JP4491391B2 - Defect inspection apparatus and defect inspection method - Google Patents

Description

  The present invention relates to a pattern inspection, a foreign matter inspection apparatus, and an inspection method for detecting a defect (such as a short circuit or disconnection) of a pattern formed on an inspection object such as a semiconductor wafer or a foreign matter on the surface of the inspection object, and particularly as a primary detector. The present invention relates to an inspection apparatus and an inspection method using an original or two-dimensional image sensor.

Defect inspection equipment that detects defects (shorts, breaks, etc.) in patterns formed on semiconductor wafers, etc. and foreign matter on the surface of the object to be inspected require higher resolution and improve throughput by miniaturizing semiconductor patterns. It has been demanded. In order to satisfy this requirement, it is essential to improve the sensitivity of the detector. With respect to this technology, conventionally, as described in Patent Document 1, a method of increasing the amount of accumulated charges by using a TDI (Time Delay & Integration) image sensor, which is a time delay integration type image sensor, as a detector, or Patent Document 2 As described above, a method of improving the quantum efficiency of photoelectric conversion and increasing the amount of detected light by using a back-illuminated TDI image sensor has been proposed.

JP 2004-301847 A JP 2001-194323 A

  In an inspection apparatus using a photoelectric conversion type image sensor as a detector, in order to improve the sensitivity of the detector several times or more, one pixel is used regardless of whether the light receiving structure of the image sensor is a front side illumination type or a back side illumination type. Since the hit sensitivity does not change, there is no choice but to increase the pixel size or increase the number of TDI stages. In this case, in order to inspect the inspection object without reducing the apparatus throughput, the size of the image sensor increases in proportion to the sensitivity improvement, so that the yield decreases due to the increase in the frequency of pixel defects due to the manufacture of large area image sensors. There were problems such as technical risks such as lens development due to significant changes in illumination and detection optical system, and development period and cost increase.

  In order to solve the above problems, an object of the present invention is to provide a method for improving the sensitivity of an inspection apparatus only by changing a detector in a defect inspection apparatus equipped with dark field illumination / reflected scattered light detection. .

  In order to achieve the above object, the present invention has a detector having an electron multiplying means at the front stage or the rear stage of a one-dimensional or two-dimensional photoelectric conversion image sensor, and illuminates illumination light from an oblique direction to be inspected. Illumination means and detection means for receiving reflected and scattered light from the inspection object obliquely or with a detector arranged above, and optically by scanning the stage on which the inspection object is mounted or scanning the detector A defect inspection method characterized by acquiring an image and inspecting a defect to be inspected by image processing or the like.

  According to the present invention, there is provided an illuminating means for illuminating illumination light from an oblique direction of the inspection object and a detecting means for receiving reflected and scattered light from the inspection object obliquely or upwardly, and mounted on the inspection object. In the inspection apparatus for dark field illumination / reflected scattered light detection that optically acquires an image by scanning the stage or scanning a detector and inspects a defect to be inspected by image processing or the like, one-dimensional or By using a detector having an electron multiplying means at the front stage or the rear stage of a two-dimensional photoelectric conversion image sensor, the SN ratio is greatly improved even if the number of incident photons is small. A foreign object can be detected, and a highly sensitive defect inspection apparatus can be provided.

  Hereinafter, the present invention will be described with reference to the drawings. FIG. 1 is a diagram showing an embodiment of the defect inspection apparatus of the present invention.

