JP2012026733A - Optical defect detection device and method, and defect observation device provided with the same - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a device capable of detecting a detect of an observation object and surely inputting the defect of the observation object within the field of view of an electron microscope, or the like in a device which observes a defect detected by an optical defect detection device or an optical visual inspection device detailedly, and to provide a device capable of detecting a defect of an object in the optical defect detection device or the optical visual inspection device.SOLUTION: In an optical microscope 6 mounted with a dark field illumination unit capable of two-wavelength illumination, a spatial distribution optical element 201 having a function for controlling the scattering direction distribution of light intensity, polarization, wavelength, or the like is inserted into a pupil surface during performing dark field observation.

Description

  The present invention relates to an optical defect detection method / optical defect detection apparatus for inspecting a sample surface or defects existing on the sample surface, or a defect observation apparatus provided with the same.

  For example, in a semiconductor device manufacturing process, if there is a foreign object or a pattern defect such as a short circuit or disconnection (hereinafter referred to as a defect including a foreign object or a pattern defect) on a semiconductor substrate (wafer), a wiring insulation failure, a short circuit, etc. Cause a defect. Further, with the miniaturization of circuit patterns formed on the wafer, finer defects also cause breakdown of capacitors and breakdown of gate oxide films. These defects are caused by various causes such as those generated from the moving parts of the transfer device, those generated from the human body, those generated by reaction inside the processing apparatus by the process gas, those mixed in chemicals and materials, etc. Are mixed in various states. For this reason, it is important for mass production of semiconductor devices to detect defects generated in the manufacturing process, quickly identify the source of the defects, and prevent the formation of defects.

  Conventionally, in order to investigate the cause of the defect, first, the defect position is identified by a defect inspection apparatus, and the defect identified by the SEM (Scanning Electron Microscope) is observed and classified in detail. There was a method to estimate the cause of the defect compared with the database.

  Here, the defect inspection apparatus is an optical defect inspection apparatus that illuminates the surface of a semiconductor substrate with a laser and observes scattered light from the defect in a dark field to identify the position of the defect, a lamp, a laser, or an electron beam. Is an optical appearance inspection apparatus or SEM inspection apparatus that identifies a defect position on a semiconductor substrate by detecting a bright-field optical image of the semiconductor substrate and comparing it with reference information. Such an observation method is disclosed in Patent Document 1 or 2.

  In addition, regarding an apparatus for observing defects in detail with an SEM, each of Patent Document 3, Patent Document 4 and Patent Document 5 uses an SEM type defect inspection using position information of defects on a sample detected by another inspection apparatus. Detect the position on the sample with the optical microscope attached to the equipment and correct the position information of the defect obtained by other inspection equipment, and then observe the defect in detail with the SEM type defect observation equipment (review) And a device for the same, and a method of optically detecting the height of the surface of the sample and observing the defect with an SEM type defect observing apparatus to adjust the surface of the sample to the focal position of the SEM.

JP-A-7-270144 JP 2000-352697 A US Pat. No. 6,407,373 JP 2007-71803 A JP 2007-235033

  When detecting defects on the surface of a semiconductor substrate using an optical defect inspection apparatus, the spot size of the laser beam for illuminating the semiconductor substrate surface with dark field illumination is increased to increase the inspection throughput. Irradiating by scanning. For this reason, the accuracy of the position coordinates obtained from the position of the laser beam spot that scans the surface of the semiconductor substrate includes a large error component.

  If the SEM is used to observe the defect in detail based on the position information of the defect including such a large error component, it is in the field of view of the SEM that is observed at a magnification much higher than that of the optical defect inspection apparatus. In some cases, defects that you want to observe do not enter. In such a case, in order to put an image of a defect to be seen in the field of view of the SEM, the defect is searched for while moving in the field of view of the SEM. However, it takes time to reduce the throughput of the SEM observation. It will cause.

  Accordingly, one of the objects of the present invention is to provide an optical defect inspection apparatus or optical appearance when a defect detected by an optical defect inspection apparatus or optical appearance inspection apparatus is observed in detail using an SEM. An object of the present invention is to provide a defect observation apparatus that can detect a fine defect detected by an inspection apparatus with high sensitivity and can be surely placed in an observation field of SEM.

  As a light source used in an optical inspection apparatus or optical appearance inspection apparatus, a single wavelength laser is generally used. The defect detection sensitivity varies depending on the illumination wavelength and the defect shape. For this reason, the light source used in the optical defect inspection apparatus or the optical appearance inspection apparatus and the defect detected by the optical defect inspection apparatus or the optical appearance inspection apparatus are mounted on the SEM for the purpose of putting them in the field of view of the SEM. Provided is a defect observation apparatus that can solve the problem that the detection sensitivity of a defect differs if the wavelength of a light source used in the defect observation apparatus is different.

  Further, in recent LSI manufacturing, the width of a wiring pattern formed on a wafer has been reduced due to miniaturization of a circuit pattern corresponding to the need for high integration. On the other hand, the height of the wiring pattern is high in order to ensure the conductivity of the wiring.

  Correspondingly, the optical defect inspection apparatus and the size of the defect to be detected are required to be miniaturized. Under these circumstances, optical defect inspection systems are working on increasing the numerical aperture (NA) of inspection objective lenses and developing optical super-resolution technology. Since NA has reached the physical limit, the essential approach is to reduce the wavelength of light used for inspection to the UV and DUV (Deep UV) region.

  However, LSI devices include memory products that are mainly formed with high-density repeating patterns and logic products that are mainly formed with non-repeating patterns, and the structure of patterns to be inspected is complex and diversified. . For this reason, it is difficult to reliably find out defects (target defects) that need to be managed when manufacturing LSI devices. Target defects that are desired to be detected include voids and scratches in the CMP process, in addition to foreign matters generated during each manufacturing process and circuit pattern shapes after etching. Furthermore, there is a short (also referred to as a bridge) between wiring patterns in a metal wiring portion such as a gate wiring or aluminum. In particular, there are many shorts between the wiring patterns that are lower in height than the wiring patterns and are difficult to detect.

  In addition, in a multi-layer wiring LSI device, in addition to the above-described miniaturization of the target defect, since the base pattern of the place where the defect occurs is diversified, it is further difficult to detect. In particular, in a process in which a transparent film such as an insulating film (here, meaning transparent to the illumination wavelength) is exposed on the outermost surface, the intensity unevenness of the interference light due to a minute film thickness difference of the transparent film is caused by optical noise. Become. Therefore, it is necessary to reveal the target defect while reducing the influence of the intensity unevenness of the interference light. In addition, in order to stably manufacture an LSI, it is necessary to accurately manage the defect status of the LSI device. To that end, it is desirable to inspect all the LSI substrates. Therefore, it is necessary to detect the target defect in a short time.

Accordingly, another object of the present invention is to provide an optical defect detection apparatus and method for detecting various defects on a wafer with high speed and high sensitivity, and a defect observation apparatus equipped with the same.
The above and other objects and novel features of the present invention will be apparent from the description of this specification and the accompanying drawings.

  In order to achieve any of the above objects, the invention disclosed in the present application is characterized by inspecting a sample under a plurality of illumination conditions or detection conditions. That is, a plurality of illumination optical systems or a plurality of detection optical systems are used to enable not only high-sensitivity detection specialized for a specific defect shape but also high-sensitivity detection of a plurality of defect shapes.

The following is a description of the outline of typical ones.
(1) An optical defect detection apparatus for inspecting a sample surface, the stage holding the sample and moving in a horizontal or horizontal / vertical direction, and irradiating the sample surface with a plurality of lights having different wavelengths from oblique directions An illumination optical system that captures light scattered from the illumination region on the sample surface with an objective lens and controls the polarization direction of the scattered light captured by the objective lens, or the objective lens The function of selecting the polarization direction of transmission of the captured scattered light, the function of selecting the transmittance of the scattered light captured by the objective lens, or the function of combining at least two of these functions An imaging optical system that controls the scattered light captured by the objective lens and forms an image of the scattered light; and It is an optical defect detecting apparatus characterized by comprising: a detection optical system and an imaging device that detects light, a.
(2) An optical defect detection apparatus for inspecting a sample surface, the stage holding the sample, an illumination optical system for irradiating light obliquely with respect to the sample surface, and the irradiation on the sample surface An objective lens that captures scattered light from a region of interest, and a ratio of scattered light captured by the objective lens to a scattered light from a defect on the sample and scattered light from the unevenness of the sample is increased. A detection optical system having a spatial distribution optical element that controls the captured scattered light and an imaging element that detects the scattered light controlled by the spatial distribution optical element, and the spatial distribution optical element is , An element for controlling the polarization direction of the scattered light captured by the objective lens, an element for selecting the polarization direction of the scattered light captured by the objective lens, or the transmission of the scattered light captured by the objective lens There either different elements for each area, or an optical defect detection apparatus, characterized in that these is a combination of at least two or more.
(3) The optical defect detection apparatus according to (2), wherein the detection optical system further includes branching means for branching scattered light captured by the objective lens into two or more optical paths, The spatial distribution optical element is disposed in each of two or more optical paths branched by the branching unit, and the imaging element is branched into the two or more optical paths and controlled by the spatial distribution optical element. An optical defect detection apparatus, which is arranged so as to detect scattered light, and further has a signal processing system for comparing or processing images detected by the arranged imaging elements.
(4) The optical defect detection apparatus according to (2), wherein the spatial distribution optical element includes a phase shifter, an optical rotator, a polarizer, a spatial filter, a neutral density filter, a phase shifter, or A phase shifter that has a function of changing the phase of the scattered light so that the ratio of the light scattered by the defects on the sample surface to the light scattered by the irregularities on the sample surface is increased, and gives a spatially different phase difference, or A distributed phase shift having a function of converting the polarization direction of light scattered by unevenness of the sample surface or light scattered by defects on the sample surface into linearly polarized light and having a spatially different fast axis and slow axis. The polarization direction is aligned so that the ratio of the light scattered by the defects on the sample surface to the light scattered by the projections or irregularities on the sample surface increases. A polarization rotator that has a higher ratio of light scattered by defects on the sample surface to light scattered by unevenness of the sample surface or of the optical rotator having spatially different fast axis and slow axis. A polarizer having a function of selectively transmitting light and having spatially different transmission axes, or a region where the ratio of light scattered by defects on the sample surface to light scattered by unevenness on the sample surface is high A spatial filter having a function of selectively transmitting light, or a function of selecting a transmittance so that a ratio of light scattered by defects on the sample surface to light scattered by unevenness on the sample surface is increased. A ratio of light scattered by defects on the sample surface to light scattered by a neutral density filter with spatially different transmittances or irregularities on the sample surface The function of controlling the polarization direction of the scattered light, the function of selecting the polarization direction of transmission of the scattered light, or the function of selecting the transmittance of the scattered light, or It is an optical defect detection device characterized in that it is constituted by any one or a combination of photonic crystals having a function combining at least two of these.
(5) The optical defect detection apparatus according to (2), wherein the spatial distribution optical element is a phase shifter, an optical rotator, a polarizer, a spatial filter, an ND filter, or irregularities on the surface of the sample. A phase shifter that has a function of changing the phase of the scattered light so as to increase the ratio of the light scattered by the defect on the sample surface to the light scattered by the surface, or provides a spatially different phase difference, or irregularities on the sample surface A quarter wave plate having a function of converting the polarization direction of the light scattered by the light or the light scattered by the defect on the sample surface into linearly polarized light and having a spatially different fast axis and slow axis, or It has a function of aligning the polarization direction so that the ratio of the light scattered by the defect on the sample surface to the light scattered by the unevenness on the sample surface is increased, and spatially different By a half-wave plate having a fast axis and a slow axis, an optical rotator constituted by either a liquid crystal or a magneto-optic modulator, or a defect on the sample surface with respect to light scattered by irregularities on the sample surface A polarizer having a function of selectively transmitting light in a polarization direction in which the ratio of scattered light is high and having spatially different transmission axes, or of the sample surface with respect to light scattered by unevenness of the sample surface Mask that has a function of selectively transmitting a region where the ratio of light scattered by a defect is high, or a combination of a polarizer and a magneto-optical modulator, a combination of a polarizer and a liquid crystal, or a polarizer and a half wavelength Spatial filter having a light-shielding part composed of either a combination of plates or a two-dimensional array shutter using MEMS, or the surface of the sample An ND filter having a function of selecting the transmittance so that the ratio of the light scattered by the defect on the sample surface to the light scattered by the convex becomes high, or having a spatially different transmittance, or between the polarizer and the liquid crystal The ratio of the light scattered by the defect on the sample surface to the light scattered by the combination or the light filter scattered by the unevenness of the sample surface or a combination of a polarizer and a magneto-optic modulator is high. The function of controlling the polarization direction of the scattered light, the function of selecting the polarization direction of transmission of the scattered light, the function of selecting the transmittance of the scattered light, or at least these A photonic crystal having a function of combining two, or a combination thereof. Is an optical defect detection apparatus, characterized in that the shall.
(6) The optical defect detection apparatus according to (2), wherein the spatial distribution optical element is a phase shifter, an optical rotator, a polarizer, a spatial filter, an ND filter, or unevenness of the sample surface. A phase shifter that has a function of changing the phase of the scattered light so as to increase the ratio of the light scattered by the defect on the sample surface to the light scattered by the surface, or provides a spatially different phase difference, or irregularities on the sample surface A quarter wave plate having a function of converting the polarization direction of the light scattered by the light or the light scattered by the defect on the sample surface into linearly polarized light and having a spatially different fast axis and slow axis, or It has a function of aligning the polarization direction so that the ratio of the light scattered by the defect on the sample surface to the light scattered by the unevenness on the sample surface is increased, and spatially different Polarized light whose ratio of the light scattered by the defect on the sample surface to the light scattered by the half-wave plate or the liquid crystal or the magneto-optic modulator having the fast axis and the slow axis, or irregularities on the sample surface A polarizer having a function of selectively transmitting light in a direction and having spatially different transmission axes, or scattering by defects on the sample surface with respect to light having a plurality of wavelengths scattered by unevenness of the sample surface by the plurality of illuminations Color filters or masks, or a combination of a polarizer and a magneto-optic modulator, or a polarizer and a liquid crystal A spatial filter having a light-shielding portion constituted by either a combination or a combination of a polarizer and a half-wave plate, or the surface of the sample An ND filter, color filter, or polarization having a function of selecting a transmittance so that the ratio of the light scattered by the defect on the sample surface to the light scattered by the convex increases and having spatially different transmittances Light that is scattered by defects on the sample surface relative to light that is scattered by unevenness of the sample surface, or a neutral density filter composed of a combination of a polarizer and a liquid crystal, or a combination of a polarizer and a magneto-optic modulator The function of controlling the polarization direction of the scattered light, the function of selecting the polarization direction of transmission of the scattered light, or the function of selecting the transmittance of the scattered light so that the ratio of , Or a photonic crystal having a function combining at least two of these, or a combination thereof An optical defect detection apparatus characterized by being configured by combining.
(7) A defect observation apparatus including the optical defect detection apparatus according to any one of (1) to (6), wherein position information of defects on the surface of the sample obtained by the optical defect detection apparatus is used. It is a defect observation apparatus characterized by having an electron microscope which observes the defect of the sample aligned based on it.
(8) A defect observation apparatus for inspecting and observing a defect on a sample surface, the stage holding the sample, an illumination optical system for irradiating light obliquely to the sample surface, and the above-mentioned sample surface An objective lens that captures scattered light from the irradiated region, and of the scattered light captured by the objective lens, the ratio of the scattered light from the defect on the sample to the scattered light from the unevenness of the sample is increased. And a detection optical system having a spatial distribution optical element that controls the captured scattered light and an imaging element that detects the scattered light controlled by the spatial distribution optical element, An electron microscope that performs alignment based on positional information of defects or foreign matter on the sample surface obtained by the optical defect detector and observes the defects, and the spatial distribution optical element Is an element for controlling the polarization direction of the scattered light captured by the objective lens, an element for selecting the polarization direction of the scattered light captured by the objective lens, or the transmission of the scattered light captured by the objective lens The defect observing apparatus is characterized in that one of elements having different rates for each region, or a combination of at least two of these elements.
(9) An optical defect detection method for detecting a defect on a sample surface, wherein the sample surface is illuminated with illumination light from an oblique direction, and scattered light from the illuminated region on the sample surface is objective lens The scattered light captured by the objective lens and split by the objective lens is split into two optical paths by either a half mirror or a dichroic mirror, and the light scattered on the sample surface using the branched first light The Fourier-transformed image is detected by the first image sensor, and based on the signal obtained from the first image sensor, the polarization direction arranged on the branched second optical path is controlled and transmitted. Controlling a spatial distribution optical element having a function of selecting a polarization direction to be transmitted and selecting a transmittance, and imaging the branched second light on a second imaging element by an imaging optical system; From the image obtained by the second image sensor Is an optical defect detecting method and detecting defects on charge surface.
(10) An optical defect detection method for inspecting a defect on a sample surface, wherein the sample surface is illuminated obliquely with illumination light, and scattered light from the illuminated area on the sample surface is objective lens And the polarization direction of the captured light is controlled by the spatial distribution optical element, the polarization direction to be transmitted is selected, the transmittance is selected, and the light transmitted through the spatial distribution optical element is transmitted by the imaging optical system. An image is formed on an image sensor, and a defect on the sample surface is detected from an image obtained by the image sensor, and optical conditions such as defect shape, defect type, illumination wavelength, illumination polarization, detection polarization, and detection conditions Is input by the GUI, and the light scattered by the defect on the sample surface with respect to the light scattered by the irregularities on the sample surface for each optical condition such as defect shape, defect type, illumination wavelength, illumination polarization, detection polarization, and detection condition The ratio of Depending on the unevenness of the surface of the sample under the conditions inputted in the GUI from a library storing either the polarization direction transmitted by the spatial distribution optical element, the direction to change the polarization direction, or the transmittance, or a combination thereof The ratio of the light scattered by the defect on the sample surface to the scattered light is increased, either the polarization direction transmitted by the spatial distribution optical element, the direction for changing the polarization direction, or the transmittance, or a combination thereof. An optical defect detection method having a function of selecting and controlling one or a combination of a polarization direction transmitted through the spatial distribution optical element, a direction for changing the polarization direction, or a transmittance. .

