JP2005274156A - Flaw inspection device - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a flaw inspection device capable of easily selecting a proper inspection condition. <P>SOLUTION: The flaw inspection device 1 comprises an illumination means for converting the illumination light from a light source to almost parallel light flux to illuminate a specimen; an image forming means for condensing the reflected light from a specimen 101 in the parallel light flux from the illumination means; an imaging means for taking the image formed by the image forming means; the wavelength selecting means arranged in the light path between the light source and the imaging means to set the center wavelength and wavelength width of the light flux in the light path; a flaw detection part for detecting the flaw of the specimen on the basis of the image taken by the imaging means; an angle setting means for setting the incident angle θi of the illumination light of the illumination means; and an inspection condition setting part for acquiring a plurality of center wavelengths and the reflectivity data of a region having no flaw of the specimen measured at an incident angle and setting the center wavelength at the time of imaging of the image used in the detection of a flaw and the incident angle based on the reflectivity data. <P>COPYRIGHT: (C)2006,JPO&NCIPI

Description

  The present invention relates to a defect inspection apparatus for detecting a defect of an object to be inspected.

  Conventionally, when a substrate such as a semiconductor is manufactured, defects may be formed on the substrate due to various causes. For example, uneven thickness of the resist applied to the substrate surface in the photolithography process and dust adhering to the resist cause defects. Examples of the defect include a defective line width of a periodic pattern on a substrate after etching, a pinhole in the pattern, and the like. In addition, a poor focus in pattern exposure causes a defect in formation of a cross-sectional shape of a resist after development. This defect can cause a defect in a later process.

  For this reason, the wafer and the substrate are subjected to defect detection and defect correction. In order to detect the defect, a defect inspection apparatus is generally used.

  For example, Patent Documents 1 and 2 show the conventional defect inspection apparatus. The conventional defect inspection apparatus of Patent Document 1 will be described below with reference to FIG. 10, and the conventional defect inspection apparatus of Patent Document 2 will be described with reference to FIG.

  The defect inspection apparatus of Patent Document 1 is an apparatus that observes a light beam regularly reflected by the subject 101 such as the wafer or the substrate having at least one periodic pattern and detects a defect on the subject 101. Specifically, the defect inspection apparatus includes a lamp house 5. The lamp house 5 includes a halogen lamp 6, a condenser lens 7, and a heat absorption filter 8. Light from the halogen lamp 6 enters the condenser lens 7 through the heat absorption filter 8 and is converted into a parallel light beam.

  The defect inspection apparatus further includes a rotating holder 9 having a plurality of interference filters and a white light illumination hole. The rotating holder 9 is configured so that any interference filter or the hole can be arranged on the optical path of the light beam that has passed through the condenser lens 7 by rotating. Each interference filter is configured to transmit only a light beam in a predetermined band centered on a predetermined wavelength. For this reason, the rotation holder 9 can change arbitrarily the wavelength width of the light beam which passes by changing an interference filter. In the present specification, in the light flux having a predetermined wavelength width, the center wavelength of the wavelength width is referred to as a center wavelength.

  Further, the defect inspection apparatus includes a condenser lens 10, an optical fiber bundle 11, a diffusion plate 12, a diaphragm 13, a collimator lens 14, and a half mirror 15. The light beam from the light source unit enters the collimator lens 14 via the half mirror 15, is converted into a parallel light beam, and is irradiated on the subject 101.

  The defect inspection apparatus further includes an imaging lens 16 and a CCD 17. The light beam regularly reflected by the subject 101 enters the imaging lens 16 through the collimating lens 14 and the half mirror 15. The imaging lens 16 forms an image of the subject 101 on the CCD 17 with the incident light beam. The CCD 17 captures an image of the subject 101 that has been formed.

  In this defect inspection apparatus, the CCD 17 captures an image of the reflected light of the light beam irradiated on the subject 101 as described above, and observes the film thickness unevenness of the subject 101 as interference fringes. Moreover, this defect inspection apparatus can also capture an image of diffracted light from the subject by tilting the optical axis of the illumination system or imaging system.

  As shown in FIG. 11, the defect inspection apparatus of Patent Document 2 includes an illumination unit 802 having a light source, and a collimator lens 803 that converts a light beam emitted from the illumination unit 802 into a parallel light beam. The illumination unit 802 and the collimator lens 803 illuminate the subject 101 at a predetermined angle θi with respect to the normal of the subject surface.

  The defect inspection apparatus further includes a collimating lens 804, an imaging lens 806, and a line image sensor 807. The collimator lens 804 converges the light beam reflected, diffracted, and scattered by the subject 101 at the inspection angle θd with respect to the normal line of the surface of the subject 101 and makes it incident on the imaging lens 806.

  This defect inspection apparatus moves the subject 101 relative to the defect inspection apparatus in a direction orthogonal to the width direction of the line image sensor 807 (indicated by an arrow AA in FIG. 11). By this movement, the defect inspection apparatus images the entire subject 101 and acquires an inspection image.

  In this inspection image, the defect can be observed by the brightness of light (reflected light) reflected by the subject 101. Specifically, defects such as film thickness unevenness appear as the brightness of the regular reflection light. In addition, defects due to poor focusing are observed as the brightness of the diffracted light. Thus, since the defect is detected by the brightness of specular reflection light and diffracted light, the object 101 is imaged at a plurality of inspection angles θd so that the object 101 can be imaged at the inspection angle θd with the clearest contrast. Take an image.