The stage 6 is composed of X, Y, Z, and θ (rotation) stages, on which a semiconductor wafer (sample) 7 that is an example of a pattern to be inspected is placed. The illumination light source 1 is composed of, for example, an ultraviolet or far ultraviolet laser light source having a wavelength of 266 nm or a wavelength of 355 nm, and is a light source that illuminates the sample 7. The ultraviolet laser light source is composed of a device that generates a third harmonic (355 nm) or a fourth harmonic (266 nm) of the fundamental wave by converting the wavelength of a solid YAG laser with a nonlinear optical crystal or the like. A laser light source having a wavelength of 193 nm, a wavelength of 195 nm, a wavelength of 248 nm, or the like may be used. Furthermore, if it exists as a laser light source, a wavelength of 100 nm or less may be used, and the resolution will be further improved. The laser oscillation mode may be continuous oscillation or pulse oscillation, but continuous oscillation is preferable because the stage is continuously run to detect an image from the object 7. The stage 6 can be controlled in the X, Y, Z, and θ directions by the control CPU 14 by a method not shown. Illumination light from the light source 1 is controlled to a light amount necessary for inspection by an ND filter 2 that restricts the light amount. The ND filter 2 can be driven by a command from the ND filter control circuit 3 by a method not shown. The beam expander 4 expands the light beam from the light source 1, and the expanded light beam sets the illumination range on the sample 7 mounted on the stage 6 by the illumination optical system 5, and irradiates the sample 7 obliquely. In addition, dark field illumination is applied. Reflected and scattered light from the sample 7 is detected by the detector 12 through the objective lens 8, the spatial filter 9, the imaging lens 11, and the like, and binarized by the image processing unit 13 to detect defects. . The spatial filter 9 can be driven by a command from the spatial filter control circuit 10 by a method not shown, and can block diffracted light from a repetitive pattern on the sample 7. The image processing result and the like are displayed on the display unit 16, and the control CPU 14 controls the information input from the input unit 15 and the data and information of the image processing unit 13, the detector 12, and the stage 6. Here, the detector 12 uses an electron multiplication type image sensor having electron multiplication means at the front stage or the rear stage of the image sensor. FIG. 2 is a diagram showing an embodiment of the configuration of the detector in the defect inspection apparatus of the present invention. As shown in FIG. 2A, the inside of the sensor package 100 is in a vacuum 105 state, and a back-illuminated CCD (Charge Coupled Device) 104 that is an image sensor is accommodated. Photons 101 incident on the detector 12 are converted to electrons 103 by the photocathode 102. The electrons 103 are accelerated by applying a voltage, and are driven into the back-illuminated CCD. At this time, secondary electrons are generated and the charge is amplified. The amplified charge is converted into a voltage and output from the output pin 107. Here, the image sensor may be a front-illuminated CCD, a front-illuminated TDI, or a back-illuminated TDI, may have an anti-blooming function, and may be a multi-tap type. The sensor structure shown in FIG.
This is a structure of an image sensor called (Electron Bombarded Charge Coupled Device), and has electron multiplying means in the front stage of the image sensor. The gain characteristic of EBCCD has a proportional relationship between the applied voltage and the electron multiplication gain as shown in FIG. FIG. 3 is a diagram showing another embodiment of the configuration of the detector in the defect inspection apparatus of the present invention. As shown in FIG. 6A, photons received by a sensor array 401 including a plurality of sensor pixels 402 are converted into electric charges, and electric charges are accumulated while a shift gate 403 that opens and closes at a period of electric charge accumulation time is closed. When opened, the charge is transferred to the shift register 404 corresponding to each sensor pixel 402. The charges transferred to the shift register 404 are sequentially transferred between the shift registers 404 while the shift gate 403 is closed, and are sent to the electron multiplication register 405. There are a plurality of electron multiplication registers 405, and a voltage is applied by the electron multiplication register control unit 406 at the time of charge transfer in each electron multiplication register 405 to generate secondary electrons inside the electron multiplication register 405. After multiplying and repeating the electron multiplication for transferring the electron-multiplied charge to the next electron multiplication register 405, the charge is sent to the output amplifier 407. Here, the image sensor may be a back-illuminated CCD, a front-illuminated CCD, a front-illuminated TDI, a back-illuminated TDI, may have an anti-blooming function, and may be a multi-tap type. good. The sensor structure shown in FIG. 1A is an image sensor structure generally called an EMCD (Electron Multiplying Charge Coupled Device), and has an electron multiplying means at the subsequent stage of the image sensor. The gain characteristic of the EMCCD has a substantially proportional relationship between the applied voltage and the logarithmic value of the electron multiplication gain as shown in FIG. By using the detector 12 having the electron multiplying means as described above, the defect detection sensitivity can be improved without changing the illumination optical system and the detection optical system including the light source. Therefore, it is possible to provide an inspection apparatus that can obtain a good detection result with a small amount and high detection sensitivity.