  According to the present invention, when a defect detected by an optical defect detection apparatus is observed in detail with an SEM or the like, the defect to be observed can be surely placed in an observation field of view such as an SEM. The throughput of the detailed inspection of the used defect can be increased. In addition, the apparatus can be configured on a small scale.

  Alternatively, various defects on the substrate can be detected with high speed and high sensitivity.

It is a block diagram which shows the structural example of the defect observation apparatus in 1st embodiment of this invention. It is a block diagram which shows the structural example of the illumination optical system shown in the present Example 1. FIG. It is a block diagram which shows the structural example of the optical height detector shown in the present Example 1. FIG. FIG. 3 is a configuration diagram illustrating a configuration example of an optical microscope illustrated in the first embodiment. 3 is a diagram illustrating an example of a mechanism that switches the spatial distribution optical element 201 illustrated in the first embodiment on an optical axis 301 of the imaging optical system 110. FIG. It is a figure which shows description of the scattered light simulation which is an example for determining the optical characteristic of the spatial distribution optical element 201 shown in the present Example 1. FIG. It is a figure which shows the example of the intensity distribution for every polarization | polarized-light component of the intensity | strength for every polarized light of the scattered light intensity from the sample surface obtained using the scattered light simulation, and the scattered light intensity from a defect. It is a figure which shows the example of scattered light intensity distribution for every polarization component of the scattered light intensity from the defect obtained using the scattered light simulation. It is the figure which showed the polarization transmission-axis direction distribution example of the spatial distribution optical element 201 shown in the present Example 1. FIG. It is a figure which shows description of the effect which controls the polarization direction of scattered light. It is a figure explaining the change of polarization at the time of applying a half-wave plate near pupil plane 302. It is a figure explaining the change of polarization at the time of applying a quarter wavelength plate near pupil plane 302. FIG. It is a figure which shows an example of the spatial distribution optical element 201 using the polarization direction control apparatus 205 using the liquid crystal which controls a polarization direction. It is a figure explaining the change of the polarization direction by a polarization direction control apparatus. It is a figure which shows an example explaining the polarization direction control apparatus using a liquid crystal. It is a figure which shows the example of the polarization direction control apparatus which has optical rotation when voltage is not applied, and loses precedence by voltage application. It is a figure which shows the example of the polarization direction control apparatus using the liquid crystal which can control a polarization direction by rubbing of an alignment film. It is a figure which shows an example of the spatial distribution optical element 201 using the polarization direction control apparatus 205 using the magneto-optical modulator using a magneto-optical effect. It is a figure which shows an example with respect to radial polarization at the time of applying the phase shifter 202 to the spatial distribution optical element 201. FIG. It is a figure which shows the example of a shape of a spatial filter. It is a figure which shows the structural example of the spatial filter which can control the light shielding area which used the liquid crystal for the spatial filter. It is a figure explaining the distribution example and effect of the transmittance | permeability of the neutral density filter from which the transmittance | permeability which changes for every area | region. It is a figure which shows description of the example of the spatial filter using the color filter which can be used for the scattered light of multiple wavelengths. It is a figure which shows the example of the spatial distribution optical element by the combination of a polarizer and a spatial filter. It is a figure which shows the positional offset amount calculation image of the defect acquired by dark field observation of the optical microscope in 1st embodiment of this invention. FIG. 4 is a diagram showing a procedure for defect observation in the first embodiment of the present invention. FIG. 3 is a diagram illustrating details of a configuration example of an optical microscope using two imaging elements in the first embodiment of the present invention. FIG. 2 is a diagram illustrating details of a configuration example of an optical microscope using one image sensor in the first embodiment of the present invention. FIG. 3 is a diagram showing details of a configuration example of an optical microscope in the first embodiment of the invention. It is a figure which shows the example of a control method of the spatial distribution optical element 201 in 1st embodiment of invention. It is a figure which shows the example of signal processing in 1st embodiment of invention. It is a figure which shows GUI in 1st embodiment of invention. It is a figure which shows the detail of the structural example of the defect observation apparatus in the 2nd Embodiment of invention. It is a figure which shows the detail of the structural example of the defect observation apparatus in the 2nd Embodiment of invention.

Hereinafter, embodiments of the present invention will be described in detail with reference to the drawings as appropriate.
FIG. 1 shows an example of the configuration of the defect observation apparatus according to the present invention. The defect observation apparatus of the present embodiment is a detailed observation of a sample holder 2 that mounts a sample 1 to be inspected, a stage 3 that can move the sample holder 2 and move the entire surface of the sample 1 under a microscope, and a sample 1 An electron microscope (hereinafter referred to as SEM) 4, an optical height detection system 5 for focusing the electron microscope 4 on the surface of the sample 1, and defects on the sample 1 by optically redetecting defects on the sample 1 At least an optical microscope (optical defect detection device) 6 for acquiring detailed position information, an optical height detection system 7 for focusing the optical microscope 6, and an objective lens 105 of the electron microscope 4 and the optical microscope 6. Control unit 10 for appropriately controlling vacuum chamber 8, stage 3 and electron microscope 4, optical height detector 5, optical height detector 7, height control mechanism 9, image sensor 111, etc. A network 13 connected to a host system such as the interface 11, the database 12, and the optical defect inspection apparatus is appropriately used.

  Furthermore, the optical microscope 6 guides the laser emitted from the dark field illumination unit 101 capable of emitting light of a plurality of wavelengths to the vacuum chamber 8 and controls the illumination position on the surface of the sample 1. Reflecting mirrors 102A and 102B, mirrors 104A and 104B for guiding the laser guided from the reflecting mirrors 102A and 102B to the sample 1 through the vacuum-sealed windows 103A and 103B, and collecting scattered light from the sample 1 or bright field observation Objective lens 105, objective lens height control mechanism 9, half mirror 108 for introducing illumination necessary for bright field observation, bright field light source 109, and objective lens 105 guided through a vacuum-sealed window 107. The imaging optical system 110 for forming an image of the sample 1 by the scattered light collected by the imaging device on the imaging device, and the imaging device 111 are appropriately used. That. In addition, although not shown here, as will be described later, the imaging optical system 110 includes a spatial distribution optical element 201 and a spatial distribution optical element switching mechanism as appropriate. The stage 3, the optical height measuring instruments 5 and 7, the SEM 4, the user interface 11, the database 12, the height control mechanism 9, and the image sensor 111 are connected to the control system 10, and the control system 10 is connected via the network 13. It is connected to a higher system (not shown).

  In the defect observation apparatus configured as described above, in particular, the optical microscope 6 has a defect position detected by the optical defect inspection apparatus on the sample 1 detected by the optical defect inspection apparatus (not shown). The height control mechanism 9 and the optical height measuring instrument 7 have a function as focusing means for performing sample focusing, and have a function of performing re-detection (hereinafter referred to as detection) using position information. The system 10 has a function as position correction means for correcting the position information of the defect based on the position information of the defect detected by the optical microscope 6, and the SEM 4 observes the defect whose position information is corrected by the control system 10. It has a configuration having a function. The stage 3 places the sample 1 and moves between the optical microscope 6 and the SEM 4 so that defects detected by the optical microscope 6 can be observed with the SEM 4.

  The objective lens 105 and the height control mechanism 9 are installed in the vacuum chamber 8. As the configuration of the height control mechanism 9, for example, a configuration using a piezo element or a Z direction along the linear guide using a stepping motor and a ball screw (direction along the optical axis 301 of the imaging optical system 110). ), Or a configuration of moving in the Z direction along a linear guide using an ultrasonic motor and a ball screw.

  The incident mirrors 102A and 102B are used to guide the light emitted from the dark field illumination unit 101 to the vacuum lamp 8 as shown in FIG. The epi-illumination mirrors 102A and 102B have a mechanism that rotates around two axes, ie, an axis along the longitudinal direction of the illustrated mirror and an axis perpendicular to the drawing in order to control the illumination position on the surface of the sample 1. May be.

Details of each part will be described below with reference to FIGS.
FIG. 2 shows a configuration example of the dark field illumination unit 101. The dark field illumination unit 101 includes, for example, a visible light laser, an ultraviolet light laser, a vacuum ultraviolet light laser, an illumination light source 112 that emits a lamp or a light emitting diode, an optical filter 113 that adjusts the intensity of illumination light, and a polarization direction of the illumination light. A wave plate 114 for focusing the illumination light on the sample 1, a nonlinear optical crystal 119 for converting the wavelength of light emitted from the illumination light source 112, a half mirror 120 for dividing the illumination light, a mirror 121 for changing the optical path, and illumination. A fiber 122 that divides light, a dichroic mirror 123 that divides illumination light by wavelength, a nonlinear optical crystal switching mechanism 124 that switches a plurality of nonlinear optical crystals, and an optical path switching mechanism 125 that moves a mirror 121A in and out to switch an optical path are used as appropriate. It is configured. The lens group 118 includes a plano-concave lens 115, an achromatic lens 116, and a cylindrical lens 117 as appropriate. This is a mechanism that can control the illumination region on the surface of the sample 1 from the entire field of view of the optical microscope 6 to the diffraction limit by selecting the lens focal length and adjusting the lens interval. The cylindrical lens 117 provides oblique illumination but a circular illumination region. Is feasible.