Then, the plurality of inspection images captured as described above are compared with a reference image that is an image obtained by capturing an image of a subject that does not have a defect, and portions having different luminance between the reference image and the inspection image are defective. Extract as
Japanese Patent Laid-Open No. 7-27709 JP-A-9-61365

  In general, a multilayer pattern is formed on the surface of the subject 101 by a photolithography process, and a resist layer is formed on the upper layer of these patterns. The reflectance of the subject 101 when imaging the subject 101 is governed by the thin film interference of the multilayer film including the resist layer and diffraction according to the periodic pattern. As a result, the luminance of the image of the subject 101 changes in a complicated manner depending on the selection of the imaging wavelength and the incident angle. Therefore, when the defect is detected by the defect inspection apparatus, it is preferable to adjust the center wavelength and the incident angle θi of the light used for the illumination according to the type of the subject. For this reason, the conventional defect inspection apparatus as shown in Patent Documents 1 and 2 is configured to be able to adjust at least one of the center wavelength and the incident angle θi.

  However, in order to detect a defect with high accuracy, it is necessary to appropriately set the center wavelength and the incident angle θi for each type of subject. If this setting is not appropriate, the conventional defect detection apparatus aims to detect defects in the upper resist layer, but also detects lower layer defects at the same time as the uppermost resist layer to be inspected. Problems arise. For example, when there is uneven thickness in the lower layer, the uneven thickness affects the intensity of the regular reflection light and diffracted light of the irradiated light due to thin film interference with the lower layer including the resist layer. In order to suppress defects such as uneven film thickness in the lower layer appearing in the inspection image, it is conceivable to widen the wavelength width of the light used for imaging, but the brightness of the defect in the resist layer is lowered and imaging is difficult. Problem arises.

Therefore, it is necessary to adjust the wavelength width together with the center wavelength of light to an optimum value according to the type of inspection pair and the defect to be detected.

  Thus, in order to set the optimal inspection conditions for inspecting the subject 101, it is necessary to set the combination of the center wavelength, the wavelength width, and the incident angle in accordance with the type of subject and the type of defect. However, in order to find and set the optimum inspection conditions for defect detection, there are many combinations, and it is very complicated for the user to determine and carry out the optimum inspection conditions. Note that it is practically impossible to image a subject under all combinations of the examination conditions and compare these data because there are many combinations of the examination conditions.

  In view of the above problems, an object of the present invention is to provide a defect inspection apparatus capable of easily selecting an appropriate inspection condition.

  In view of the above problems, the defect inspection apparatus of the present invention has the following configuration in order to solve the above problems.

  The defect inspection apparatus of the present invention is configured to illuminate a subject with illumination light from a light source as a substantially parallel light beam, and to collect reflected light from the subject out of the parallel light flux from the illumination unit, An image forming means for forming an image of a subject, an image pickup means for picking up an image formed by the image forming means, and an optical path between the light source and the image pickup means are disposed in the optical path. A wavelength selecting means for setting a center wavelength and a wavelength width of the light beam, a defect detecting section for detecting a defect of the object from an image picked up by the image pick-up means, and a plurality of the center wavelengths and the incident angles measured above. Inspection conditions for acquiring reflectance data of a portion of the subject that does not have a defect and setting the center wavelength and the incident angle at the time of capturing an image used for detection of the defect based on the reflectance data A setting unit.

  The present invention can provide a defect inspection apparatus capable of easily selecting an appropriate inspection condition.

    Hereinafter, embodiments of the present invention will be described with reference to the drawings.

  A defect inspection apparatus according to an embodiment of the present invention will be described with reference to FIGS. FIG. 1 is a schematic side view showing a defect inspection apparatus 1 according to the present embodiment. FIG. 2 is a schematic side view showing a wavelength selection unit of the defect inspection apparatus 1 in FIG. FIG. 3 is a schematic block diagram showing a control system of the defect inspection apparatus 1 in FIG. The defect inspection apparatus 1 according to the present embodiment inspects the subject 101 that is a multilayer substrate. In the subject 101, a resist layer having a plurality of periodic patterns is formed on the uppermost layer.

The defect inspection apparatus 1 includes a stage 2, an illumination unit 3, an imaging unit 4, an imaging unit 70, and a control system 200.
The stage 2 is a support mechanism that holds and moves the subject 101. In the present embodiment, the stage 2 is configured so that the subject 101 can be adsorbed and held and moved along two predetermined axes orthogonal to each other. In this specification, one axis is defined as the X axis, and the other axis is defined as the Y axis. In the present specification, an axis perpendicular to the X and Y axes is taken as a Z axis. Accordingly, the Z axis substantially coincides with the normal direction of the surface of the subject 101 supported by the stage 2.

  In addition, the defect detection apparatus is provided with a not-shown subject transport unit provided with a substrate transport robot that transports a semiconductor wafer substrate or FPD glass substrate to be inspected onto the stage 2. The subject transport unit includes a stocker (cassette) that stores the subject 101 before the examination and a stocker (cassette) that contains the subject 101 after the examination.

  The illumination unit 3 is an illumination unit that illuminates the subject 101, and includes a light source 31, a wavelength selection unit 32, an optical fiber 33, and a collimating lens 34. The light source 31 is a known lighting device. In the present embodiment, the light source 31 is a halogen lamp and outputs white light. Light from the light source 31 enters the wavelength selector 32. Specifically, the light from the light source 31 enters an incident slit 32a of a wavelength selection unit 32 described later.