FIG. 4 is a diagram showing the relationship between the output signal (S) and noise (N) of the CCD sensor and EBCCD sensor. As conditions for calculating the output signal and noise, the quantum efficiency (QE) of the CCD sensor is 50%, the readout noise of the CCD sensor is 80e, the quantum efficiency (η) of the photoelectric conversion surface in the previous stage of EBCCD is 20%, The electron multiplication gain (G _EB ) was set to 1000 times, the dark current (N _D ) of the photoelectric conversion surface in front of EBCCD was set to 300e, and the number of saturated electrons of the CCD sensor was set to 100000e. Further, the noise increase ratio (NF) with respect to the electron multiplication gain, the output signal and noise of the CCD sensor, and the output signal and noise of the EBCCD were calculated by the equations (1) to (5).

The horizontal axis of the graph shown in the figure is the number of incident photons (photons) (N _S ), the vertical axis is the number of output signals and noise output electrons from each sensor, and the right side is the output signal and noise of each sensor. This is the S / N ratio (S / N). As shown in the figure, the output signal of EBCCD is much larger than the output signal of CCD due to electron multiplication, but it is a part of noise, and shot noise generated at the square root of the output signal when electrons are generated. Is also a ratio between the output signal and noise, and the signal-to-noise ratio, which is greatly involved in defect detection, does not differ greatly when the number of incident photons is small, and is highly sensitive when the number of incident photons is large. Since the EBCCD reaches the saturation level earlier, the CCD has a better SN ratio. For this reason, even if an EBCCD having an electron multiplication function and high sensitivity is used as a detector, it seems that the effect of improving the detection sensitivity of the defect inspection apparatus is thin. However, this SN ratio varies greatly depending on the inspection method. Hereinafter, the relationship between the detection method and the SN ratio will be described with reference to FIGS. FIG. 5 is a diagram showing an embodiment of a defect detection method by bright field illumination / reflected light detection. When the inspection target is, for example, a semiconductor wafer, the pattern influence can be reduced by comparing the inspection target chip with the adjacent chip in pattern inspection for detecting pattern defects (short circuit, disconnection, etc.) and foreign matters. Remove and detect pattern defects and foreign matter. At this time, in the defect detection method based on bright field illumination / reflected light detection, basically a lot of reflected light is received by the detector, so that an overall bright output signal can be obtained as shown in FIG. This figure is an output signal when a defect is present, the horizontal axis is the scan position on the inspection target when the stage carrying the inspection target is scanned, and the vertical axis is the output signal of the detector. In order to detect a defect whose output signal is different by δ with respect to the overall brightness S ₁ shown in the figure without being influenced by the pattern and discriminate it from shot noise (the square root of S ₁ ), FIG. Compare (differ) with the output signal of the adjacent chip as shown. The signal after the difference is shown in FIG. As shown in the figure, in order to discriminate the defect output (S) and the noise (N) from the signal after the difference and detect only the defect, it is necessary that S> N. Indicates that the conditions shown are necessary.

  That is, the increase amount of the output signal due to the defect needs to be larger than the shot noise when the output signal is high (the number of incident photons is large), which is the same as the output signal (S) and noise (N ) Means that the S / N ratio (dynamic range) must be large.