  As shown in FIG. 2 (a), the first configuration example of the dark field illumination unit 101 includes two illumination light sources 112A and 112B and each other emitted from the illumination light sources so that two different wavelengths can be illuminated. Optical filters 113A and 113B arranged in the respective optical paths so that the illumination intensity of the illumination lights 701A and 701B having different wavelengths can be adjusted, and the respective optical paths so that the deflection directions of the two illumination lights 701A and 701B can be adjusted. And two lens groups 118A and 118B arranged in respective optical paths so as to focus the two illumination lights 701A and 701B on the sample 1, respectively, and having two wavelengths different from each other. The surface of the sample 1 can be illuminated simultaneously with illumination light.

  FIG. 2B shows a second configuration example of the dark field illumination unit 101. In FIG. 2B, the illumination light 701 emitted from the illumination light source 112 is divided into two illumination lights 701A and 701B by the half mirror 120 so that illumination with two different wavelengths can be performed, and the divided illumination light 701A, Each of 701B is converted into a different wavelength by using different nonlinear optical crystals 119A and 119B. The divided 701B may be guided to the nonlinear optical crystal 119B using a mirror 121 or the like as appropriate. The converted illumination lights 701A and 701B having different wavelengths illuminate the surface of the sample 1 using different optical filters 113A and 113B, wavelength plates 114A and 114B, and lens groups 118A and 118B, respectively. The surface of the sample 1 can be illuminated at the same time by wavelength illumination.

  FIG. 2C shows a third configuration example of the dark field illumination unit 101. In FIG. 2C, the illumination light 701 emitted from the illumination light source 112 is divided into two illumination lights 701A and 701B by the fiber 122 so that illumination with two different wavelengths can be performed, and the divided illumination lights 701A and 701B are divided. Each is converted into a different wavelength by using different nonlinear optical crystals 119A and 119B. The converted illumination lights 701A and 701B having different wavelengths illuminate the surface of the sample 1 using different optical filters 113A and 113B, wavelength plates 114A and 114B, and lens groups 118A and 118B, respectively. The surface of the sample 1 can be illuminated at the same time by wavelength illumination. Alternatively, a nonlinear fiber or a photonic crystal fiber may be used for the nonlinear optical crystals 119A and 119B and may be built in the fiber 122.

  FIG. 2D shows a fourth configuration example of the dark field illumination unit 101. In FIG. 2D, the illumination light 701 emitted from the illumination light source 112 capable of oscillating multiple wavelengths so that illumination with two different wavelengths can be performed is divided into two illumination lights 701A and 701B with different wavelengths by the dichroic mirror 123. is doing. The divided illumination lights 701A and 701B having different wavelengths illuminate the surface of the sample 1 using different optical filters 113A and 113B, wavelength plates 114A and 114B, and lens groups 118A and 118B, respectively. The surface of the sample 1 can be illuminated with wavelength illumination.

  FIG. 2E shows a fifth configuration example of the dark field illumination unit 101. FIG. 2 (e) illuminates the surface of the sample 1 using illumination light 701 emitted from an illumination light source 112 capable of oscillating multiple wavelengths so that illumination with two different wavelengths is possible, and a plurality of wavelengths different from each other. The surface of the sample 1 can be illuminated by illumination.

  FIG. 2F shows a sixth configuration example of the dark field illumination unit 101. In FIG. 2 (f), the illumination light 701 emitted from the illumination light source 112 is used for the nonlinear optical crystals 119A and 119B using the nonlinear optical crystal switching mechanism 124 so that illumination with two different wavelengths can be performed, and the illumination light 701 is different. The surface of the sample 1 can be illuminated with two different wavelength illuminations that can be converted into wavelengths.

FIG. 2G shows a seventh configuration example of the dark field illumination unit 101. FIG. 2G shows the illumination light 701 emitted from the illumination light source 112 so that two different wavelengths can be illuminated, and the mirror 121A is switched by the optical path switching mechanism 125 to switch the optical path, and the illumination light 701A, By using different nonlinear optical crystals in 701B, the surface of the sample 1 can be illuminated with two different wavelength illuminations.
Note that either one of the nonlinear optical crystals 119A and 119B in FIGS. 2B, 2C, and 2G may not have a nonlinear optical crystal.

  The illumination light source 112 is a laser oscillator, a broadband white lamp, or a light emitting diode. The laser oscillator oscillates, for example, 405 nm, 455 nm, 488 nm, 532 nm visible light (400 nm-800 nm), 400 nm or less ultraviolet light, or 200 nm or less vacuum ultraviolet light. But it can be used. As these selection methods, if a continuous wave laser is used, it is cheap and stable, and a small device can be realized. The wavelength of the illumination light source 112 is not limited to the above wavelength. A long wavelength laser may be used as the illumination light source 112, and the surface of the sample 1 may be irradiated with the short wavelength illumination light 701 after being converted into a second harmonic, a third harmonic, a fourth harmonic, or the like using a nonlinear optical crystal 119. . Further, a laser capable of multi-wavelength oscillation may be used. To detect defects on or near the sample surface, the surface of the sample 1 is irradiated with a short wavelength ultraviolet laser, vacuum ultraviolet laser, or visible short wavelength blue laser to detect defects or crystal defects inside the sample. For this purpose, it is preferable to irradiate the surface of the sample 1 with a visible laser or an infrared laser.

  When sensitivity for detecting defects on the surface of the sample 1 or near the surface of the sample 1 is required, ultraviolet light is used. In this case, the objective lens 105, the vacuum sealing window 107, the half mirror 108, and the imaging optical system 110 are made of synthetic quartz or the like. An ultraviolet region compatible optical element or a reflective optical element may be used. If further sensitivity is required, vacuum ultraviolet light is used. In that case, the objective lens 105, the vacuum sealing window 107, the half mirror 108, and the imaging optical system 110 are optical elements corresponding to the vacuum ultraviolet region such as fused silica or reflective optics. Further, the entire optical path in the optical microscope 6 is installed in a vacuum or in, for example, a nitrogen gas atmosphere in order to prevent vacuum ultraviolet rays from being absorbed along with propagation. Since the purpose is to propagate vacuum ultraviolet rays, the gas to be filled is not limited to nitrogen.

  When the sample 1 is a mirror wafer, P-polarized laser light is used for irradiation of the sample 1, and S-polarized laser light is used when the surface of the sample 1 is covered with a metal thin film. P-polarized light or S-polarized linearly polarized light is used for more efficiently observing scattered light and realizing observation with good S / N. That is, when observing a specular wafer, P-polarized illumination is suitable because S-polarized light has poor scattering ability and the amount of absolute scattered light decreases, resulting in poor efficiency. On the other hand, when observing a metal thin film, etc. If the P-polarized illumination is used, the scattered light from the substrate becomes strong, and minute defects or minute foreign objects cannot be observed, so that the S-polarized illumination is suitable.

  In order to suppress scattered light from the substrate, it is desirable to illuminate the substrate surface at a low elevation angle of about 10 °. The mirror 104 has a mechanism (not shown) that moves together with the objective lens 105 so that the field of the objective lens 105 can be illuminated even when the objective lens 105 moves up and down. Alternatively, the mirror 104 may be a mechanism (not shown) that is independently movable so that the illumination position in the field of view of the objective lens 105 can be changed.

  FIG. 3 shows a configuration example of the optical height detectors 5 and 7. The optical height detectors 5 and 7 include a light source 150 that emits height measurement light, a condensing lens 151 that condenses the height measurement light emitted from the illumination 150, and light collected by the condensing lens 151. The slit 152 illuminated by 1, the image of the light transmitted through the slit 152 as the height measurement light (image of the slit 152) is imaged on the surface of the sample 1, and the height measurement light reflected by the sample 1 And a detector 155 that detects the height measurement light collected by the condenser lens 154 and converts it into an electrical signal. Information on the height measurement light converted into an electrical signal by the detector 155 is sent to the control system 10 to calculate the height. As the detector 155, a two-dimensional CCD or a line sensor, a 2-division or 4-division position sensor is used.

  FIG. 4 shows a configuration example of the optical microscope 6. The optical microscope 6 includes a dark-field illumination unit 101, epi-illumination mirrors 102A and 102B, mirrors 104A and 104B, a sample for irradiating the same spot on the surface of the sample 1 with two lights having different wavelengths emitted from the dark-field illumination unit. The objective lens 105 that collects the light from 1, the height control mechanism 9 (not shown) that adjusts the height of the objective lens, the bright field illumination unit 126 for bright field observation of the sample 1, and the objective lens An imaging optical system 110 that collects the collected light, image sensors 111A and 111B that acquire the collected light, an image processing unit 133 that processes an image acquired by the image sensor 111, and an image acquired by the image sensor 111 The image display unit 134 for displaying the image and the image storage unit 135 for storing the image acquired by the image sensor 111 are appropriately used. The arrangement of the imaging element 111 may be a conjugate position with the sample surface or a conjugate position with the pupil plane of the objective lens.

  The imaging optical system 110 connects lenses 131A and 131B that extract the pupil plane 302C of the objective lens 105, a dichroic mirror 132 that splits the wavelength of the light beam, and images of the branched sample 1 having different wavelengths onto the image sensors 111A and 111B, respectively. The image forming lenses 130A and 130B to be imaged and the spatial distribution optical elements 201A and 201B inserted into the pupil surfaces 302A and 302B of the objective lens taken out by the branched lenses 131A and 131B having different wavelengths are appropriately used. The dichroic mirror 132 may branch the light beam by taking it in and out of the optical axis of the imaging optical system, or may branch the light beam using a mirror for the dichroic mirror.

  As a method of splitting the light beam, there are a polarization branching unit based on a difference in polarization direction, a wavelength branching unit based on a difference in light wavelength, and a spatial branching unit based on a difference in spatial location.

  In this embodiment, a plurality of spatial distribution optical elements having different characteristics are held by the spatial distribution optical element holder as the spatial distribution optical elements 201A and 201B (in the example shown in FIG. 5A, the presence or absence of the spatial distribution optical element 201E). In the example shown in FIG. 5B, four types (201A to 201D) of switchable spatial distribution optical element holders 136A and 136B are inserted into the pupil planes 302A and 302B. Further, the spatial distribution optical element 114 may not be disposed on the optical axis 301 of the imaging optical system 110.

  The height control mechanism 9 (not shown) is connected to the control system 10 (not shown), and the image sensors 111A and 111B are connected to the image processing unit 133.

  The lenses 131A and 131B are used to draw out the pupil plane 302C of the objective lens 105 to the outside and form it inside the imaging optical system 110, and drive the spatial distribution optical element holders 136A and 136B to The spatial distribution optical element 201 selected from among the plurality of spatial distribution optical elements 201A to 201D held by the spatial distribution optical element holders 136A and 136B is inserted on the pupil planes 302A and 302B taken out inside.

  In this embodiment, the lenses 131A and 131B and the imaging lens 130 are a set of three, and the image of the sample 1 is formed on the detection surfaces of the imaging elements 111A and 111B, respectively. In this embodiment, the imaging optical system 110 uses two lenses 131, but one lens 131 may be used and can be selected as appropriate.

  In this embodiment, the bright-field illumination unit 126 illuminates the light captured by the condensing lens 127 on the surface of the sample 1 and the condensing lens 127 that captures light emitted from the bright-field light source 128. A half mirror 129 used for the above is used as appropriate. The ratio of reflection and transmission of the half mirror 129 may be arbitrary. However, when the light intensity of the bright field light source 128 is sufficiently ensured, it is desirable that the scattered light from the defect be guided to the imaging optical system 110 and the image sensor 111, and the bright field illumination unit is movable. If not used, the optical axis 301 may be removed.

  The bright field light source 128 can be a lamp or a laser. In the case of using a laser, the condensing lens 127 is not necessary, and by replacing the half mirror 129 with a dichroic mirror, illumination can be brightened and more scattered light can be guided to the image sensor 111. Alternatively, when performing dark field observation, a mechanism (not shown) for removing the half mirror 129 from the optical axis 301 of the imaging optical system 110 and the objective lens 105 may be provided. In that case, there is an advantage that more scattered light can be guided to the image sensor 111.

  FIG. 5A shows a mechanism for switching the presence / absence of the spatial distribution optical element 201 by moving the spatial distribution optical element 201 inserted into the pupil planes 302A and 302B of the objective lens 105 on and off the optical axis 301 of the detection optical system. is there. This is a mechanism for sliding the spatial distribution optical element holder 136 </ b> A so that the spatial distribution optical element 201 </ b> E is taken in and out of the optical axis of the imaging optical system 110. Although FIG. 5A shows the case where there is one spatial distribution optical element 201E, a plurality of spatial distribution optical elements 201 may be provided.

  FIG. 5B shows a mechanism for switching the spatial distribution optical elements 201A to 202D inserted in the pupil planes 302A and 302B of the objective lens 105 on the optical axis 301 of the imaging optical system 110. The switching mechanism includes a spatial distribution optical element holder 136B in which a plurality of spatial distribution optical elements 201A to 201D having different characteristics are arranged, and a rotation driving unit 137 for rotating the spatial distribution optical element holder 136B. The spatial distribution optical element holder 136B is a mechanism for switching to any one of the plurality of spatial distribution optical elements 201A to 201D in accordance with the type of minute defect to be detected. Here, the case where there are four spatial distribution optical elements is shown, but the present invention is not limited to this, and a necessary number of elements can be appropriately applied.