  The wavelength selection unit 32 is a monochromator whose transmission wavelength width is variable by changing the widths of the openings of the entrance slit 32a and the exit slit 32a, for example. The wavelength selector 32 includes an entrance slit 32a, a concave diffraction grating 32b, and an exit slit 32c. The arrangement of the entrance slit 32a and the concave diffraction grating 32b is set so that the light beam that has passed through the entrance slit 32a enters the concave diffraction grating 32b. The concave diffraction grating 32b is configured to be rotatable in the direction of the arrow shown by a movable device (not shown), and the center wavelength of light incident on the exit slit 32c can be selected by rotating the concave diffraction grating 33b.

  The optical fiber 33 enters the light from the exit slit 32c from the incident end 33a, and emits the incident light from the exit end 33b. The emission end 33 b is disposed on the focal point of the collimating lens 34. The collimating lens 34 shapes the light that has passed through it into linear parallel light. The collimating lens 34 illuminates a linear region along the direction of the subject 101 that intersects the optical axis. In other words, the collimating lens 34 is arranged so that the width direction is parallel to the Y axis.

  The illuminating unit 3 has a rotation mechanism (not shown) so that the angle (incident angle) θi formed by the optical axis of the collimating lens 34 and the normal line (Z axis) of the surface to be inspected of the subject 101 can be changed. The output end 33b of the optical fiber 33 is supported by the portion.

  The illumination unit 3 having the above configuration transmits the collimating lens 34 and illuminates the subject 101 with linear light. In this specification, an area illuminated by the illumination unit 3 is an illumination area. The illumination area of the illumination unit 3 is imaged on the imaging unit 4 and imaged by the imaging unit 70.

  The imaging unit 4 includes a collimating lens 41 and an imaging lens 43. The collimating lens 41, the imaging lens 43, and an imaging unit switching unit 73, which will be described later, can change the angle (inspection angle) θd formed by the optical axis of the collimating lens 41 and the normal line (Z axis) of the subject 101. , And is supported by a support portion having a rotation mechanism (not shown). The collimating lens 41 collects the reflected light having the inspection angle θd out of the reflected light generated from the illumination area and guides it to the imaging lens 43. In this specification, the term “reflected light” includes specularly reflected light, diffracted light, and scattered light.

  The imaging lens 43 is arranged so that the entrance pupil is arranged at a substantially focal position of the collimating lens 41. The imaging lens 43 forms an image of the light flux from the collimating lens 41 on the corresponding imaging unit 70.

  The imaging unit 70 includes an imaging unit 71 that is a known imaging device, a polychromator 72 that splits incident light and measures the spectral intensity of each wavelength, and an imaging unit switching unit 73.

  The imaging unit 71 has an imaging surface 71a (see FIG. 4) for capturing an image by the imaging lens 43. In the present embodiment, the imaging unit 71 is a line image sensor, and can capture an image over the entire width of the subject at a time. The imaging unit 71 is supported by the imaging unit switching unit 73 so as to be movable in the direction of the arrow d2 illustrated.

  The polychromator 72 is a spectral intensity measuring means for spectrally analyzing the received light and measuring the spectral intensity collectively. The polychromator 72 includes a movable fixed portion 72a, an optical fiber 72b, and a polychromator body 72c. The movable fixed portion 72a supports the incident end 72d of the optical fiber 72b so as to be movable in the direction of the arrow d1 shown in the drawing along the Y axis. Further, the movable fixing portion 72a is supported by the imaging unit switching portion 73 so as to be movable in the direction of the arrow d2 shown in the figure. The optical fiber 72b has an emission end connected to the polychromator body 72c, and sends light incident from the incident end 72d to the polychromator body 72c. The polychromator main body 72 c measures the spectral intensity of the incident light and sends the measurement result to the control system 200.

  The imaging unit switching unit 73 includes a driving unit (not shown) that drives the imaging unit 71 and the movable fixing unit 72a. The imaging unit switching unit 73 moves the imaging unit 71 and the movable fixing unit 72a along the driving direction d2 orthogonal to the longitudinal direction d1 by the driving unit. With this configuration, the imaging unit switching unit 73 is configured such that one of the imaging unit 71 and the movable fixed unit 72a is arranged at the imaging position of the subject by the imaging lens 43 by driving the driving unit according to the command of the control system 200. These are selectively moved as required. Specifically, when the imaging unit switching unit 73 performs imaging by the imaging unit 71, the imaging unit 71 controls the imaging unit 71 so that the imaging surface 71a of the imaging unit 71 forms an image 80 by the imaging lens 43 shown in FIG. The imaging unit 71 is moved so as to match the position. At this time, the polychromator 72 is disposed at a position shifted from the image 80 in the driving direction d2. When switching from measurement to measurement by the polychromator 72, the drive unit of the imaging unit switching unit 73 is controlled so that the incident end 72d of the polychromator 72 coincides with the position of the image 80 shown in FIG. The movable fixed portion 72a is moved. At this time, the imaging surface 71a is arranged at a position shifted from the image 80 in the driving direction d2. Furthermore, the imaging unit switching unit 73 controls the driving unit to move the incident end 72d to a desired measurement position along the longitudinal direction d1 and to stop the position after the movement.