On the other hand, when the inspection object is, for example, a semiconductor wafer by dark-field illumination / reflected scattered light detection, a pattern filter that detects pattern defects (such as short-circuits and disconnections) and foreign objects, and foreign object inspection are performed using a spatial filter. Since the diffracted light from the upper repetitive pattern is shielded and only the reflected scattered light from the defect or foreign matter is detected by the detector, an overall dark output signal is obtained as shown in FIG. This figure is an output signal when a defect is present, the horizontal axis is the scan position on the inspection target when the stage carrying the inspection target is scanned, and the vertical axis is the output signal of the detector. In order to detect a defect whose output signal is different by δ with respect to the overall brightness S ₁ shown in the figure by distinguishing it from shot noise (the square root of S ₁ ), the adjacent chip as shown in FIG. Compare (differ) with the output signal. The signal after the difference is shown in FIG. As shown in the figure, in order to discriminate the defect output (S) and the noise (N) from the signal after the difference and detect only the defect, it is necessary that S> N. Indicates that the conditions shown are necessary.

  That is, the increase amount of the output signal due to the defect needs to be larger than the shot noise when the output signal is low (the number of incident photons is small), which is equal to the output signal (S) and noise (N ) Means that the S / N ratio (dynamic range) need not be large.

FIG. 7 is a diagram showing an example of the ratio (SN ratio) between the output signal (S) and noise (N) of the detector in the case of bright field illumination / reflected light detection and dark field illumination / reflected scattered light detection. . A CCD sensor and an EBCCD sensor are used as detectors, and the conditions for calculating the output signal and noise are as follows: the quantum efficiency (QE) of the CCD sensor is 50%, the readout noise of the CCD sensor is 80e, and the preceding stage of EBCCD. The quantum efficiency (η) of the photoelectric conversion surface is 20%, the electron multiplication gain (G _EB ) is 1000 times,
The dark current (N _D ) on the photoelectric conversion surface in the previous stage of EBCCD was 300e, and the number of saturated electrons of the CCD sensor was 100,000e. Here, the SN ratio in the bright field illumination / reflected light detection is (Equation 1) to
The SN ratio (S / N) is calculated from the output signal (S) and noise (N) of the CCD and EBCCD calculated by the equation (5), and the output signal of the SN ratio in the dark field illumination / reflected scattered light detection. (S) is calculated by the equations (2) and (4), and the noise (N) is the same as the readout noise of the CCD sensor because the light shot noise at the dark level can be ignored, and the SN ratio is calculated. From the figure, in bright field illumination / reflected light detection, there is no significant difference in the signal-to-noise ratio between the case where CCD is used as the detector (CCD (S / N)) and the case where EBCCD is used (EBCCD (S / N)). However, the sensitivity cannot be significantly improved. On the other hand, in the case of using a CCD as a detector (CCD (S / N _R )), the SN ratio in bright field illumination / reflected light detection is not much different in dark field illumination / reflected scattered light detection, but EBCCD is used. (EBCCD (S / N _R )), the SN ratio is high even if the number of incident photons is small, and the SN ratio is remarkably improved. For this reason, it becomes possible to detect minute defects and foreign matters having a small number of incident photons, and it is possible to provide a highly sensitive defect inspection apparatus.

FIG. 8 is a diagram showing an embodiment of a method for calculating the electron multiplication gain of the detector in the defect inspection apparatus of the present invention. This figure shows SN characteristics when EBCCD is used as a detector. The horizontal axis represents the electron multiplication gain, and the left side of the vertical axis represents the output signal (S) and noise (N) when the incident photon (N _S ) is 300e. ) And the right side is the SN ratio (S / N), which is the ratio of the output signal (S) and noise (N). As conditions for calculating the output signal and noise, the incident photon (N _s ) is 300e, the quantum efficiency (QE) of the CCD sensor is 50%, the readout noise of the CCD sensor is 80e, the photoelectric conversion surface of the front stage of the EBCCD is The quantum efficiency (η) is 20%, the electron multiplication gain (G _EB ) is 1 to 1000 times, the dark current (N _D ) of the photoelectric conversion surface in front of EBCCD is 300e, and the saturation electron number of the CCD sensor is 100000e. . Further, the noise increase ratio (NF) with respect to the electron multiplication gain, the output signal, and the noise were calculated by the equations (Equation 1), (Equation 4), and (Equation 5). As shown in the figure, for example, when the SN ratio necessary for detecting a defect having an incident photon number of 300e is 400, the necessary electron multiplication gain is calculated by the equation (8). To do.

