferably, the sampling time interval is at least 5 within the time when one point on the object to be inspected passes through the illumination spot even at the outer peripheral part on the object to be inspected rotated by the main scanning. It is desirable that the sampling is performed at a point or more.

  The particle size calculation circuit may be configured to calculate the size of a foreign substance or a defect using a result obtained by adding a plurality of digital data sampled from the electrical signal within a time for passing through the illumination spot. desirable.

  Further, the particle size calculation circuit is a result obtained by subjecting the size of the foreign matter or defect to digital filtering on a plurality of digital data sampled from the electrical signal within the time of passing through the illumination spot. It is desirable to make a calculation using

  The digital filter is preferably a convolution filter.

  It is desirable that the digital filter is configured to have a frequency band limiting effect (function) effective in reducing shot noise included in the electrical signal.

  In addition, the particle size calculation circuit calculates the size of the foreign matter and the defect with respect to a plurality of digital data obtained by sampling the electrical signal by the A / D conversion circuit within the time required to pass through the illumination spot. A circuit that detects that the input signal of / D conversion is saturated (the signal amplitude is greater than or equal to a predetermined value), a circuit that counts the number of signal amplitudes greater than or equal to the predetermined value, and the signal amplitude is constant A circuit that detects two or more values in the base portion from a smaller signal, a result of counting the number of times that the signal amplitude is equal to or greater than a predetermined value, and two or more amplitudes in the base portion where the signal amplitude is smaller than a certain value It is desirable to use a value result to calculate the maximum amplitude value and the position coordinates of the maximum amplitude from the input signal in which the input signal is saturated (the signal amplitude is a certain value or more).

  Further, since it becomes possible to A / D convert a small defect signal at a frequency larger than the resolution of the rotation angle detector (encoder) of the rotary stage and detect the defect position Δθ in the period of the angle detection signal, high accuracy. The defect detection position information can be obtained.

  In addition, since the waveform is collected by oversampling at a high frequency, even when the detection signal is partially saturated, the maximum (peak) value can be fitted using a non-saturated signal, and the signal is saturated. Even when a large defect is input, the defect size and the defect position can be accurately detected.

  Furthermore, since the frequency for A / D conversion is made larger than the frequency of the angle detection signal, signal processing such as filtering can be performed on the A / D conversion data inside the angle detector, so that noise ( Noise) can be removed and detected with high accuracy.

The figure which showed the rotation angle of the semiconductor wafer in the unit time in the Example of this invention. The figure which shows the waveform of the defect detection signal of the outer peripheral part in an Example of this invention, and an inner peripheral part. The figure which showed the method of signal sampling of this invention in the inner periphery in the Example of this invention. The figure which showed the signal sampling method of this invention in the outer periphery in the Example of this invention. The figure which shows one Example outline | summary of (DELTA) (theta) calculation method of this invention. The figure which shows another Example of (DELTA) (theta) calculation method of this invention. The figure which shows the waveform which the scattered light intensity signal in the Example of this invention is saturated. The figure which shows the point when A / D-converting the saturation signal in the Example of this invention with the sampling signal for A / D conversion. The figure which shows one Example of (DELTA) (theta) detection circuit of this invention. The figure which shows one Example of the fitting circuit of the saturation signal of this invention. The figure which shows another one Example of this invention. The figure which shows another Example of (DELTA) (theta) detection circuit of this invention. The figure which shows the whole structure of the apparatus using the foreign material / defect detection method in the Example of this invention. The figure which shows the structure of the signal processing part of the semiconductor inspection apparatus in the Example of this invention. The figure which showed the positional relationship of the wafer in the Example of this invention, a light source, and a light receiving element. It is a figure which shows the data format of the defect detection data of the inspection apparatus in the Example of this invention.