  Note that the imaging unit switching unit 73 and the polychromator main body 72c are connected to the control system 200, and are controlled in driving, and send captured images and measurement results to the control system 200. With the above configuration, the defect inspection apparatus 1 is configured to rotate the incident angle θ1 on the illumination side and the inspection angle θd on the imaging side, so that the illumination unit 3 illuminates the subject 101 at an arbitrary incident angle θi and performs imaging. The unit 70 can receive the reflected light from the illumination area at an arbitrary inspection angle θd.

Next, the control system 200 will be described with reference to FIG.
The control system 200 includes a stage control unit 201, a drive circuit 202, an optical system control unit 203, a subject transport control unit 204, an image interface (image I / F) 205, an image processing device 206, and a monitor 207. 210, an image storage device 208, a host computer 209, a keyboard 211, and a sequencer 213.

The stage control unit 201 is connected to the stage 2 and controls the driving of the stage 2.
The drive circuit 202 is connected to the imaging unit 70 and controls the driving of the imaging unit 70 in synchronization with the control of the stage control unit 201.

  The optical system control unit 203 is connected to the illumination unit 3 and the imaging unit 4 and controls the driving thereof. More specifically, the optical system control unit 203 controls the light amount, wavelength width, and center wavelength of the illumination light of the illumination unit 3, and the angles of the illumination unit 3 and the imaging unit 4 with respect to the subject (incident angle). θi and inspection angle θd) are controlled.

The subject transport control unit 204 is connected to the above-described subject transport unit, and controls the driving of the subject transport unit.
An image interface (image I / F) 205 is connected to the imaging unit 70 and the image processing device 206. The image acquired by the imaging unit 70 is transferred to the image processing device 206 via the image I / F 205.

  The image processing apparatus 206 is connected to the host computer 209, and the drive is controlled by the host computer 209. The image processing apparatus 206 reconstructs an image for each line sent from the imaging unit 70 that is a line image sensor, and creates an entire image of the subject 101. Further, the image processing device 206 performs image processing on the whole image and extracts defects. In addition, the image processing apparatus 206 sends defect data such as the type, number, position, and area of defects related to the extracted defects to the host computer 209. A monitor 207 and an image storage device 208 are further connected to the image processing device 206. A monitor 207 is an image display device that displays an image or the like created by the image processing device 206.

  The image storage device 208 is a data storage device for storing the inspection image processed by the image processing device 206 and the processed image. The object 101 and the inspection image, the extracted defect image, and the non-defective non-defective object 101. The reference image etc. which imaged are stored.

  The host computer 209 is connected to the imaging unit 70, the image processing device 206, the monitor 210, the keyboard 211, and the sequencer 213. The host computer 209 controls driving of the image processing apparatus 206 and the sequencer 213 and displays necessary settings for this control on the monitor 210 as a control menu. In addition, the host computer 209 acquires the measurement result of the polychromator 72.

The keyboard 211 is an input device for inputting commands and information necessary for the control menu and / or other controls to the host computer 209.
The sequencer 213 is connected to the stage control unit 201, the drive circuit 202, the optical system control unit 203, and the subject transport control unit 204, and sends control target drive sequences controlled by them. Also, the sequencer 213 sends a drive command from the host computer 209 to them.

  The host computer 209 has a storage unit (not shown), and control information necessary for the control is stored in the storage unit. The control information includes, for example, inspection conditions for each type of subject (such as optical system settings, inspection area, image processing conditions, acceptance criteria when performing defect quality determination), inspection data, and the like. This control information can be stored in the storage unit in advance before inspection, or can be input by the keyboard 211 or the like at the time of inspection and stored in the storage unit.

Below, operation | movement of the defect inspection apparatus 1 of the said structure is demonstrated.
When the defect inspection apparatus 1 inspects the subject 101, first, inspection condition setting processing for setting inspection conditions (recipe) is performed. In this inspection condition setting process, a part (for example, one line or a plurality of lines) of the subject 101 is imaged, and based on the captured image, the center wavelength and wavelength width of the light source 31 set at the time of inspection, and the illumination unit 3. The incident angle θi and the inspection angle θd are calculated. In this inspection condition setting process, the defect inspection apparatus 1 of the present embodiment sets the inspection conditions in the procedure shown in FIG. The host computer 209 has a processing procedure for performing the inspection condition setting process in the storage unit, and automatically calculates the inspection condition according to this processing procedure. The reason for setting the inspection conditions will be described below with reference to FIG.

[Subject placement step 501]
In the subject placement step 501, the host computer 209 issues a drive command to the subject transport control unit 204 and the stage control unit 201 via the sequencer 213. Upon receiving this drive command, the subject transport control unit 204 drives the subject transport unit, takes out one subject 101 from the stocker, and places it on the stage 2. Subsequently, a subject imaging step 502 is performed. In the present embodiment, the subject 101 is a substrate to be inspected having a defect.

[Subject imaging step 502]
In the subject imaging step 502, the host computer 209 issues a drive command to the drive circuit 202, drives the imaging unit switching unit 73, and places the imaging unit 71 at the imaging position. At the same time, the host computer 209 issues a drive command to the optical system control unit 203 and sets the arrangement of the illumination unit 3 and the imaging unit 4 so that the incident angle θi and the inspection angle θd match. In addition, the host computer 209 sets the wavelength of the light source and the incident angle θi used during imaging in this step. In this subject imaging step, in order to capture the entire subject for the purpose of confirming the arrangement of the pattern of the subject, the wavelength of the light source and the incident angle may be set to preset reference values.