  The spatial distribution optical element holder shown in FIGS. 5 (a) and 5 (b) is a spatial distribution optical element in order to avoid disturbing an acquired image when performing bright field observation or not using the spatial distribution optical element. The position of the holder 136 is set at a place where the spatial distribution optical element 201 is not installed and observed. Or it switches to the place where the parallel flat glass of the same thickness as the spatial distribution optical element 201 was installed in the spatial distribution optical element holder 136. The reason why the parallel flat glass having the same thickness as that of the spatial distribution optical element 201 is installed is to avoid that the optical path length changes when the spatial distribution optical element 201 is removed and the image of the sample 1 is not formed on the imaging element 111. It is. Alternatively, a mechanism for adjusting the position of the imaging lens 130 or the imaging element 111 that forms an image and forming an image on the imaging element 111 without using parallel flat glass may be used.

Next, the spatial distribution optical element 201 will be described with reference to FIGS.
First, the scattered light simulation and terms required in an example of a method for determining the spatial distribution optical element 201 will be described with reference to FIGS. In the scattered light simulation, the sample 1 is illuminated with a laser as the illumination light 312 obliquely from above, and the light scattered from the minute foreign matter or minute defect placed on the sample 1 is the optical closest to the sample 1 of the imaging optical system. The intensity distribution and polarization distribution of scattered light on the surface closest to the sample 1 of the element (pupil plane 302) is calculated.

The specularly reflected light on the surface of the sample 1 of the illumination light 312 in FIG. A surface that is perpendicular to the sample 1 and includes the illumination light 312 is an incident surface 315.
FIG. 6B shows the intensity distribution in the pupil plane 302. An axis 307 in the intensity distribution indicates an axis in which the illumination incident surface 315 is made to correspond to the pupil plane 302. An arrow 312 indicates the incident direction of the illumination light, and an arrow 313 indicates the regular reflection direction of the illumination light. An axis 314 in the intensity distribution indicates an axis in which a plane perpendicular to the sample 1 and perpendicular to the incident surface 307 is associated with the pupil plane 302. Scattering of the region 317 on the illumination light 312 side divided on the pupil plane 302 by the axis 314 is called backscattering, and scattering of the region 316 on the specularly reflected light 313 side divided on the pupil plane 302 by the axis 314 is called forward scattering.
FIG. 6C shows an example of the polarization direction of light on the pupil plane 302. Here, light that vibrates in a direction parallel to the incident surface is X-polarized light 318, light that vibrates in the direction perpendicular to the X-polarized light is Y-polarized light 319, light that vibrates in the radial direction on the pupil plane 302 is radial polarized light 320, and pupil plane The light oscillating concentrically on 302 is referred to as azimuth polarized light 312.

  FIG. 7A shows the distribution pNp (r, θ) and azimuth polarized light (S polarized light) component of the radial polarized light (P polarized light) component of the scattered light from the irregularities on the substrate surface, calculated using the scattered light simulation. The distribution pNs (r, θ), the X-polarization distribution pNx (r, θ), and the Y-polarization distribution pNy (r, θ) are shown in FIG. 7B. The radial polarization (P-polarization) of the scattered light from the defect is shown in FIG. A component distribution pSp (r, θ), an azimuth polarization (S polarization) component distribution pSs (r, θ), an X polarization distribution pSx (r, θ), and a Y polarization distribution pSy (r, θ) are shown. These distributions are obtained by using the Stokes vectors obtained by obtaining the Stokes vectors of scattered light from minute irregularities on the surface of the substrate and the Stokes vectors of scattered light from foreign matter to be detected with high sensitivity by the scattered light simulation. It is done. In addition, the polarization | polarized-light to obtain | require is not restricted to these, The linearly polarized light in which the angle of polarization inclined in the range of (pi) to-(pi) may be sufficient, and elliptical (circle) polarized light may be sufficient.

  FIG. 7A shows the intensity distribution pNp of the radially polarized light, the intensity distribution pNs of the azimuth polarized light, the intensity distribution pNx of the X polarized light, and the intensity distribution pNy of the Y polarized light of the scattered light (illumination wavelength 405 nm) from the minute irregularities on the substrate surface. . FIG. 7B shows a radial polarization intensity distribution pSp of scattered light from a spherical foreign substance having a diameter of 18 nm, an azimuth polarization intensity distribution pSs, an X polarization intensity distribution pSx, and a Y polarization intensity distribution pSy. An axis 307 in the intensity distribution of each polarization indicates an axis in which the incident plane of illumination corresponds to the pupil plane 302. An arrow 312 indicates the incident direction of the illumination light, and an arrow 313 indicates the regular reflection direction of the illumination light.

  In the intensity distribution of each polarized light in FIG. 7, the region 308 is a region where the scattered light intensity is strong, the region 309 is a region where the scattered light intensity is slightly strong, the region 310 is a region where the scattered light intensity is slightly weak, and the region 311 is the weak scattered light intensity. Although the regions are shown, these indicate the relative relationships of the intensities in the same distribution, and even in the same region between the distributions, they do not necessarily indicate the same intensity (for example, the region 308 of the intensity distribution of radial polarization). And the X-polarized light intensity distribution region 308 do not necessarily show the same intensity).

  According to the scattered light intensity distribution of each polarized light shown in FIG. 7A, scattered light from minute irregularities on the substrate surface is strong on the illumination incident 312 side (backscattering), and the backscattered polarized light is strong in radial polarization. I understand. Further, according to the scattered light intensity distribution of each polarized light shown in FIG. 7B, it can be understood that the scattered light from the minute foreign matter is almost isotropic and the radial polarized light is strong. Therefore, the ratio of the scattered light from the substrate surface to the scattered light from the foreign material can be increased by appropriately setting and arranging the polarizing filter based on these results, and a high S / N defect detection is possible. It becomes.

  FIGS. 8A and 8B show the scattered light intensity distribution of the scattered light from the defect calculated using the scattered light simulation. These distributions are obtained by a scattered light simulation. In addition, the polarization | polarized-light to obtain | require is not restricted to these, The linearly polarized light in which the angle of polarization inclined in the range of (pi) to-(pi) may be sufficient, and elliptical (circle) polarized light may be sufficient.

  FIG. 8A shows a scattered light intensity distribution fSA (r, θ) of scattered light from crystal defects. FIG. 8B shows a scattered light intensity distribution fSB (r, θ) of scattered light from a spherical defect having a diameter of 18 nm. An axis 307 in the intensity distribution of each polarization indicates an axis in which the incident plane of illumination corresponds to the pupil plane 302. An arrow 312 indicates the incident direction of the illumination light, and an arrow 313 indicates the regular reflection direction of the illumination light.

  In the intensity distribution of each polarization in FIG. 8, the region 308 is a region where the scattered light intensity is strong, the region 309 is a region where the scattered light intensity is slightly strong, the region 310 is a region where the scattered light intensity is slightly weak, and the region 311 is the weak scattered light intensity. Although the regions are shown, these indicate the relative relationships of the intensities in the same distribution, and even in the same region between the distributions, they do not necessarily indicate the same intensity (for example, the regions in FIG. 8A). 308 and the region 308 in FIG. 8B do not necessarily show the same intensity).

According to the scattered light intensity distribution shown in FIG. 8A, it is understood that the scattered light from the crystal defect is strong on the illumination incident 312 side (backscattering). Further, according to the scattered light intensity distribution shown in FIG. 8B, it can be seen that the scattered light from the minute foreign matter is almost isotropic. As described above, the intensity distribution of scattered light differs depending on the defect shape. Therefore, based on these results, by appropriately setting and arranging a polarizing filter, scattering from the substrate surface with respect to scattered light from foreign matter The light ratio can be increased, and a high S / N defect can be detected. The defects for which the scattered light intensity distribution is obtained are not limited to these, and may be a defect shape to be detected such as a polishing defect or a stacking fall. Further, not only the scattered light intensity distribution but also the intensity distribution for each polarization of each defect and the intensity distribution for each polarization component of the scattered light intensity from each defect as shown in FIG. The desired polarized light is not limited to these, and may be linearly polarized light whose polarization angle is in the range of π to −π, or may be elliptical (circular) polarized light.
Optical characteristics of various spatial distribution optical elements 201 to be described below or arrangement of phase shifters, inclination of slow axis and inclination of fast axis of wave plate, optical rotation direction by polarization direction control device constituting spatial distribution optical element 201 The determination method of the light shielding region in the spatial filter, the transmission polarization axis direction of the polarizer, the transmittance of the neutral density filter, the applied voltage to the liquid crystal element, the applied voltage to the magneto-optic modulator, the optical characteristics of the photonic crystal, etc. The determination is made based on the scattered light intensity distribution obtained by the scattered light simulation or actual measurement described with reference to FIGS.

Next, a method for determining the optical characteristics of the spatial distribution optical element 201 and its effect will be described.
A method example for determining the polarization transmission axis distribution h (r, θ) and the effect of using the polarizer 231 will be described.
First, the scattered light intensity distribution fS (r, θ) from the minute defect or minute foreign matter to be detected with high sensitivity by the scattered light simulation, the radial polarization distribution pSp (r, θ) of the scattered light, and the S polarization distribution pSs ( r, θ), and the scattered light intensity distribution fN (r, θ) from the minute irregularities on the substrate surface, the P-polarized light distribution pNp (r, θ) and the S-polarized light distribution pNS (r, θ) of the scattered light. Ask.

  The polarization transmission axis direction distribution h (r, θ) of the spatial distribution optical element 201 is a polarization axis distribution that most blocks scattered light from minute irregularities on the substrate surface, that is, h (r) that minimizes the wrinkle of (Equation 1). , Θ), or a polarization axis distribution that most transmits scattered light from a minute defect or minute foreign matter, that is, h (r, θ) that maximizes Λ in (Equation 2), or scattered light from minute irregularities on the substrate surface. Is determined as h (r, θ) that maximizes Ω in (Equation 3).

これは、瞳面３０２上の領域を複数の領域に分割し、その分割された各領域それぞれにおいて、（数１）中のΠが最小化するｈ（ｒ，θ）、または（数２）のΛを最大化するｈ（ｒ，θ）、または（数３）のΩを最大化するｈ（ｒ，θ）をそれぞれ決定してもよい。

This is because the region on the pupil plane 302 is divided into a plurality of regions, and h (r, θ) in (Equation 1) or (Equation 2) in which the wrinkles in (Equation 1) are minimized in each divided region. H (r, θ) that maximizes Λ or h (r, θ) that maximizes Ω in (Equation 3) may be determined.

  Thus, the scattered light from the sample 1 can be suppressed by selecting the vibration direction of light, that is, the polarization direction. Therefore, by appropriately setting and arranging a polarizer on the pupil surface 302 or in the vicinity of the pupil surface 302, the ratio of the scattered light from the substrate surface to the scattered light from the foreign matter can be increased, and a high S / N ratio can be obtained. Defect detection is possible.

  FIGS. 9A and 9B show spatial distribution optical elements 201A and 201B in which the polarization transmission axis direction of the spatial distribution optical element 201 is uniform. FIG. 9A has a polarization transmission axis 327A in the Y direction, and transmits Y-polarized light. FIG. 9B has a polarization transmission axis 327B in the radial direction, and transmits the radially polarized light.

  FIGS. 9C and 9D show spatial distribution optical elements 201C and 201D in which the polarization transmission axis direction of the spatial distribution optical element 201 is distributed differently depending on the region.

  9E, 9F, 9G, and 9H, the polarization transmission axis direction of the spatial distribution optical element 201 divides the spatial distribution optical element into regions and is uniform in each of the divided regions. Spatial distribution optical elements 201E, 201F, 201G, and 201H having various polarization transmission axis directions are shown. In FIGS. 9F and 9H, the spatial distribution optical element 201 using the polarization transmission axis 327E parallel to the region into which the polarization transmission axis that transmits the radially polarized light is obtained in the approximately divided region is used. ing.

  The spatially-distributed polarizing element 201 in which the transmission polarization axis direction is distributed in the same plane is a combination of linear polarizers, a photonic crystal, a wire grid polarizer, a combination of a liquid crystal and a polarizer, or a magneto-optic modulator and a polarization. Realized by a combination of children. Here, the photonic crystal is an optical element having a fine structure having a refractive index different from that of the light wavelength, and the wire grid polarizer has optical anisotropy by periodically arranging conductive thin wires. Polarizing element.