Explanation of symbols

DESCRIPTION OF SYMBOLS 101 ... Wafer, 202 ... Outer peripheral foreign matter detection waveform, 203 ... Inner peripheral foreign matter detection waveform, 204, 301 ... A / D conversion sampling signal, 401 ... Maximum value of scattered light signal, 402 ... Measurement result of pulse interval of θ encoder signal 403 ... Peak detection position time from the θ encoder,
404 ... A / D conversion circuit sampling frequency, 602 ... scattered light peak value when not saturated, 801 ... scattered light signal, 802 ... scattered light maximum value within the θ pulse period, 803 ... scattered light within the θ pulse period Δθ of maximum value, 805... Amplification circuit, 806... A / D conversion circuit, 808, 903... Sampling signal generation circuit, 809... Maximum position detection circuit, 812 ... sampling clock, 901. Detection circuit, 904... Maximum position detection circuit, 1001... Digital filter circuit, 1002... Addition circuit, 1204 .. Chuck, 1200 .. Foreign matter or defect, 1202 ... Light source of illumination light, 1203. Stage, 1208 ... Z stage.

Claims (28)

An inspection object moving stage in which main scanning is rotational movement at a constant rotational angular speed and sub-scanning is translational movement, a laser light source, and a laser beam emitted from the laser light source having a predetermined size on the surface of the inspection object Illuminating means for irradiating as an illumination spot; scattering / diffraction / reflected light detecting means for detecting light that is scattered, diffracted, and reflected at the illumination spot and converting the light into an electrical signal; and A / D conversion means for converting to data, angle coordinate detection means for detecting angle coordinate information in the main scanning direction of the position irradiated with the illumination spot on the surface of the object to be inspected, and foreign matter or defect from the digital data The position coordinate value on the surface of the object to be inspected is calculated based on the information from the particle size calculating means for calculating the size of the object and the angle coordinate detecting means. And a goods-defect coordinate calculating means,
The A / D conversion means samples the electrical signal at a constant sampling time interval, and one unit time of the sampling time interval is smaller than the update time of angle information from the angle coordinate detection means. Foreign matter / defect detection device. In claim 1,
The A / D conversion means samples the electrical signal at a constant sampling time interval, and based on the sampled data, the angle information from the angle coordinate detection means is updated every unit (one unit) time, A foreign matter / defect detection apparatus comprising means for obtaining a maximum value of sampled A / D conversion data. In claim 1 or 2,
It has means for measuring and recording the time when the angle information from the angle coordinate detection means is updated, and means for measuring and recording the time when the A / D conversion means samples the maximum value of the electric signal. Foreign matter / defect inspection equipment. In any of claims 1, 2 and 3,
From the time when the angle information is updated and the time when the A / D conversion means samples the maximum value of the electric signal, until the time when the maximum value of the electric signal is sampled after the angle information is updated A foreign matter / defect inspection apparatus comprising means for calculating In claim 3,
The time when the angle information from the angle coordinate detection means is updated and the time when the A / D conversion means samples the maximum value of the electric signal, and the A / D conversion means samples the electric signal. A foreign object / defect detection apparatus characterized by measuring based on a sampling time interval. In any of claims 1, 2 and 3,
The position coordinate value of the foreign object or defect on the surface of the object to be inspected, the latest angle coordinate information from the angle coordinate detection means at the time when the foreign object or defect is detected, and the time when the latest angle coordinate information is generated The digital data corresponding to the point in time when the foreign matter or defect is detected is calculated using both of the elapsed time information until the time when the maximum data is sampled in the A / D conversion means. Foreign matter / defect detection device. Inspected object moving stage in which main scanning is rotational movement and sub-scanning is translational movement, laser light source, and a laser beam emitted from the laser light source is irradiated to a predetermined size illumination spot on the surface of the inspected object Illuminating means for detecting, scattered / diffracted / reflected light detecting means for detecting light scattered / diffracted / reflected at the illumination spot and converting it into an electrical signal, and A for converting the electrical signal into digital data / D conversion means, angle coordinate detection means for detecting angle coordinate information in the main scanning direction of the position where the illumination spot is irradiated on the surface of the object to be inspected, and the size of the foreign matter or defect is calculated from the digital data Particle size calculating means for calculating the position coordinate value of the foreign object or defect on the surface of the object to be inspected based on information from the angle coordinate detecting means; Have,
In the foreign matter / defect detection apparatus for detecting foreign matter and defects existing on the surface of the inspected object or in the vicinity of the surface while main-scanning the inspected object at a substantially constant rotational angular velocity,