  Subsequently, the host computer 209 drives the stage 2 along the X axis, and moves the subject along its longitudinal direction. In synchronization with this, the host computer 209 causes the imaging unit 71 to image the subject 101 line by line along the width direction of the subject. In addition, the host computer 209 sequentially sends captured images to the image processing apparatus 206 via the image I / F 205. When imaging of the entire surface of the subject 101 is completed by driving the imaging unit 71 and the stage 2, the image processing apparatus 206 forms a whole subject image obtained by imaging the whole subject 101, and this whole subject image is hosted. Send to computer 209. Subsequently, a reference position setting step 503 is performed.

[Reference position setting step 503]
In this reference position setting step 503, when the host computer 209 selects an appropriate center wavelength and incident angle θi, a reference position that is a measurement location of the subject is set. At the time of this setting, the host computer 209 detects the position where an examination region, for example, a pattern having a relatively large area, is formed from a plurality of patterns on the subject 101 with reference to the entire subject image. Then, the host computer 209 sets a position having no defect in the detected pattern as a reference position. Since a pattern having a relatively large area is important in the subject, it is preferable to set this pattern at the examination position. However, another position on the subject 101 can be set as the reference position.

  This reference position can be set to a position free of defects detected by performing image processing on the entire subject image, and can be set by the operator by looking at the large pattern displayed on the monitor 207. Is also possible. Subsequently, a reference position moving step 504 is performed.

[Reference position moving step 504]
In the reference position moving step 504, the host computer 209 issues a drive command to the stage control unit 201, and the stage 2 is arranged so that the reference position set in the reference position moving step 503 is imaged by the imaging unit 71. Move. That is, the stage 2 moves the subject 101 so that the reference position is arranged in the illumination area. Subsequently, the imaging unit switching unit 73 moves the movable fixing unit 72a along the driving direction d2 so that the incident end 72d is disposed in the region of the image 80. After the movement of the movable fixed portion 72a is completed, the incident end 72d is moved along the longitudinal direction d1, and is arranged at a position where the reference position is imaged. Subsequently, a reflectance measurement step 505 is performed.

[Reflectance measurement step 505]
In this reflectance measurement step 505, the host computer 209 issues a drive command to the optical system control unit 203, and sets the wavelength selection unit 32 so that white light from the light source 31 enters the optical fiber 33. This setting is achieved by rotating the concave diffraction grating 32b to an angle for regular reflection so that the light beam from the entrance slit 32a converges to the exit slit 32c. Subsequently, while the incident angle θi and the inspection angle θd are matched, the angle is moved from a small angle to a large angle. Along with this movement, the polychromator 72 measures the spectral reflection intensity at the reference position, and takes in the reflectance data as the measurement result into the host computer 209. By this measurement, the reflectance corresponding to each wavelength at a wavelength having a predetermined width is obtained for each incident angle.

  The reflectance data reflects the spectral intensity of the light source 31, the spectral sensitivity of the polychromator 72, and the spectral reflectance or transmittance of each optical element in between. For this reason, in the subsequent correction step 506, the above-described reflectance data is corrected.

[Correction process 506]
In the correction step 506, the host computer 209 takes out the correction coefficient for each wavelength and incident angle θi prepared in advance from the storage unit, and sets the reflectance corresponding to each wavelength and incident angle of the reflectance data. Multiply to correct the reflectance data. This correction coefficient is obtained as follows and stored in the storage unit. In order to obtain this correction coefficient, a flat plate (for example, single crystal silicon) having a known spectral refractive index is prepared, and the reflectance data of this flat plate is measured in the same procedure as the reflectance measurement step 505, The reflectance at the wavelength and the incident angle θi is calculated from the spectral refractive index. By dividing the calculated values of reflectance at each wavelength and incident angle by the measurement data, a correction coefficient for each wavelength and incident angle θi can be obtained. After correcting the reflectance data measured in the reflectance measurement step 505 as described above, a center wavelength incident angle setting step 507 is subsequently performed.

[Center wavelength incident angle setting step 507]
In the center wavelength incident angle setting step 507, the center wavelength and the incident angle θi in the inspection conditions are obtained from the reflectance data corrected by the correction step 506, and set as the inspection conditions.

  Note that when detecting a defect, the defect of the subject includes a reference image obtained by imaging a (normal) subject having no defect, and an inspection image that is an image of the entire subject obtained by imaging the subject 101. Are detected by the change of these luminance values. Therefore, in order to detect a defect more reliably, it is preferable that the defect and the portion having no defect have a large luminance difference. For this reason, the central wavelength and the incident angle θi are selected such that a luminance difference can clearly appear when comparing the defect to be inspected with a normal part on the subject. In order to obtain such a center wavelength and incident angle θi, the center wavelength incident angle setting step 507 is performed as follows.