After calculating the necessary electron multiplication gain, it is converted into an applied voltage, and the applied voltage of EBCCD is adjusted to the converted voltage. Thereby, the defect inspection apparatus which can obtain the detection sensitivity which can detect the defect of a desired defect size can be provided. FIG. 9 is a diagram showing an embodiment of the configuration of the defect inspection apparatus of the present invention. The stage 6 is composed of X, Y, Z, and θ (rotation) stages, on which a semiconductor wafer (sample) 7 that is an example of a pattern to be inspected is placed. The illumination light source 1 is composed of, for example, an ultraviolet or far ultraviolet laser light source having a wavelength of 266 nm or a wavelength of 355 nm, and is a light source that illuminates the sample 7. The ultraviolet laser light source is composed of a device that generates a third harmonic (355 nm) or a fourth harmonic (266 nm) of a fundamental wave by converting the wavelength of a solid YAG laser with a nonlinear optical crystal or the like. A laser light source having a wavelength of 193 nm, a wavelength of 195 nm, a wavelength of 248 nm, or the like may be used. Furthermore, if it exists as a laser light source, a wavelength of 100 nm or less may be used, and the resolution will be further improved. The laser oscillation mode may be continuous oscillation or pulse oscillation, but continuous oscillation is preferable because the stage is continuously run to detect an image from the object 7. The stage 6 can be controlled in the X, Y, Z, and θ directions by the control CPU 14 by a method not shown. Illumination light from the light source 1 is controlled to a light amount necessary for inspection by an ND filter 2 that restricts the light amount. The ND filter 2 can be driven by a command from the ND filter control circuit 3 by a method not shown. The beam expander 4 expands the light beam from the light source 1, and the expanded light beam sets the illumination range on the sample 7 mounted on the stage 6 by the illumination optical system 5, and irradiates the sample 7 obliquely. In addition, dark field illumination is applied. Reflected and scattered light from the sample 7 passes through an objective lens 8, a spatial filter 9, an imaging lens 11, and the like, and a detector 12 comprising an electron multiplication type image sensor having electron multiplication means at the front stage or the rear stage of the image sensor. The image processing unit 13 performs binarization processing and the like to detect defects. The spatial filter 9 can be driven by a command from the spatial filter control circuit 10 by a method not shown, and can block diffracted light from a repetitive pattern on the sample 7. The image processing result and the like are displayed on the display unit 16, and the control CPU 14 controls the information input from the input unit 15 and the data and information of the image processing unit 13, the detector 12, and the stage 6. When the applied voltage for determining the electron multiplication gain of the detector 12 is set to a desired value, the SN ratio calculation unit 17 calculates an SN ratio for detecting a desired defect, via the control CPU 14. The calculated S / N ratio is sent to the electron multiplication gain control unit 18, and the electron multiplication gain control unit 18 calculates a necessary applied voltage from the S / N ratio to control the applied voltage. Here, the applied voltage calculation function is performed by the control CPU 14, and the applied voltage of the detector 12 may be directly controlled by the control CPU 14, and the SN ratio may be directly calculated by the control CPU 14. Further, the defect output signal (S) and noise (N) at the time of calculating the S / N ratio may be calculated from the number of incident photons as shown in FIG. 8, or measured values in this apparatus may be used. . Further, the result of calculating the SN ratio and the necessary electron multiplication gain or applied voltage for each defect size by any of the methods described above may be stored in advance in a memory or the like in the control CPU 14. It should be noted that parameters such as the calculation of the SN ratio, the necessary electron multiplication gain, and the defect size necessary for calculating the applied voltage may be input via the input unit 15. The image sensor used for the detector 12 may be a back-illuminated CCD, a front-illuminated CCD, a front-illuminated TDI, or a back-illuminated TDI, and may have an anti-blooming function. It may be a mold. With the above configuration, the defect inspection apparatus of the present invention can obtain detection sensitivity capable of detecting a defect having a desired defect size or the number of incident photons, and the image sensor can use a multi-tap detector. Since the operation speed of the detector can be improved, a defect inspection apparatus capable of high-speed inspection can be provided.