The method for determining the optical rotation angle η (θ) of the optical rotator that controls the polarization direction of the scattered light and the effect of using the optical rotator will be described.
FIGS. 10A, 10B, and 10C show examples of effects obtained by controlling the polarization direction of scattered light. A point 326 in FIG. 10 indicates a point projected on the optical axis pupil image of the imaging optical system. A point 322 indicates an arbitrary point A on the pupil plane 302, and a point 323 indicates an arbitrary point B that is the same distance as the distance from the point 326 to the point 322 on the pupil plane. An arrow 324 indicates the vibration direction of light at a certain point 322, and an arrow 325 indicates the vibration direction of light at a certain point 323, respectively. The image on the pupil plane 302 can be Fourier transformed to obtain a sample image. When the image on the pupil plane 302 is Fourier-transformed, the light on the points 322 and 323 is superimposed. FIG. 10A shows the vibration direction of light at certain points A and B on the pupil plane 302 of the scattered light at a certain moment. The light vibration direction 324A and the point B at any point A are shown in FIG. The light vibration directions 325A are in different directions. FIG. 10B shows the vibration of light by inserting an optical rotator so that the vibration direction 324A of the light at a certain point 322 and the vibration direction 325A of the light at a certain point 323 are the same. As an example in which the direction is controlled, FIG. 10C shows that the vibration direction 324A of light at a certain point 322 and the vibration direction 325A of light at a certain point 323 are 180 degrees different from each other. An example in which an optical rotator is inserted and the vibration direction of light is controlled is shown. When the pupil image 302 shown in FIG. 10B is Fourier-transformed, the light in the vibration direction 324B of light at a certain point 322 and the light in the vibration direction 325B of light at a certain point 323 are superimposed. As a result, the strengthening peak intensity increases. On the other hand, when the pupil image 302 shown in FIG. 10C is Fourier-transformed, the light in the vibration direction 324B of the light at a certain point 322 and the light in the vibration direction 325B of the light at a certain point 323 are Overlapping each other cancels each other and weakens each other.

  The optical rotation angle η (θ) of the optical rotator is an optical rotation angle η at which light oscillating in the h (r, θ) direction cancels out by superimposing h (r, θ) that minimizes the wrinkles in (Equation 1). The angle of rotation η (θ) or (number) in which the light oscillating in the h (r, θ) direction is strengthened by superimposing h (r, θ) that maximizes Λ in (θ) or (Equation 2). The rotation angle η (θ) at which the light oscillating in the h (r, θ) direction is strengthened may be determined by superimposing h (r, θ) that maximizes Ω in 3). This is because the region on the pupil plane 302 is divided into a plurality of regions, and h (r, θ) that minimizes the wrinkles in (Equation 1) is superimposed on each of the divided regions. Vibration in the h (r, θ) direction by superimposing the optical rotation angle η (θ) that cancels light oscillating in the (r, θ) direction, or h (r, θ) that maximizes Λ in (Equation 2) Rotating angle η (θ) where the light to be strengthened or h (r, θ) that maximizes Ω in (Equation 3) is superposed so that the light oscillating in the h (r, θ) direction is strengthened Each η (θ) may be determined.

  In this way, by changing the vibration direction of light, that is, the polarization direction, the scattered light from the sample 1 can be suppressed or the scattered light from the defect can be amplified. Therefore, by appropriately setting and arranging an optical rotator on or near the pupil plane 302, the ratio of the scattered light from the substrate surface to the scattered light from the foreign matter can be increased, and a high S / N ratio can be obtained. Defect detection is possible.

Next, an example of the effect when a half-wave plate is used for the configuration of the spatial distribution optical element 201 will be described taking radial polarization as an example.
FIG. 11 is a diagram for explaining the change in polarization when the half-wave plate 203 is applied near the pupil plane 302. By controlling the electric field vector with the half-wave plate 203 in the configuration of the spatial distribution optical element 201, the scattered light from the sample is suppressed, and the scattered light from the defect or foreign matter desired to be detected is strengthened from the sample. The ratio of the scattered light intensity of the defect or foreign matter to the scattered light intensity of can be increased.

  As a specific example, the half-wave plate 203 is applied to the radial polarization 331A on the pupil plane 302. In FIG. 11, reference numeral 203 denotes a half-wave plate, arrow 303 denotes a slow axis of the half-wave plate 203, and arrow 304 denotes a fast axis of the half-wave plate 203. As shown in FIG. 7 (b), the scattered light from the minute foreign matter is close to radial polarization. Therefore, in the direction symmetrical to the axis 307 corresponding to the incident optical axis of the illumination on the pupil plane, the oscillation direction of the electric field vector. Are opposed to each other, and strength reduction may occur due to superposition. This is an example in which a half-wave plate 203 is applied in order to suppress this strength reduction. With respect to the electric field vector 329A inclined at an angle 328 with respect to the axis 307 corresponding to the incident optical axis of the illumination on the pupil plane, the half-wave plate 203 in the state 203A is set to 1 at an angle 328 with respect to the X axis. By arranging the half-wave plate 203B with the angle 330 phase axis 303 of / 2 inclined, the electric field vector 329B parallel to the axis 307 corresponding to the incident optical axis of the illumination on the pupil plane can be obtained. . Thus, by arranging the spatial distribution optical element 201 using a half-wave plate in the pupil plane 302 or in the vicinity of the pupil plane, the illumination optical axis of illumination is symmetrical with respect to the axis 393 corresponding to the pupil plane. The direction of vibration of the electric field vector in a certain direction is parallel, the polarization on the pupil plane is 331B, and the peak intensity of scattered light due to superposition can be increased.

Further, before using the polarizer or optical rotator, the effect of the optical rotator or polarizer can be enhanced by converting elliptically or circularly polarized light into linearly polarized light using a phaser.
As a specific example, FIG. 12 shows an application example of the quarter-wave plate 204 for the elliptically polarized light 331 on an arbitrary pupil plane. In FIG. 12, 204 indicates a quarter wavelength plate, arrow 303 indicates a slow axis of the quarter wavelength plate 204, and arrow 304 indicates a fast axis of the quarter wavelength plate 203, respectively. The quarter wavelength plate 204A is tilted at an angle of 332 with respect to the axis 307 corresponding to the entrance plane of illumination light corresponding to the pupil plane 307, the long axis of the elliptically polarized light 329C and the slow axis of the short axis of the elliptically polarized light 329C. The elliptical polarized light 329C can be changed to the linearly polarized light 329D by inclining the quarter wavelength plate 204A at an angle 332 and arranging the quarter wavelength plate 204B so that 303 matches. The linearly polarized light converted from the elliptically polarized light is selectively transmitted by the polarizer of the spatial distribution optical element 201, or the polarization direction of the linearly polarized light converted from the elliptically polarized light is controlled by the optical rotator of the spatial distribution optical element 201. It is possible to suppress a decrease in the amount of scattered light from the defect or foreign matter caused by the distributed optical element 201.

Next, an example of the effect when using liquid crystal in the configuration of the spatial distribution optical element 201 will be described.
First, an example in which a polarization direction control device 205 using liquid crystal instead of the half-wave plate 203 in FIG. 11 and the quarter-wave plate 204 in FIG. 12 is arranged will be described with reference to FIGS. By controlling the voltage applied to the liquid crystal or the direction of rubbing applied to the alignment film, a precise polarization direction control that cannot be realized by the half-wave plate 203 or the quarter-wave plate 204 using a crystal of crystal. Is possible. Alternatively, a plurality of polarization direction control devices 205 may be arranged two-dimensionally to form one spatial distribution optical element 201. At that time, the plurality of polarization direction control devices 205 constituting the spatial distribution optical element 201 can respectively control the optical rotation angle, and can adjust the optical rotation angle in time according to the scattered light distribution.

  FIG. 13A shows a spatial distribution optical element 201 using a polarization direction control device 205A configured by appropriately using the liquid crystal 215, the alignment films 216 and 217, and the electrodes 218A and 218B sandwiched between the outermost layers 214A and 214B. An example is shown, and the arrangement of liquid crystal molecules in the liquid crystal 215 is controlled by applying a voltage between the electrodes 218A-218B to control the polarization direction.

  FIG. 14A is a diagram for explaining the change of the electric field vector by the polarization direction control 205, and the electric field vector 334 at an arbitrary point at an arbitrary point 333 is changed to an electric field vector 335 using the polarization direction control device 205. Here, an angle difference 336 between the electric field vector 334 and the electric field vector 335 is defined as an optical rotation angle.

Here, the liquid crystal is an intermediate state between the liquid and the crystal, and is a crystal having both the fluidity of the liquid and the anisotropy of the crystal, and the liquid crystal has no chirality with the optical rotatory liquid crystal having the chirality. There is a liquid crystal without optical rotation.
In the case of a liquid crystal having chirality, the liquid crystal molecules 219A in contact with the substrate as illustrated in FIG. The direction of rotation is determined by the chirality of the liquid crystal. When light is transmitted through the liquid crystal 215A, the polarization direction of the light rotates in accordance with the liquid crystal molecules 220 and the polarization state can be changed.

  When a voltage is applied to the liquid crystal 215A having this chirality by using the power source 221, the liquid crystal molecules arranged in the horizontal state rise up vertically to become like 215B, and as the liquid crystal molecules rise, the optical rotation is lost, and the liquid crystal 215B has an optical rotation. There is no. The polarization direction can be changed by controlling the rising angle of the liquid crystal molecules according to the magnitude of the applied voltage.

  In addition, when there are liquid crystal molecules in the middle between the state parallel to the alignment film and the state perpendicular to the alignment film, the scattered light that has passed through the liquid crystal can be elliptically polarized light instead of linearly polarized light. By combining with a four-wave plate, the light that has been elliptically polarized by the liquid crystal may be converted to linearly polarized light. In addition, when optical rotation is unnecessary, such as during bright field observation, ON or OFF of optical rotation can be easily selected by applying or not applying voltage.

  FIG. 13B shows a spatial distribution optical system using a polarization controller 205B using a liquid crystal that rubs the alignment film 214A, controls the alignment of liquid crystal molecules in the liquid crystal 215 by applying a voltage, and controls the polarization direction. 5 is a diagram showing an example of an element 201. FIG.

  When there is no applied voltage, the liquid crystal 215 is aligned along the rubbing direction of the alignment film 214A, and the alignment of the liquid crystal molecules in the liquid crystal 215 is parallel between the alignment film 216 and the outermost layer 214B without rotating. When a voltage is applied between the electrodes 218A and 218B, the liquid crystal molecules in the vicinity of the electrodes rotate in the plane, but the liquid crystal molecules in the vicinity of the alignment film 216 are aligned along the rubbing direction. When the liquid crystal 215 is rotated and light is transmitted through the liquid crystal 215 in this state, the polarization direction of the liquid crystal molecules and thus the polarization direction of the light is rotated, and the polarization state can be changed. The polarization direction can be changed by controlling the rotation angle of the liquid crystal molecules in the plane parallel to the alignment film 216 by the magnitude of the applied voltage. When the scattered light that has passed through the liquid crystal can be elliptically polarized light instead of linearly polarized light, the light that has been elliptically polarized by the liquid crystal may be made linearly polarized light in combination with a quarter-wave plate. In addition, when optical rotation is unnecessary, such as during bright field observation, ON or OFF of optical rotation can be easily selected by applying or not applying voltage.

  FIG. 13C shows a polarization direction control using a liquid crystal that controls the polarization direction by controlling the alignment of the liquid crystal molecules in the liquid crystal 215 according to the rubbing direction of one or both of the alignment films 216 and 217. It is a figure which shows an example of the spatial distribution optical element 201 using the apparatus 205B.

  In FIGS. 13B and 13C, by selecting the rubbing direction, it is possible to create a distributed wavelength plate that can realize a polarization state desired to be obtained precisely. Here, the rubbing direction is a direction direction of rubbing treatment in which the alignment film is rubbed with a cloth wound around a roller, and liquid crystal molecules are arranged in parallel with the rubbing direction.

  Next, FIG. 15 shows a liquid crystal 215 of a polarization direction control device using a chiral liquid crystal capable of controlling an optical rotation angle by using an electrode. A plurality of electrodes for applying a voltage to the liquid crystal 215 are used, and the voltage to be applied is controlled at each location of the liquid crystal 215 in accordance with the desired polarization. FIGS. 15A and 15B show an example in which a single liquid crystal 215a having optical rotation and no division is used. When the liquid crystal 215 is not divided, the thickness of the liquid crystal 215 is determined so that an optical rotation angle in the range of 2π to 0 or 0 to −2π is obtained.

  An example of the voltage applied to the liquid crystal is that the scattered light from the minute foreign matter is close to radial polarization, and therefore, in the direction symmetric with respect to the axis 307 corresponding to the incident optical axis of the illumination on the pupil plane, the oscillation direction of the electric field vector Since the intensity may decrease due to overlapping, the voltage applied to each electrode may be different so that the polarization direction can be obtained by adjusting the direction of the electric field vector to suppress the decrease in the peak intensity of scattered light from foreign matter. To do. Alternatively, the voltage of each electrode is controlled so as to obtain a polarization direction that weakens scattered light on the sample surface. As a result, the ratio of the peak intensity of the scattered light from the foreign material to the peak intensity of the scattered light on the sample surface can be increased.

  Further, the liquid crystal 215 can be divided into several parts, a voltage is applied to each of the divided liquid crystals, the direction of liquid crystal molecules in the liquid crystal can be controlled, and the optical rotation direction can be controlled. FIG. 15C shows an example in which the liquid crystal is divided into two at an angle. The ranges of the optical rotation controllable angles of the two-divided liquid crystals 215b and 215c in FIG. 15C may be π to 0 and 0 to −π, respectively. FIG. 15D shows an example in which the liquid crystal 215 is divided into four at an angle. The maximum value and combination of the optical rotation controllable angles of the liquid crystal 215 are determined according to the number of divisions. By dividing, it is possible to control each divided area, so that the optical rotation angle can be controlled more precisely as a whole. The number of divisions and the division method are not limited to this, and can be set as appropriate according to the scattered light distribution.