  First, the reflectance distribution at the reference position is examined from the corrected reflectance data. Note that this reference position is a normal part having no defect of the subject 101 as described above. Therefore, the reflectance distribution of the normal part can be examined by this. This distribution is shown in FIG. In FIG. 7, the reflectance of each wavelength and the incident angle θi is converted to gray scale as shown in the legend 602, and the horizontal axis is set to the wavelength and the vertical axis is set to the incident angle θi. Shown. Based on the distribution of the reflectance data, a function of the reflectance characteristics at the reference position corresponding to the incident angle and wavelength (a function based on the reflectance data) can be obtained. In this legend, the area painted in white indicates the position of the local maximum value of the function based on the reflectance data.

  In general, the reflectance distribution of the subject 101 as shown in FIG. 7 changes in a complicated manner depending on the type of the subject and the imaging position. Further, the defect reflectance data changes in a complicated manner depending on the type of defect when compared with the reflectance data of a normal part. Although the reflectance data shows such a complicated change, the reflectance at a typical defect changes to some extent with respect to the reflectance of the surface of a normal subject. For example, in the defect (line width defect) in which the line width of the periodic pattern of the resist is narrower than a predetermined width, in the distribution, the value of the maximum value of the reflectance in the spectral reflectance and the position of the maximum value Shifts. This also applies to defects such as film thickness unevenness. Therefore, the difference in reflectance between the defect and the normal part can be easily made by comparing the incident angle θi and the wavelength at the maximum value of the reflectance data of the normal part.

  Further, when the incident angle θi is determined in the reflectance data, the reflectance distribution at the incident angle θi can be shown as a curve graph (reflectance curve) by connecting them in order of wavelength values. For example, the reflectance curve in the region of the maximum value pointed out by reference numeral 603 in FIG. 7 is obtained by plotting each reflectance when the incident angle θi is set to 60 degrees in order of wavelength value. It is formed. A reflectance curve at an incident angle of 60 degrees created in this way is indicated by reference numeral 702 in FIG. The reflectance curve 703 is a reflectance curve at a normal site (reference position). In FIG. 8, a reflectance curve 703 at an incident angle of 60 degrees in the reflectance data of the line width defect is indicated by a broken line. Even when the two are compared, the reflectance curve 703 of the defect is changed in the height and position of the peak of the maximum value with respect to the reflectance curve 702 of the normal part. Specifically, when the reflectance curve 702 and the reflectance curve 703 are compared at the peak wavelength of the reflectance curve 702, these differences can be detected as a difference in reflectance. When these are compared at the wavelength 705 at the intersection of the reflectance curve 703 and the reflectance curve 702, there is no difference in reflectance. Therefore, it can be seen that a normal object and a defective object cannot be compared depending on the selection of the wavelength. Thus, it can be understood that the reflectance data of the defect is useful because it is highly likely that the difference in peak is clear by comparing it with the maximum value of the reflectance in the reflectance data of the normal part. The defect reflectance data can be measured by performing the reflectance measurement step 505 and the correction step 506 on the defect to be measured.

  In this step, specifically, the maximum value is selected from other than that except for the incident angle θi and the both ends in the wavelength measurement range, which may contain errors. The host computer 209 stores the incident angle θi and wavelength at the maximum value selected as described above as inspection conditions (incidence angle θi and center wavelength) in specular reflection imaging at the time of inspection. As described above, the host computer 209 according to the present embodiment sets an inspection condition for setting the center wavelength and the incident angle θi at the time of imaging (inspection) of an image used for detection of the defect based on reflectance data. It is a setting part. Subsequently, a variable wavelength width imaging step 508 is performed.

[Variable wavelength imaging process 508]
In the variable wavelength width imaging step 508, the host computer 209 first sets the wavelength selected by the wavelength selection unit 32 to the center wavelength obtained in the center wavelength incident angle setting step 507. At the same time, the illumination unit 3 is similarly set to the incident angle θi obtained in the center wavelength incident angle setting step 507. Correspondingly, the inspection angle θd of the imaging unit 4 is also set. Further, the imaging unit 70 sets the imaging unit switching unit 73 so that the imaging unit 71 can capture an image.

  In this state, the wavelength selection unit 32 images the entire surface of the subject 101 with the wavelength width being minimized. After completing the imaging, the wavelength width is increased in order, and imaging of the entire surface of the subject 101 is repeated up to a predetermined maximum width. Each captured image is stored in the host computer 209. Note that the minimum and maximum wavelength widths to be imaged and the increment of the wavelength width after imaging may be stored in advance in the host computer 209 or may be input by the operator via the keyboard. It is. Subsequently, a wavelength width setting step 509 is performed.

[Wavelength setting step 509]
In this wavelength width setting step, each image captured in the wavelength width variable imaging step 508 is displayed on the monitor 207. From the display image, the user selects an image captured with the minimum wavelength width from images in which the influence of defects below the resist layer, such as uneven film thickness, is sufficiently suppressed. The host computer 209 stores the wavelength width of the image and sets it in the wavelength selection unit 32.
In this way, the inspection condition setting process ends.

  Next, the operation of the defect inspection apparatus 1 after the inspection conditions are set as described above will be described. When an operation menu is displayed on the monitor 210 and the operator instructs the start of the examination together with the type of the subject 101 using the keyboard 211, a condition corresponding to the subject 101 is selected from the examination conditions stored in the host computer 209. Searched. The searched conditions are transmitted to the optical system control unit 203 via the sequencer 213, and the illumination system 3, the imaging unit 4, and the imaging unit 70 are set by the optical system control unit 203. At this time, the imaging unit switching unit 73 arranges the imaging unit 71 at the imaging position of the imaging unit 4.