It is a figure which shows one Example of the defect inspection apparatus of this invention. It is a figure which shows one Example of a structure of the detector of the defect inspection apparatus of this invention. It is a figure which shows the other Example of a structure of the detector of the defect inspection apparatus of this invention. It is a figure which shows one Example of the signal characteristic of the detector of the defect inspection apparatus of this invention. It is a figure which shows the defect detection conditions of the apparatus which performs bright field illumination and reflected light detection. It is a figure which shows the defect detection conditions of the apparatus which performs the dark field illumination and reflected scattered light detection of this invention. It is a figure which shows one Example of the signal characteristic of the detector by the difference in a detection system. It is a figure which shows one Example of the electron multiplication gain setting method of this invention. It is a figure which shows the other Example of the defect inspection apparatus of this invention.

Explanation of symbols

DESCRIPTION OF SYMBOLS 1 ... Light source, 2 ... ND filter, 3 ... ND filter control circuit, 4 ... Beam expander, 5 ... Illumination optical system, 6 ... Stage, 7 ... Sample, 8 ... Objective lens, 9 ... Spatial filter, 10 ... Spatial filter Control circuit, 11 ... imaging lens, 12 ... detector, 13 ... image processing unit, 14 ... control CPU, 15 ... input unit, 16 ... display unit, 17 ... SN ratio calculation unit, 18 ... electron multiplication gain control unit DESCRIPTION OF SYMBOLS 100 ... Sensor package 101 ... Photon 102 ... Photoelectric surface 103 ... Electron 104 ... Back side illumination type CCD 105 ... Vacuum 106 ... Applied voltage 107 ... Output pin 401 ... Sensor array 402 ... Sensor pixel 403 ... shift gate, 404 ... shift register,
405: Electron multiplication register, 406: Electron multiplication register controller, 407: Output amplifier.

Claims (8)