  Further, the liquid crystal 215 is not limited to a single layer, and may be used by being stacked on a plurality of layers. When a plurality of layers are stacked, a voltage may be applied to each layer, or a voltage may be applied together. FIGS. 15E, 15 </ b> F, and 15 </ b> G are examples showing a case where a liquid crystal that is divided into two by an angle and stacked in two layers is used. FIG. 15E shows a division method. FIG. 15E shows the upper liquid crystal, and FIG. 15F shows the lower liquid crystal. In this example, the upper liquid crystals 215i and 215h are liquid crystals having a positive optical rotation 337d, and the lower liquid crystals 215k and 215j are liquid crystals having a negative optical rotation 337e. In both the upper layer and the lower layer, the incident optical axis of illumination is divided into two with respect to the axis 307 corresponding to the pupil plane, and different voltages can be applied to the liquid crystals 215i, 215h, 215k, and 215j. For example, the optical rotation in the positive direction is made upward in FIG. 15C with respect to the axis 307 in which the incident optical axis of illumination corresponds to the pupil plane, and the incident optical axis of illumination corresponds to the pupil plane. When it is desired to obtain the optical rotation in the negative direction on the lower side in FIG. 15C with respect to the axis 307, it is sufficient to apply a voltage to the liquid crystals 215i and 215j and not apply a voltage to the liquid crystals 215h and 215k. By superposing liquid crystals in a plurality of layers, it becomes possible to control the optical rotation angle more precisely.

  Here, since the liquid crystal 215 of the polarization control device of FIG. 15A uses a liquid crystal having optical rotation, the alignment films 216 and 217 of the polarization control device of FIG. Or an alignment film without rubbing is used for either of them.

  FIG. 16 shows an example in which radial polarization is converted to X polarization using a polarization control device using liquid crystal having no optical rotation and having a thickness adjusted to have a function of a quarter-wave plate. Show. Since the scattered light of minute foreign matter is close to radial polarization, the direction of vibration of the electric field is opposite in the direction symmetric with respect to the axis 307 corresponding to the incident optical axis of the illumination on the pupil plane, and the intensity is reduced by superposition. In order to suppress this, a liquid crystal having no optical rotation 337 is used. In the case of this polarization control device, when no voltage is applied to the liquid crystal 215m, the optical rotation direction can be changed from radial polarization to X polarization, while when the voltage is applied, the optical rotation direction can be controlled. If exceeded, the optical rotation will be lost. Here, the lower alignment film 217b is rubbed so as to obtain the desired optical rotation angle, and the upper alignment film 216b is used without being rubbed.

Next, a polarization control device using liquid crystal that is selected for an optical rotation angle at which no voltage is applied to the liquid crystal and the rubbing direction to the alignment film is desired will be described with reference to FIG. In the liquid crystal without chirality, liquid crystal molecules are aligned along the rubbing direction of the upper alignment film 216 and the lower alignment film 217. FIG. 17A shows how liquid crystal molecules are arranged when the rubbing direction of the lower alignment film 217 and the rubbing direction of the upper alignment film 216 are different by π / 2. In the case of the alignment film arrangement shown in FIG. 17A, liquid crystal molecules are aligned along the alignment film, and an optical rotation angle of π / 2 is obtained.
Accordingly, the polarization controller by selecting the rubbing direction of the upper alignment film 216 and the rubbing direction of the lower alignment film 217 is a half-wave plate having a spatially different distribution of the slow axis direction and the fast axis direction. It is possible to realize a more precise polarization control device than the realization of.

17 (b), (c), (d), (e), (f), (g), and (h) are directions of polarization for obtaining the rubbing directions of the upper alignment film 216 and the lower alignment film 217. It is a figure which shows the example of the polarization direction control apparatus performed according to. 17 (c) and 17 (d) show the rubbing direction of the alignment film when the radial polarization shown in FIG. 17 (b) is converted into X polarization (FIG. 17 (g)) and Y polarization (FIG. 17 (h)). Also shown in FIGS. 17 (e) and (f).
Here, the control of the polarization direction by the rubbing direction will be described with reference to FIG. Since the scattered light of minute foreign matter is close to radial polarization, the direction of vibration of the electric field is opposite in the direction symmetric with respect to the axis 307 corresponding to the incident optical axis of the illumination on the pupil plane, and the intensity is reduced by superposition. In order to suppress this, an example of a polarization control device using liquid crystal having no optical rotation 337 is shown. The rubbing direction of the alignment films 216 and 217 is determined so as to be equal to the optical rotation angle at which an inclination difference between the rubbing directions of the alignment films 216 and 217 is desired at an arbitrary point on or near the pupil plane. An electric field vector 339A at an arbitrary time at an arbitrary point 338A is changed into an electric field vector 339B using the polarization direction controller 205. At points 338A and 338B, the angle difference in the rubbing direction between the alignment films 216A and 217A is π / 4. Therefore, the electric field vector 340A changes its π / 4 polarization direction, and X-polarized linearly polarized light 340B is obtained. In FIGS. 17D, 17E, and 17F, the polarization direction is controlled by the rubbing direction in the same procedure.

Next, an example of an effect when the polarization control device using the magneto-optical effect is used for the configuration of the spatial distribution optical element 201 will be described.
18A shows a spatial distribution in which a polarization direction control device using a transparent magnetic material using a magneto-optic effect is arranged instead of the half-wave plate 203 in FIG. 11 and the quarter-wave plate 204 in FIG. An example of the optical element 201 is shown. In this example, the polarization direction is controlled using Faraday rotation by controlling the magnetization direction of the transparent magnetic body 222 sandwiched between the transparent substrates 223 and 224. A plurality of polarization direction control devices 205 using a magneto-optic modulator may be arranged two-dimensionally to form one spatial distribution optical element 201. At that time, the plurality of polarization direction control devices 205 constituting the spatial distribution optical element 201 can control the optical rotation angle in accordance with the scattered light distribution by controlling the optical rotation angle respectively.

  FIGS. 18B and 18C show an example using a transparent magnetic material obtained by dividing the incident optical axis of illumination symmetrically with respect to an axis 307 corresponding to the pupil plane. By magnetizing the transparent magnetic bodies 222A and 222b arranged symmetrically with respect to the axis 307 corresponding to the incident optical axis of the illumination on the pupil plane, the incident optical axis of the illumination on the pupil plane 112 is magnetized. A symmetrical optical rotation angle 339 can be obtained with respect to the corresponding axis 307. The direction of vibration of the electric field vector in a direction symmetric with respect to the axis 307 corresponding to the incident optical axis of the illumination on the pupil plane opposes, and when the scattered light is superimposed, the intensity of Y-polarized light in the scattered light of the defect decreases. Although it occurs, it is possible to suppress a decrease in the peak intensity of the scattered light by selecting the magnetization direction so that the direction of the electric field vector is aligned in one direction by the polarization direction control device using the magneto-optical effect. The direction of magnetization need not be parallel to the pupil plane. Moreover, the division | segmentation number of a transparent magnetic body may be single or plural. Furthermore, the number of layers of the transparent magnetic material is not limited to a single layer, and a plurality of layers may be used in an overlapping manner.

  The magnetization direction is controlled by applying an external magnetic field, applying stress to the crystal by a piezoelectric actuator, or applying an electric field, applying an external magnetic field, and applying stress to the crystal by a piezoelectric actuator. When optical rotation is not required, such as during bright field observation, optical rotation can be easily removed by applying no stress, no electric field, or no external magnetic field.

  The polarization direction of the scattered light from the foreign matter is substantially equal to the vibration direction and the intensity with respect to the axis 307 corresponding to the illumination incident surface on the pupil plane 302, so that the illumination incident optical axis corresponds to the pupil plane 302. When some of the light cancels each other by causing the light in the vibration direction symmetric with respect to the axis 307 to interfere, the scattered light from the foreign matter can be obtained by applying the phase shifter to the pupil plane 302 and the vicinity of the pupil plane 302. It becomes possible to increase the strength. In addition, a phase shifter can be applied to suppress the scattered light intensity from the substrate surface, or to suppress the scattered light intensity from the substrate surface and increase the scattered light intensity from the foreign matter.

  Here, the effect of the phase shifter will be described in more detail with reference to FIG. 19, taking radial polarized light as an example. As a result of the scattered light simulation, the scattered light of the minute foreign matter is close to the radial polarized light as shown in the radial polarization distribution of FIG. 7B. The strength is decreasing. A phase shifter was applied to radial polarized light for the purpose of suppressing the intensity drop.

  FIG. 19A shows a phase difference between a plane perpendicular to a substrate including an axis 307 in which the incident optical axis of illumination corresponds to the pupil plane in the vicinity of the pupil plane of the optical microscope or a plane on the pupil plane corresponding thereto. This is an application example of radial polarization when a phase shifter 202a for creating π is arranged. 19B, a phase shifter 202b that creates a phase difference π is arranged in the upper and lower regions in FIG. 19 with an axis 307 corresponding to the incident optical axis of the illumination on the pupil plane as a boundary near the pupil plane of the optical microscope. This is an example of application of radial polarization in the case of the above. In FIG. 19A, it is conceivable that the components in the Y direction of certain radial polarizations 341a and 341b cancel each other, and if the light of certain radial polarizations 341a and 341b interfere with each other, the peak of the scattered light intensity in the Y direction decreases. For this radially polarized scattered light, a phase shifter that creates a phase difference π in the upper and lower regions in FIG. 19 with an axis 307 in the vicinity of the pupil plane of the optical microscope associating the incident optical axis of illumination with the pupil plane. When arranged, the state becomes 341c and 341d, the components in the Y direction of 341c and 341d strengthen each other, and the peak can be increased.

  In FIG. 19B, it is conceivable that the components in the X direction of certain radial polarizations 341e and 341f cancel each other, and if the light of certain radial polarizations 341e and 341f interfere with each other, the peak of the scattered light intensity in the X direction decreases. In addition, a phase shifter that creates a phase difference π in the left and right regions in FIG. 19 with an axis 314 in the vicinity of the pupil plane of the optical microscope as the boundary between the illumination incident plane and the plane perpendicular to the sample surface on the pupil plane. When arranged, the state becomes 341g and 41h, and the components in the X direction of 341g and 341h strengthen each other, and the peak can be increased when they interfere with each other.

Next, an effect of using the spatial filter and an example of a method for determining the light shielding region g (r, θ) of the spatial filter will be described.
As a spatial filter, the ratio of the amount of scattered light on the substrate surface and the amount of scattered light on the substrate surface is derived by scattered light simulation or actual measurement, the ratio is larger than a certain threshold value, and the scattered light from the defect or foreign material and the sample are transmitted. By shielding the area where the ratio of scattered light from the surface is smaller than a certain threshold, by removing the area where the ratio of the scattered light from the defect or foreign material and the scattered light from the sample surface is small, the defect or the entire pupil plane The ratio of the amount of scattered light from the foreign matter to the amount of scattered light from the sample can be increased.

As a method of determining the light shielding region g (r, θ) of the spatial filter, for example, there is a method of optimizing the light shielding region g (r, θ) so as to maximize the Ψ shown in (Equation 4).

  Further, the region on the pupil plane 302 is divided into a plurality of regions, and in each of the divided regions, a region where the ratio of the scattered light from the sample to the scattered light from the foreign material is small is a light shielding region g (r, θ). May be determined.

  Thus, by appropriately setting and arranging the spatial filter, the ratio of the scattered light from the sample to the scattered light from the foreign material can be increased, and a high S / N defect can be detected. A spatial filter intended to remove diffracted light from the pattern may be used in the configuration of the spatial distribution optical element 201.

FIG. 20 shows an example of the shape of the spatial filter. The example of the spatial filter shown in FIG. 20 is arranged on or near the pupil plane, and the black portion 342 is a light shielding portion and the white portion 343 is an opening portion, and indicates a light beam transmission portion.
FIG. 20A is an example of the spatial filters 225A to 225D in which a region where the scattered light intensity from the sample is strong is shielded for the purpose of increasing the ratio of the scattered light from the sample to the scattered light from the defect or foreign matter. FIG. 20B shows an example of the shape of the spatial filter 225 arranged near the pupil plane for the purpose of excluding scattered light resulting from the pattern. Here, an example in which nine small openings are provided in the central part (225E), an example in which a large opening is provided in the central part (225F), an example in which an intermediate opening is provided in the central part (225G), and two small openings. Although the example (225H) provided individually was shown, it is not restricted to this, You may provide an opening in a vertical or horizontal stripe in combination with these, and a numerical aperture, a shape, and a magnitude | size can be set suitably. Since the image on the pupil plane represents the angle component of the diffraction and scattered light of the object to be inspected, the diffraction of the object to be inspected and the scattered light are discarded by determining how many openings are provided at which position. Selection can be made.

  As the spatial filter 225, a light shielding mask, a combination of a liquid crystal and a polarizer, a combination of a magneto-optical modulator and a liquid crystal, or a MEMS may be used. A combination of liquid crystal and a polarizer, or a combination of a magneto-optic modulator and a polarizer, or a two-dimensional array shutter using MEMS may be able to control a light shielding region depending on the presence or absence of an applied voltage.