  Subsequently, the host computer 209 takes out one subject 101 to be examined from a stocker (not shown) and places it on the stage 2. The stage 2 conveys the subject 101 along the X axis at a constant speed, and images the subject 101 one line at a time by the imaging unit 71 in synchronization with this conveyance. The captured line-shaped image is sent to the image processing apparatus 206 via the image I / F 205, and an inspection image that is an image of the entire subject is formed. The image processing device 206 calls a reference image from the image storage device 208, compares it with the inspection image, and detects a defect. Further, the image processing apparatus 206 sends data such as the position and area of the detected defect to the host computer 209. The host computer 209 compares the number of defects, the area, and the like with the acceptance criteria of the subject 101 and makes a pass / fail determination. When the examination is completed, the subject 101 is classified as good or bad and is transported to a stocker for storing the subject after the examination (not shown). In this way, the examination of one subject 101 is completed.

  As described above, the defect inspection apparatus 1 according to the present embodiment measures the reflectance data of the normal part (the part having no defect) of the subject 101, and based on this reflectance data, the detection target is detected. Inspection conditions including the incident angle θi of the illumination unit 3 and the center wavelength, which are suitable for detecting a defect, can be determined. Thereby, the defect inspection apparatus 1 of the present embodiment can easily set complicated inspection conditions, and can reliably detect a target defect as a result of the setting.

  The imaging unit 71 of the present embodiment is a line image sensor. At the same time, the stage 2 can move the subject 101 relative to the imaging unit 71 in the length direction of the imaging region of the line image sensor and in a direction substantially perpendicular to the length direction. With this configuration, the defect inspection apparatus 1 according to the present embodiment has an area image sensor while the stage 2 and the imaging unit 71 are driven in synchronization to reliably capture the entire surface of the subject. Compared to, it is possible to reduce the size. However, the imaging unit 71 can be replaced with an area image sensor that can collectively image the entire surface of the subject 101.

The polychromator 72 according to the present embodiment is a spectral intensity measuring unit that spectrally measures the spectral intensity of the light in the wavelength band included in the light source. The defect inspection apparatus 1 can measure the reflectance data by this spectral intensity measuring means. For this reason, the defect inspection apparatus 1 can acquire the reflectance data of a predetermined wavelength range collectively, and can perform a quicker measurement.
The host computer 209 automatically sets the center wavelength and the incident angle in accordance with the processing procedure of the inspection condition setting process. Therefore, the defect inspection apparatus 1 according to the present embodiment can easily set complicated inspection conditions.

  In the automatic setting of the inspection condition, the host computer 209 sets the wavelength and the incident angle θi at the extreme value of the reflectance data to the center wavelength and the incident angle θi of the inspection condition. Thereby, the defect inspection apparatus 1 of this Embodiment can obtain | require and set more reliably the center wavelength and incident angle (theta) i from which the difference of the reflectance of a defect and a normal site | part tends to come out at the time of inspection.

  In the present embodiment, for example, when the reflectance data of a defect and a normal part are compared, and the shift amount between the peak of the maximum value and the position is small, the wavelength 704 in the vicinity of the maximum value or the minimum value. Then, as shown in FIG. 9, the difference in reflectance does not appear remarkably. In such a case, it is advantageous to use a region having a large spectral reflectance (slope with respect to wavelength and incident angle) between the maximum value and the minimum value, and a function based on the reflectance data is differentiated, The extreme incident angle θi and the wavelength are set as the incident angle θi and the center wavelength at the time of inspection. With this selection, an appropriate incident angle θi and a center wavelength as indicated by reference numeral 706 in FIG. 9 are selected. This selection is preferable because the inspection can be performed with a larger difference in reflectance than when the wavelength 704 is selected.

  Further, in the present embodiment, the imaging unit 71 images the subject 101 with a plurality of wavelength widths in cooperation with the wavelength selection unit 32. These images are compared with each other, and the wavelength width of light used for imaging is determined based on the comparison result. Since the defect inspection apparatus 1 of this Embodiment can determine a wavelength width based on such a comparison result, it can select an appropriate wavelength width more reliably.

  In the present embodiment, these images are displayed on the monitor 207 and the wavelength width of the selected image is set as the inspection condition so that the operator can confirm images of a plurality of wavelength widths. Can do. Therefore, the operator can set an appropriate wavelength width only by selecting an image that can obtain a preferable inspection result. Further, the image processing apparatus 206 applies image processing for defect detection to these images, selects an image with few false detections, and sets the wavelength width obtained by capturing the selected image as the wavelength width of the inspection condition. It is also possible to do. Thereby, the defect inspection apparatus 1 can automatically select the wavelength width.

  In the present embodiment, the imaging position of the imaging unit 71 and the measurement position of the polychromator 72 are set to be the same. However, it is also possible to arrange a half mirror or the like in the image forming unit 4, branch the optical path, distribute the imaging unit 71 and the polychromator 72 to the branched optical path, and divide the measurement positions. In this way, when the imaging position of the imaging unit 71 and the measurement position of the polychromator 72 are set to different positions, the imaging unit switching unit 73 can be omitted. In addition, the imaging unit 71 and the polychromator 72 may be driven by separate driving units, and the imaging unit switching unit 73 may be omitted.