A detector including a photoelectric conversion image sensor having electron multiplying means;
Illuminating means for illuminating illumination light from an oblique direction to be inspected;
Detecting means for receiving reflected and scattered light from the inspection object obliquely or above with the detector disposed above,
Scanning means for optically acquiring an image with the detector by scanning the inspection object;
The photoelectric conversion image sensor has a photoelectric conversion unit in the preceding stage, and the photoelectric conversion image sensor
And a detector having a voltage application unit that applies a voltage to electrons converted by the photoelectric conversion unit, and a structure that maintains a vacuum state between the photoelectric conversion unit, and
An applied voltage control unit for controlling an applied voltage of the voltage applying unit;
An SN ratio calculation unit for calculating the output signal of the detector, noise, and the ratio of the output signal to the noise from the number of photons incident on the detector and the electron multiplication gain controllable by the applied voltage;
The output signal input by the external input means is compared with the target value of the noise ratio, and the applied voltage control unit is controlled so that the ratio of the output signal and noise is equal to or higher than the target value, and the electron multiplication gain is set. A defect inspection apparatus comprising: an electron multiplication gain control unit for adjustment. A detector including a photoelectric conversion image sensor having electron multiplying means;
Illuminating means for illuminating illumination light from an oblique direction to be inspected;
Detection in which reflected and scattered light from the inspection object is received by the detector arranged obliquely or above
Means,
Scanning means for optically acquiring an image with the detector by scanning the inspection object;
It has a photoelectric conversion part in the front | former stage of the said photoelectric conversion image sensor, it has a structure which maintains between the said photoelectric conversion image sensor and the said photoelectric conversion part in a vacuum state, and it is voltage to the electron converted in the said photoelectric conversion part a detector having a voltage application unit for applying a
A plurality of units arranged in series between the charge transfer unit and the readout register of the photoelectric conversion image sensor
With an electronic multiplication register
A voltage application unit for applying a voltage to the electron multiplier register;
An applied voltage control unit for controlling an applied voltage of the voltage applying unit;
An SN ratio calculating section for calculating the output signal of the detector, the noise, and the ratio of the output signal to the noise from the number of photons incident on the detector and the electron multiplication gain controllable by the applied voltage; and an external input means An electron that compares the output signal input by the target value of the noise ratio and controls the applied voltage control unit to adjust the electron multiplication gain so that the ratio of the output signal and noise is equal to or greater than the target value. A defect inspection apparatus comprising a multiplication gain control unit. A detector including a photoelectric conversion image sensor having electron multiplying means;
Illuminating means for illuminating illumination light from an oblique direction to be inspected;
Detection in which reflected and scattered light from the inspection object is received by the detector arranged obliquely or above
Means,
Scanning means for optically acquiring an image with the detector by scanning the inspection object;
The photoelectric conversion image sensor has a photoelectric conversion unit in the preceding stage, and the photoelectric conversion image sensor
And a detector having a voltage application unit that applies a voltage to electrons converted by the photoelectric conversion unit, and a structure that maintains a vacuum state between the photoelectric conversion unit, and
An applied voltage control unit for controlling an applied voltage of the voltage applying unit;
An SN ratio calculation unit for calculating the ratio of the output signal and the noise from the output signal when the particle having a known size such as the standard particle attached to the inspection object is detected by the detector and the noise at this time;
The output signal input by the external input means is compared with the target value of the noise ratio, and the applied voltage control unit is controlled so that the ratio of the output signal and noise is equal to or higher than the target value, and the electron multiplication gain is set. A defect inspection apparatus comprising: an electron multiplication gain control unit for adjustment. A detector including a photoelectric conversion image sensor having electron multiplying means;
Illuminating means for illuminating illumination light from an oblique direction to be inspected;
Detection in which reflected and scattered light from the inspection object is received by the detector arranged obliquely or above
Means,
Scanning means for optically acquiring an image with the detector by scanning the inspection object;
The photoelectric conversion image sensor has a photoelectric conversion unit in the previous stage, and the photoelectric conversion image sensor
And a detector having a voltage application unit that applies a voltage to electrons converted by the photoelectric conversion unit, and a structure that maintains a vacuum state between the photoelectric conversion unit, and
A plurality of units arranged in series between the charge transfer unit and the readout register of the photoelectric conversion image sensor
The electronic multiplication register
A voltage application unit for applying a voltage to the electron multiplier register;
An applied voltage control unit for controlling an applied voltage of the voltage applying unit;
An SN ratio calculation unit for calculating the ratio of the output signal and the noise from the output signal when the particle having a known size such as the standard particle attached to the inspection object is detected by the detector and the noise at this time;
The output signal input by the external input means is compared with the target value of the noise ratio, and the applied voltage control unit is controlled so that the ratio of the output signal and noise is equal to or higher than the target value, and the electron multiplication gain is set. A defect inspection apparatus comprising: an electron multiplication gain control unit for adjustment. In the defect inspection apparatus in any one of Claims 1-4 ,
The defect inspection apparatus, wherein the photoelectric conversion image sensor is a time delay integration type (TDI) image sensor. 6. The defect inspection apparatus according to claim 5, wherein the time delay integration (TDI) image sensor is an anti-blooming TDI image sensor. TDI type (TDI) image sensor according to claim 5 or 6, wherein the defect inspection apparatus which is a back-illuminated TDI image sensor. The image sensor according to any one of claims 5 to 7 , wherein the image sensor is a multi-tap image sensor that is divided into a plurality of pixels in the pixel direction and that can be read in parallel for each of the divided several pixel units (taps). Defect inspection equipment.

Images