FIG. 21 shows a configuration example of a spatial filter that can control the light shielding region.
FIG. 21A shows an example of the spatial filter 225 configured by a combination of the liquid crystal and the polarizer shown in FIG. In this configuration, an optical rotator using the liquid crystal shown in FIG. 13A is inserted between the polarizer 226 and the polarizer 227. The liquid crystal shown in FIG. 21 is a liquid crystal capable of obtaining an optical rotation angle of π / 2 when no voltage is applied. FIG. 21B is a diagram schematically illustrating a configuration example 228 in which the polarizer is arranged so that the polarization transmission axis of the polarizer 226 and the polarization transmission axis of the polarizer 227 are in the same direction. In the configuration example 228 in which the polarizer is arranged so that the polarization transmission axis of the polarizer 226 and the polarization transmission axis of the polarizer 227 are in the same direction, the light transmitted through the polarizer 227 when no voltage is applied between the electrodes 218A-218B. Is optically rotated by an optical rotation angle π / 2, so that it is shielded from light without passing through the polarizer 226. When a voltage is applied between the electrodes 218A-218B, the liquid crystal loses optical rotation and the light transmitted through the polarizer 227 does not rotate, so that it passes through the polarizer 226. Next, FIG. 21C schematically shows a configuration example 229 in which the polarizer is arranged so that the angle difference between the polarization transmission axis of the polarizer 226 and the polarization transmission axis of the polarizer 227 is π / 2. It is. In the configuration example 229 in which the polarizer is arranged so that the angle difference between the polarization transmission axis of the polarizer 226 and the polarization transmission axis of the polarizer 227 is π / 2, the voltage is not applied between the electrodes 218A-218B. Is transmitted through the polarizer 226 because the light transmitted through the optical axis is rotated by the optical rotation angle π / 2. When a voltage is applied between the electrodes 218A-218B, the liquid crystal loses optical rotation and the light transmitted through the polarizer 227 does not rotate, so that the light is blocked without passing through the polarizer 226. In this way, transmission and light shielding can be selected by turning the voltage on and off.

  The spatial filter 205 may be configured using an element capable of providing optical rotation, such as a liquid crystal or a magneto-optical modulator, between the polarizers. A plurality of spatial filters 205 may be arranged two-dimensionally to form one spatial distribution optical element 201. At that time, each of the plurality of spatial filters 205 constituting the spatial distribution optical element 201 can control the light shielding or transmission, and can control the light shielding region in a timely manner according to the scattered light distribution.

Next, the effect of using a neutral density filter having a different transmittance for each region on or near the pupil surface will be described.
FIG. 22 shows an example of transmittance distribution when the neutral density filter 230 having different transmittance is used for each region on the pupil plane or in the vicinity of the pupil plane. FIG. 22 shows an example in which a neutral density filter having a transmittance proportional to the inverse of the intensity distribution of scattered light from the sample is used. The region 344 in FIG. 22 is a region with a high transmittance, the region 345 is a region with a slightly high transmittance, the region 346 is a region with a slightly low transmittance, the backscattering has a large amount of scattered light from the sample, and the region 347 has a transmittance. It is a low area.

  Next, the case where the neutral density filter 230 having the transmittance of the reciprocal of the scattered light from the sample is used on the pupil plane will be described. Scattered light from a defect or foreign material in the forward scattering region on the pupil plane is Sf, scattered light from the sample in the forward scattering region is Nf, scattered light from the defect or foreign material in the backscattering region is Sb, and from the sample in the backscattering region Nb is the scattered light from the sample, Ns is the scattered light from the sample in the side scattering region, and Ss is the scattered light from the defect or foreign material in the side scattering region.

When a neutral density filter having a transmittance that is the reciprocal of the scattered light from the sample is used, the scattered light from the defect or foreign matter is normalized by the scattered light from the sample in each region. The ratio S / N of the scattered light from the sample to the scattered light from the defect or foreign matter normalized by the scattered light from the sample is (Equation 5).

  From (Equation 5), when a neutral density filter having a transmittance proportional to the inverse of the intensity distribution of scattered light from the sample is used, the ratio of the scattered light from the sample to the scattered light from the defect or foreign matter should be increased. You can see that

Further, the neutral density filter 230 having different transmittances for each region on the pupil plane or in the vicinity of the pupil plane is used to scatter from the sample with respect to scattered light from defects or foreign matters in each region on the divided pupil plane or in the vicinity of the pupil plane. A neutral density filter 230 having a transmittance proportional to the light ratio may be used.
Further, as the neutral density filter, a combination of a polarizer and a liquid crystal, or a combination of a polarizer and a magneto-optic modulator may be used. This uses the same configuration as the spatial filter described in FIG. 21, and the amount of light transmitted through the polarizer changes by controlling the optical rotation angle of the liquid crystal molecules with the magnitude of the voltage applied to the liquid crystal or magneto-optic modulator. . Using this, a neutral density filter capable of controlling the transmittance by applying a voltage can be realized.

Next, an example of a spatial filter or a neutral density filter corresponding to a plurality of wavelengths arranged on or near the pupil plane will be described.
FIGS. 23A, 23B, and 23C are diagrams illustrating an example of a spatial filter using a color filter in the case where light having a wavelength of 405 nm (blue) and light having a wavelength of 532 nm (green) is used as illumination light. It is.

  FIG. 23A shows scattered light from a particulate defect when light having a wavelength of 405 nm is used as illumination light, and FIG. 23B shows scattering from a crystal defect when light having a wavelength of 532 nm is used as illumination light. FIG. 23 (c) shows an example of a spatial filter when light having a wavelength of 405 nm is used as illumination light, and FIG. 23 (d) shows a spatial filter when light having a wavelength of 532 nm is used as illumination light. Show. Light having a wavelength of 405 nm is blue light, and the complementary color of blue is yellow. Light having a wavelength of 532 nm is green light, and the complementary color of green is magenta. 23A and 23B, a region 308 is a region where the scattered light intensity is strong on the pupil plane, a region 309 is a region where the scattered light intensity is slightly high on the pupil plane, and a region 310 is scattered on the pupil plane. The region 311 is the region where the light intensity is slightly weak, the region 311 is the region where the scattered light intensity is weak, the region 353 in FIGS. 23C and 23D is the light shielding portion of the spatial filter, and the region 354 is the spatial filter. Each of the transmission parts (openings) is shown.

  FIG. 23 (e) suppresses scattered light from the sample at wavelengths of 405 nm and 532 nm by using a particle defect using light with a wavelength of 405 nm as illumination light, crystal defects using light at a wavelength of 532 nm as illumination light, It is an example of the spatial filter 225 which can enlarge the ratio of the scattered light from a sample with respect to the scattered light from a defect or a foreign material. A region that blocks the scattered light from the illumination light having a wavelength of 532 nm and the scattered light from the illumination light having a wavelength of 405 nm transmits the region 353, and the scattered light from the illumination light having a wavelength of 532 nm and the scattered light from the illumination light having a wavelength of 405 nm are transmitted. The region transmits the scattered light from the illumination light having a wavelength of 532 nm through the region 354 and blocks the scattered light from the illumination light having a wavelength of 532 nm while blocking the scattered light from the illumination light having a wavelength of 532 nm. Regions 356 indicate the regions through which scattered light from the illumination light is transmitted. A region 353 is a light filter, a region 354 is transparent, a region 355 is yellow, and a region 356 is a spatial filter 225 configured by a magenta color filter.

Further, a color filter may be used as the neutral density filter 230 corresponding to a plurality of wavelengths that change the transmittance of each region.
The spatial distribution optical element 201 inserted into the pupil plane includes a phase shifter 202, a half-wave plate 203, a quarter-wave plate 204, a polarization direction control device 205 using a liquid crystal or a magneto-optic modulator, a polarizer 231, Any combination of the spatial filter 225, the neutral density filter 230, the neutral density filter using the color filter, and the spatial filter using the color filter may be used, or any of them may be formed on the same substrate. .

  FIG. 24A shows the ratio of the scattered light from the defect or foreign matter and the scattered light obtained from the scattered light simulation described in FIG. 7 to the radial polarization intensity distribution, azimuth polarization intensity distribution, X polarization intensity distribution, Y Derived by each of the polarization intensity distributions, light in a region larger than a certain threshold is transmitted, and the ratio of scattered light from a defect or foreign material to scattered light from a sample is derived to shield light in a region smaller than any certain threshold It is a figure which shows the example of the spatial distribution optical element by the combination of a polarizer and a spatial filter. In addition, when the ratio of the amount of scattered light from the foreign material and the amount of scattered light on the substrate surface is larger than an arbitrary threshold value for both X-polarized light and Y-polarized light, the polarizer and the light shielding plate are not used.

  The spatial distribution optical element 201i, which is a combination of a polarizer and a spatial filter, includes a polarizer and regions 232a and 232b having no light shielding plate, light shielding plates 233a and 233b, and a polarizer having a transmission polarization axis inclined by π / 2 with respect to X-polarized light. 234a, a combination of a polarizer 234b having a transmission polarization axis inclined by π / 2 with respect to X-polarized light and a polarizer 235a / 235b having a transmission polarizer transmitting X-polarized light.

In addition, the spatial distribution optical element 201j using a combination of another polarizer and a spatial filter derives a ratio between the amount of scattered light from the defect and the amount of scattered light from the sample in the radial polarization from the result of the scattered light simulation in FIG. We examined a combination of a polarizer and a light-shielding plate that transmits light in a region where the ratio of the light is larger than a certain threshold and shields the region where the ratio of scattered light from the defect and scattered light from the sample is smaller than an arbitrary threshold. It is an example of the spatial distribution optical element 201 obtained as a result, the light shielding plates 233c and 233d, a polarizer 234c having a transmission polarization axis inclined by π / 2 with respect to X-polarized light, and a transmission inclined by π / 2 with respect to X-polarized light. It depends on the combination of the polarizer 234d having the polarization axis.
Here, the polarizers 234a and 234c having a transmission polarization axis tilted by π / 2 with respect to the X polarization and the polarizers 234b and 234d having a transmission polarization axis tilted by π / 2 with respect to the X polarization are radial (radial direction). The ratio of the amount of scattered light from the defect to the amount of scattered light from the sample surface is better when a polarizer having a transmission polarization axis is used.

  The spatial distribution optical element 201k by the combination of the polarizer and the spatial filter derives the ratio of the amount of scattered light from the defect and the amount of scattered light from the sample in the X-polarized light based on the result of the scattered light simulation used to obtain FIG. Consider a combination of a polarizer and a light-shielding plate that transmits light in a region where an arbitrary ratio is larger than a certain threshold and shields light in a region where the ratio of scattered light from a defect and scattered light from a sample is smaller than an arbitrary threshold. This is an example of the spatial distribution optical element 201 obtained as a result, and is based on a combination of a light shielding plate 233e and a polarizer 235c having a transmission polarizer that transmits X-polarized light.

  The spatial distribution optical element 201l by the combination of the polarizer and the spatial filter derives the ratio of the scattered light amount from the defect and the scattered light amount from the sample in the Y-polarized light from the result of the scattered light simulation used to obtain FIG. Consider a combination of a polarizer and a light-shielding plate that transmits light in a region where an arbitrary ratio is larger than a certain threshold and shields light in a region where the ratio of scattered light from a defect and scattered light from a sample is smaller than an arbitrary threshold. This is an example of the spatial distribution optical element 201 obtained as a result, and is based on a combination of a light shielding plate 233f and polarizers 236a and 236b each having a transmission polarizer that transmits Y-polarized light. The combination of the polarizer and the light shielding plate described above is not limited to these, and can be appropriately selected according to the scattered light distribution.

  From the result of the scattered light simulation used to obtain FIG. 7, the ratio of the amount of scattered light from the defect to the amount of scattered light from the sample is derived in azimuth polarization, and light in an area where the arbitrary ratio is larger than a certain threshold is transmitted. However, as a result of examining a combination of a polarizer and a light shielding plate that shields a region where the ratio of the scattered light from the defect to the scattered light from the sample is smaller than an arbitrary threshold, there are no coordinates on the pupil plane that exceed the threshold. It was.

  Next, FIG. 24B shows an example of the spatial distribution optical element 201 by a combination of a polarizer and a light shielding plate obtained by dividing the pupil plane or the vicinity of the pupil plane into arbitrary regions. By dividing the pupil plane or the vicinity of the pupil plane into arbitrary regions, it can be realized more easily than the spatial distribution optical element of FIG. A region 237 in FIG. 24B is a region having a transmission polarization axis inclined by π / 4 from the axis 307 projected from the illumination incident surface on the pupil plane, and a region 238 is an axis projected from the illumination incident surface on the pupil plane. Each region having a transmission polarization axis tilted by −π / 4 from 307 is shown.

  Spatial distribution optical elements 201m and 201n are examples in which the pupil plane is divided into eight in the radial direction, and radial polarization, azimuth polarization, X-polarization, and Y-polarization polarizers and light-shielding plates are appropriately combined for each of the divided eight regions. is there. 201o and 201p divide the pupil plane into 8 parts in the radial direction and 2 parts in the circumferential direction. For each of the 16 divided areas, a polarizer and a light shielding plate for radial polarization, azimuth polarization, X polarization, and Y polarization are appropriately used. This is a combined example. In addition, although the example divided into 8 was shown here, the division | segmentation number and the division | segmentation method are not restricted to this, It can set suitably according to scattered light distribution.

  Here, the spatial distribution optical element 201 may control the polarization direction with a wave plate and select polarized light with a polarizer. By combining the wave plate and the polarizer, it is possible to simplify the transmission polarization axis direction of the polarizer used for the purpose of increasing the ratio of the scattered light amount from the foreign material to the scattered light amount from the substrate surface. For example, a distribution half-wave plate is placed in the regions 234e and 234f that transmit the radially polarized light of the spatial distribution optical element 201m so that the vibration direction of the electric field is converted from radial polarization to Y polarization. You may use the spatial distribution optical element by the combination of the wavelength plate, the polarizer of Y polarization, and the light-shielding plate.