  In this embodiment, when measuring reflectance data, the wavelength and the measurement interval of the incident angle θi can be set arbitrarily. Furthermore, the measurement speed can be shortened by increasing these measurement intervals and appropriately interpolating between the data.

  Further, the variable wavelength width imaging step 508 of the present embodiment can be replaced as in the following variable wavelength width imaging step 508 '. In the variable wavelength width imaging step 508 ′, as in the variable wavelength width imaging step 508, the angles θi and θd between the illumination unit and the imaging unit are set to the angle of the inspection condition set in the central wavelength incident angle setting step 507. Further, the wavelength width of the wavelength selector 32 is set to the minimum. In this state, the wavelength selection unit 32 captures an image of the entire subject surface with respect to each center wavelength by the imaging unit 71 while changing the center wavelength in the vicinity of the center wavelength at the time of examination. The group of images taken in this way is integrated. When such a variable wavelength width imaging step 508 ′ is performed, the defect inspection apparatus 1 can acquire an image corresponding to an arbitrary wavelength width by integrating the image group.

  Further, the reflectance measurement step 505 of the present embodiment can be replaced with the following reflectance measurement step 505 '. In the reflectance measurement step 505 ′, the imaging unit switching unit 73 is driven, and the imaging unit 71 is disposed at the image forming position of the image by the imaging unit 4. At the same time, the incident angle θi of the illumination unit 3 and the inspection angle θd of the imaging unit 4 are set to the minimum. Subsequently, the wavelength selector 32 changes the center wavelength from the minimum to the maximum with the wavelength width minimized. In synchronization with this, the imaging unit (linear image sensor) 71 captures a one-line image including the reference position in the width direction of the subject 101 for each center wavelength. After this imaging is completed, the same imaging is repeated while changing the incident angle θi and the inspection angle θd with a constant change width. After all the imaging is completed, the luminance of the pixel including the reference position is taken into the host computer 209 as reflectance data. According to this reflectance measuring step 505 ', the polychromator 72 can be omitted.

FIG. 1 is a schematic side view showing a defect inspection apparatus according to an embodiment of the present invention. FIG. 2 is a schematic side view showing the wavelength selection unit in FIG. FIG. 3 is a block diagram showing the control system in FIG. 4 is a schematic front view showing the imaging unit switching unit in FIG. 1 during imaging. FIG. 5 is a schematic front view showing the imaging unit switching unit in FIG. 1 at the time of measurement. FIG. 6 is a flowchart showing the inspection condition setting process. FIG. 7 is a schematic diagram showing reflectance data. FIG. 8 is a schematic diagram showing a reflectance curve. FIG. 9 is a schematic diagram showing a reflectance curve when the shift amount is small. FIG. 10 is a schematic side view showing a conventional defect inspection apparatus. FIG. 11 is a schematic perspective view showing another conventional defect inspection apparatus.

Explanation of symbols

  DESCRIPTION OF SYMBOLS 1 ... Defect inspection apparatus, 101 ... Test object, 2 ... Stage, d1 ... Longitudinal direction, d2 ... Drive direction, 200 ... Control system, 201 ... Stage control part, 202 ... Drive circuit, 203 ... Optical system control part, 204 ... Subject transport control unit, 205 ... image interface, 206 ... image processing apparatus, 207 ... monitor, 208 ... image processing apparatus, 209 ... host computer, 210 ... monitor, 211 ... keyboard, 213 ... sequencer, 3 ... illumination part, 31 DESCRIPTION OF SYMBOLS ... Light source, 32 ... Wavelength selection part, 32a ... Incidence slit, 32b ... Concave diffraction grating, 32c ... Output slit, 33 ... Optical fiber, 33a ... Incident end, 33b ... Other end, 34 ... Collimating lens, 4 ... Imaging part, DESCRIPTION OF SYMBOLS 41 ... Collimating lens, 43 ... Imaging lens, 501 ... Subject placement process, 502 ... Subject imaging process, 503 ... Reference position setting process 504: Reference position moving step, 505 ... Reflectance measuring step, 506 ... Correction step, 507 ... Center wavelength incident angle setting step, 508 ... Variable wavelength width imaging step, 509 ... Wavelength width setting step, 602 ... Legend, 70 ... Imaging unit, 702, 703 ... reflectance curve, 71 ... imaging unit, 71a ... imaging surface, 72 ... polychromator, 72a ... movable fixed part, 72b ... optical fiber, 72c ... polychromator body, 72d ... incident end, 73 ... imaging Switching unit, 8 ... heat absorption filter, 80 ... image, θd ... inspection angle, θi ... incident angle

Claims (1)

Illuminating means for illuminating the subject with illumination light from the light source as a substantially parallel light beam;
Imaging means for condensing the reflected light from the subject out of the parallel luminous flux from the illuminating means, and forming an image of the subject;
Imaging means for capturing an image formed by the imaging means;
A wavelength selection unit that is disposed in an optical path between the light source and the imaging unit and sets a center wavelength and a wavelength width of a light beam in the optical path;
A defect detection unit that detects a defect of the subject from an image captured by the imaging unit;
An angle setting means for setting an incident angle of illumination light of the illumination means;
Obtaining reflectance data of a portion of the subject that does not have a defect measured at a plurality of center wavelengths and incident angles, and based on the reflectance data, the image at the time of capturing an image used for detection of the defect A defect inspection apparatus comprising: an inspection condition setting unit that sets a center wavelength and an incident angle.

Images