The A / D conversion means samples the electrical signal at a substantially constant sampling time interval, and the sampling time interval is also on the outer periphery of the inspected object rotated by the main scanning. A foreign matter / defect detection apparatus, wherein at least two points are sampled within a time during which one point passes through the illumination spot. In claim 7,
As for the sampling time interval, at least 5 points or more are sampled within a time when one point on the inspection object passes through the illumination spot even in the outer peripheral portion on the inspection object rotated by the main scanning. Foreign matter / defect detection device characterized by the above. In claim 7 or 8,
The particle size calculating means calculates the size of a foreign object or a defect using a result obtained by adding a plurality of digital data sampled from the electrical signal within a time for passing through the illumination spot.・ Defect detection equipment. In claim 7,
The particle size calculation means uses the result obtained by subjecting a plurality of digital data sampled from the electrical signal within a time for passing through the illumination spot to digital filter processing for the size of the foreign matter or defect. A foreign object / defect detection device characterized by In claim 10,
The foreign matter / defect detection apparatus, wherein the digital filter is a convolution filter. In claim 10 or 11,
The foreign matter / defect detection device, wherein the digital filter has a frequency band limiting function. In claim 7,
The particle size calculating means converts the size of the foreign matter and the defect, and the A / D conversion for the plurality of digital data sampled by the A / D conversion means within the time for passing through the illumination spot. Means for detecting that the signal amplitude of the input signal is greater than or equal to a predetermined value, means for counting the number of signal amplitudes greater than or equal to the predetermined value, and two points from the signal whose signal amplitude is smaller than a certain value to the base portion Using the means for detecting the above values, the result of counting the number of times that the signal amplitude is greater than or equal to a predetermined value, and the amplitude value result of two or more points in the base portion where the signal amplitude is smaller than a certain value, the signal amplitude is A foreign matter / defect detection device, wherein a maximum amplitude value and a position coordinate of a maximum amplitude are calculated from the input signal that is equal to or greater than a certain value. Inspected object moving stage in which main scanning is rotational movement and sub-scanning is translational movement, laser light source, and a laser beam emitted from the laser light source is irradiated to a predetermined size illumination spot on the surface of the inspected object Illuminating means for detecting, scattered / diffracted / reflected light detecting means for detecting light scattered / diffracted / reflected at the illumination spot and converting it into an electrical signal, and A for converting the electrical signal into digital data / D conversion means, angle coordinate detection means for detecting angle coordinate information in the main scanning direction of the position where the illumination spot is irradiated on the surface of the object to be inspected, and the size of the foreign matter or defect is calculated from the digital data Particle size calculating means for calculating the position coordinate value of the foreign object or defect on the surface of the object to be inspected based on information from the angle coordinate detecting means; Have,
In the foreign matter / defect detection apparatus for detecting foreign matter and defects existing on or near the surface of the inspected object while main-scanning the inspected object at a constant rotational angular velocity,
The A / D conversion means is configured such that one unit time of the sampling time interval is smaller than an angle at which the object to be inspected is rotated, and an update time of angle information from the angle coordinate detection means. And obtaining means for obtaining the maximum value of the sampled A / D conversion data at every step of updating the angle information from the angle coordinate detection means, and sampling the maximum value of the electrical signal. Means for measuring and recording the measured time, and the means for obtaining the maximum value is calculated using a result obtained by subjecting a plurality of digital data sampled from the electrical signal to digital filter processing. Foreign matter / defect detection device characterized by Irradiate the surface of the object to be inspected with laser light, detect at least scattered light and convert it into an electrical signal, convert the electrical signal into digital data, and inspect the object to be inspected at the position irradiated with the laser light. Detecting angle information, calculating the size of the foreign matter or defect from the digital data, and calculating the position of the foreign matter or defect on the surface of the inspected object based on the angle information,
In the conversion to the digital data, the electric signal is sampled at a constant sampling time interval, and the sampling time interval is smaller than the update interval of the angle information. In claim 15,
In the conversion, the electric signal is sampled at a constant sampling time interval, and based on the sampled data, a maximum value of the sampled data is obtained for each update interval of the angle information. . In claim 15 or 16,
A time for updating the angle information is recorded, and a time for sampling the maximum value of the electric signal is recorded. In any of claims 15, 16 and 17,
Based on the time when the angle information is updated and the time when the maximum value of the electrical signal is sampled, the time from when the angle information is updated until the maximum value of the electrical signal is sampled is calculated. Foreign matter / defect inspection method. In claim 17,
A foreign matter / defect detection method, wherein the time when the angle information is updated and the time when the maximum value of the electric signal is sampled are measured with reference to the sampling time interval for sampling the electric signal. In any of claims 15, 16 and 17,
The position coordinate value of the foreign object or defect on the surface of the object to be inspected, the latest angle coordinate information from the angle coordinate detection means at the time when the foreign object or defect is detected, and the time when the latest angle coordinate information is generated The digital data corresponding to the point in time when the foreign matter or defect is detected is calculated using both of the elapsed time information until the time when the maximum data is sampled in the A / D conversion means. Foreign matter / defect detection method. Laser light is irradiated on the surface of the object to be inspected, at least scattered light is detected and converted into an electrical signal, the electrical signal is converted into digital data, and the inspection object at the position where the laser light is irradiated Detecting angle information, calculating the size of the foreign matter or defect from the digital data, and calculating the position of the foreign matter or defect on the surface of the inspected object based on the angle information,
In the conversion to the digital data, the electrical signal is sampled at a substantially constant sampling time interval, and the sampling time interval is at least two points within the time when one point on the object to be inspected is irradiated with the laser light. A foreign matter / defect detection method characterized by performing the above sampling. In claim 21,
The sampling time interval is such that at least 5 points are sampled within a time period during which one point on the object to be inspected passes through the illumination spot on the outer periphery of the object to be inspected rotated by the main scanning. A foreign matter / defect detection method characterized by being configured as described above. In claim 21 or 22,
The particle size calculation means is configured to calculate the size of the foreign matter or defect using a result obtained by adding a plurality of digital data sampled from the electrical signal within the time of passing through the illumination spot. Characteristic foreign object / defect detection method. In claim 21,
The particle size calculation means uses the result obtained by subjecting a plurality of digital data sampled from the electrical signal within a time for passing through the illumination spot to digital filter processing for the size of the foreign matter or defect. A foreign matter / defect detection method characterized by being configured to calculate the   25. The foreign object / defect detection method according to claim 24, wherein the digital filter is a convolution filter. In claim 24 or 25,
A foreign matter / defect detection method, wherein the digital filter limits a frequency band so as to reduce shot noise included in the electrical signal. In claim 21,
The particle size calculating means converts the size of the foreign matter and the defect, and the A / D conversion for the plurality of digital data sampled by the A / D conversion means within the time for passing through the illumination spot. Means for detecting that the signal amplitude of the input signal is greater than or equal to a predetermined value, means for counting the number of signal amplitudes greater than or equal to the predetermined value, and two points from the signal whose signal amplitude is smaller than a certain value to the base portion Using the means for detecting the above values, the result of counting the number of times that the signal amplitude is greater than or equal to a predetermined value, and the amplitude value result of two or more points in the base portion where the signal amplitude is smaller than a certain value, the signal amplitude is A foreign matter / defect detection method, wherein a maximum amplitude value and a position coordinate of the maximum amplitude are calculated from the input signal that is equal to or greater than a certain value. Laser light is irradiated on the surface of the object to be inspected, at least scattered light is detected and converted into an electrical signal, the electrical signal is converted into digital data, and the inspection object at the position where the laser light is irradiated Detecting angle information, calculating the size of the foreign matter or defect from the digital data, and calculating the position of the foreign matter or defect on the surface of the inspected object based on the angle information,
In the conversion to the digital data, the electric signal is sampled at a substantially constant sampling time interval, and one unit time of the sampling time interval is an angle of rotation of the object to be inspected, angle information from the angle coordinate detecting means, It is configured to be smaller than one unit time to be updated, and based on the sampled data, the maximum value of the sampled digital data is obtained for each unit time in which the angle information is updated, and the maximum value is sampled. A foreign matter / defect detection method characterized by measuring the time when

Images