  Further, instead of the half-wave plate of the spatial distribution optical element 201, a polarization direction control device using liquid crystal as described above or a polarization direction control device using a transparent magnetic material using a magneto-optical effect may be used. In this case, the polarization direction can be controlled more precisely than the half-wave plate. Moreover, ON or OFF of optical rotation can be easily selected. Further, a phase shifter may be used in combination with a spatial distribution optical element that is a combination of a wavelength plate, a polarizer, and a light shielding plate. By combining the phase shifter with the spatial distribution optical element, when the scattered light transmitted through the spatial distribution optical element is caused to interfere, a decrease in intensity caused by superposition can be suppressed, and the peak intensity can be increased by superposition. Alternatively, a photonic crystal having a function of a phase shifter, a polarizer, a light shielding plate, a wave plate, or a combination thereof may be used as the spatial distribution optical element. By using a photonic crystal, a spatial distribution optical element having a precise polarization selectivity and a polarization direction control function can be realized.

  Since the intensity distribution of scattered light differs depending on the optical properties such as the shape, size, refractive index, etc. of the minute foreign matter or minute defect to be detected, the characteristics of the spatial distribution optical element 201 inserted into the imaging optical system pupil plane 302 are shown in FIG. The shape is not limited to that shown in FIGS.

  The size of the foreign matter to be detected is reduced, or the scattered light intensity from the foreign matter is reduced by using the spatial distribution optical element 201 or both, so that the weak scattered light intensity from the fine foreign matter is doubled or imaging is performed. In order to suppress noise caused by the element 111, a high-sensitivity imaging element 111 may be used. By using the high-sensitivity image sensor 111, it is possible to increase the ratio of scattered light from the defect to noise caused by the image sensor. For example, the high-sensitivity imaging device 111 includes a cooled CCD camera (Cooled CCD Camera), an ICCD camera (Intensified CCD Camera), a SIT camera (Silicon Integrated CCD Camera), an EB-CCD camera (Electron Bombardment CCD Camera), and an EM CCD CCD. A camera (Electron Multiplier CCD Camera) or the like may be used as appropriate.

  The various spatial distribution optical elements described above can be used alone or in appropriate combination, and can be applied to the inspection apparatuses of the above-described embodiments as well as the inspection apparatuses of the embodiments described later.

  The operation in the configuration of the defect observation apparatus shown in FIG. 1 will be described. First, the sample 1 is transferred onto the sample holder 2 in the vacuum chamber 8 through a load lock chamber (not shown). Then, the sample 1 is moved into the field of view of the optical microscope 6 under the control of the stage 3. At this time, the sample 1 may be displaced from the focal position of the optical microscope 6. When the height of the sample 1 is deviated from the focal position, the objective lens 105 and the mirror 104 are moved to Z (height) using the height control mechanism 9 so that the sample 1 is set at the focal position of the optical microscope 6. Move in the direction. A method of determining the amount of movement in the Z direction will be described later.

  A defect on the sample 1 placed on the stage 3 of the defect observation apparatus shown in FIG. 1 is observed using position information of the defect on the sample 1 detected in advance by another defect inspection apparatus (not shown). In order to achieve this, wafer alignment that matches the reference position of the sample 1 with the reference of the stage 3 must be performed. This wafer alignment is performed using a bright field observation image. At the time of bright field detection, illumination light is emitted from the bright field illumination 109, reflected by the half mirror 108, and irradiated onto the sample 1 using the objective lens 105. The reflected light from the sample 1 passes through the imaging optical system 110 and forms an image on the image sensor 111. Here, the bright field light source 109 may be a lamp, for example. In the bright field observation of the present embodiment, the spatial distribution optical element 201 inserted into the imaging optical system 110 is switched to a parallel plate glass having the same thickness. When alignment is performed with the outer shape of the sample 1 (for example, orientation flat or notch if the sample 1 is a wafer), the positioning points of the sample 1 and several images of the outer shape may be acquired and processed.

  After wafer alignment, the position of the defect on the wafer is moved into the field of view of the optical microscope 6 in accordance with the position information of the defect detected in advance by the defect inspection apparatus, and the defect image is acquired by the dark field observation method of the optical microscope 6 To do. At this time, when the height of the sample 1 deviates from the focal position of the optical microscope 6 at each defect position, focusing is performed by a method described later.

Here, the dark field observation method will be described. In the dark field observation method, illumination light is emitted from the dark field illumination unit 101. Although the illumination light may be laser light or lamp light, it is desirable to use laser light because the laser light can increase the illuminance.
The light emitted from the illumination unit 101 is reflected by the epi-illumination mirror 102 and redirected in the Z direction, guided through the vacuum-sealed window 103 to the vacuum tube 8, and redirected by the mirror 104 to change the direction of the optical microscope 6. The surface of the sample 1 at the focal position is irradiated. The light scattered by the sample 1 is collected by the objective lens 105, guided to the imaging optical system 110, imaged at the imaging position of the image sensor 111, converted into an electrical signal by the image sensor 111, and controlled. Sent to system 10.

  The image acquired by the dark field detection method of the optical microscope 6 is accumulated in the control system 10 as a grayscale image or a color image. In the control system 10, as shown in FIG. 25, the center position of the visual field range 357 of the SEM 4 and the Y-direction position shift amount 359 and the X-direction position shift amount 360 of the defect 361 are calculated, and these shift amounts are registered as coordinate correction values. . Thereafter, using the coordinate correction value, the sample 1 is moved on the stage 3 so that the defect 361 enters the field of view 357 of the SEM 4 and observed with the SEM 4. The image of the observed defect is transmitted to the control system 10, and processing such as display on the user interface 11, registration in the database 12, and automatic defect classification is performed.

The flow of defect observation will be described with reference to FIG.
First, the sample 1 is aligned (6001). This is performed by the method described above by bright field observation with the optical microscope 6. Next, the stage 3 is moved so that the defect to be observed on the sample 1 enters the field of view of the optical microscope 6 using the position information of the defect detected in advance by another defect inspection apparatus (6002). Next, the objective lens 105 is moved by the height control mechanism 9 to perform focusing (6003).

  If a defect is searched from the image acquired by the optical microscope 6 and the image sensor 111 (6004) and a defect is detected (6005-YES), the defect detection position by the optical microscope 6 and other defect inspection devices are detected in advance. Based on the difference from the position information of the detected defect, using the position information of the defect detected in advance by another defect inspection apparatus, the deviation amount of the visual field position of the SEM 4 with respect to the defect when this defect is to be observed with the SEM 4 is calculated. (6006). Based on the calculated deviation amount, the position information of the defect detected in advance by the other defect inspection apparatus is corrected (6007), and the defect whose position information is corrected is moved to the field of view of the SEM 4 for observation. (6008). At this time, the observed information is sent to the control system 10 and registered in the database 11. When there are a large number of defects to be observed, a representative number of them are extracted and obtained by detecting the extracted defects with positional information detected in advance by another defect inspection apparatus and the optical microscope 6. The amount of deviation between the position of the defect detected in advance by another defect inspection apparatus and the visual field position of the SEM 4 is obtained from the position information of each defect. Using the information of the obtained deviation amount, the positional information obtained by detecting in advance by another defect inspection apparatus is also corrected for defects that have not been detected by the optical microscope 6 other than a few representative points.

Next, when other defect information is not necessary (6009-NO), the observation is ended (6010). When observation is necessary (6009-YES), other defect position information to be observed is acquired (6013). Returning to the procedure (6002) of moving another defect to be observed within the field of view of the optical microscope 6 described above, the process proceeds. In addition, when a defect cannot be detected by the above-described defect detection procedure (6005-NO), it is considered that the defect is outside the field of view of the optical microscope 6. Therefore, even if the periphery of the field of view of the optical microscope 6 is searched. Good. When searching for the peripheral portion (6012-YES), for example, the sample 1 is moved by an amount corresponding to the visual field (6011), and processing is performed from the above-described defect detection procedure. Further, when the vicinity search is not performed (6012-NO), the process proceeds according to the procedure.
There is also a method in which a defect position correction amount for each defect is calculated in advance and registered in a database, and after the calculation of position correction amounts for a plurality of defects or all defects is completed, inspection is performed by the SEM 4.

  A second configuration example and a third configuration example of the optical microscope 6 in the present embodiment will be described with reference to FIGS. Hereinafter, with respect to the modified examples, differences from the first embodiment will be mainly described, and description of similar parts will be omitted. The second and third configuration examples are different from the first embodiment in that the lens 131 is not used and the bright field illumination unit is not used. Therefore, there exists an advantage which becomes a simple structure shown to Fig.27 (a), (b). The spatial distribution optical element 201 is disposed on the pupil plane of the objective lens or in the vicinity of the pupil plane. 27A and 27B is the arrangement position of the dichroic mirror 123 for branching the optical path of the imaging optical system. In FIG. 27A, the optical path of the image forming optical system is divided by the dichroic mirror 123, and the image forming lenses 130A and 130B are respectively arranged, and each image is formed on the image sensor. In FIG. 27B, the optical path of the imaging optical system is branched by a dichroic mirror 123 after a single imaging lens 130 and imaged on each imaging device more simply. In the configuration shown in FIGS. 27A and 27B, the same reference numerals as those in FIG. 4 have the same functions as those described with reference to FIG. Although not shown here, the illumination optical system having the configuration shown in FIGS. 1 and 2 is used.

  A fourth configuration example of the optical microscope 6 in the present embodiment will be described with reference to FIG. This embodiment is different from the first embodiment in that the arrangement position of the spatial distribution optical element 201 and the bright field illumination unit are not used. For this reason, as shown in FIG. 27C, the same spatial distribution optical element 201 can be used for each branched light beam that forms an image on different imaging elements 111A and 111B. There is an advantage that the difference due to the influence of aberration or the like can be reduced. Note that in the configuration shown in FIG. 27C, components having the same numbers as those in FIG. 4 have the same functions as those described with reference to FIG. As in the above, although not shown here, the illumination optical system having the configuration shown in FIGS. 1 and 2 is used.

  A fifth configuration example of the optical microscope 6 in the present embodiment will be described with reference to FIG. This embodiment is different from the first embodiment in that the lens 131 of the microscope 6 is not used, a dichroic mirror is not used, and there is one image sensor. Therefore, the simple configuration shown in FIG. Detection under different detection conditions or illumination conditions is possible by switching the spatial distribution element, using a spatial distribution optical element corresponding to multiple wavelengths, or switching the illumination light. The dichroic mirror 132 may branch the light beam by taking it in and out of the optical axis of the imaging optical system, or may branch the light beam by using a mirror for the dichroic mirror. Note that in the configuration shown in FIG. 28A, the same reference numerals as those in FIG. 4 have the same functions as those described with reference to FIG. Although not shown here, the illumination optical system having the configuration shown in FIGS. 1 and 2 is used.

  A sixth configuration example of the optical microscope 6 in the present embodiment will be described with reference to FIG. This embodiment is different from the first embodiment in that the dichroic mirror of the microscope 6 is not used and that there is one image sensor. Therefore, the simple configuration shown in FIG. Detection under different detection conditions or illumination conditions is possible by switching the spatial distribution element, using a spatial distribution optical element corresponding to multiple wavelengths, or switching the illumination light. In the configuration shown in FIG. 28B, components having the same numbers as those in FIG. 4 have the same functions as those described with reference to FIG. Although not shown here, the illumination optical system having the configuration shown in FIGS. 1 and 2 is used.

  A seventh configuration example of the optical microscope 6 in the present embodiment will be described with reference to FIG. This embodiment is different from the first embodiment in that one illumination light source is used. Therefore, there is an advantage that the configuration can be simplified as shown in FIG. 29 and the apparatus can be made compact. It is possible to detect under different detection conditions or illumination conditions by switching the illumination wavelength using the wavelength modulation means for the illumination light, but it is possible to use illumination light of only a single illumination wavelength without wavelength conversion means. Good. The dichroic mirror 132 may branch the light beam by taking it in and out of the optical axis of the imaging optical system, or may branch the light beam using a mirror for the dichroic mirror.

Next, a method for controlling the optical characteristics of the spatial distribution optical element 201 according to the first embodiment of the present invention will be described with reference to FIG.
The light scattered on the surface of the sample 1 is collected by the objective lens, passes through the lenses 131A and 131B and the spatial distribution optical element 201, and is imaged on the imaging element 111A by the imaging lens 130A. The half mirror 120 is arranged after 131B, and the other light branched and branched is configured to form the pupil image of the objective lens 105 on the image sensor 111B using the imaging lens 130B. The pupil image signal obtained by the image sensor 111B is received and processed by the signal processing unit 139, the optical characteristics of the spatial distribution optical element 201 are determined, and the light shielding region or transmission of the spatial distribution optical element 201 is determined according to the scattered light distribution. By controlling the controllable optical characteristics such as the rate and the optical rotation angle using the control unit 138, the ratio of the scattered light from the sample to the scattered light from the defect or foreign matter can be detected largely.

  For example, a pupil image of scattered light from the sample is acquired by the image sensor 111B, and the transmittance is obtained depending on the magnitude of the applied voltage using a liquid crystal and a polarizer so as to obtain a transmittance proportional to the inverse of the scattered light from the sample. By controlling the transmittance of the spatial distribution optical element 201 that can control the light, and detecting the scattered light from the sample 1 using the spatial distribution optical element 201A whose transmittance is controlled, the scattered light from the defect or foreign matter The ratio of scattered light from the sample to can be increased. Note that the mirror 120 may branch the light beam in and out of the optical axis of the imaging optical system, or the light beam may be branched using a dichroic mirror as the mirror.

Next, an example of image processing using the acquired images of a plurality of optical conditions using the first embodiment of the present invention will be described with reference to FIG. FIG. 31A is a diagram for explaining an example of signal processing that enables highly sensitive detection of particle defects and crystal defects using two-wavelength illumination lights having different wavelengths. A defect map 1A, a defect map 1B mainly including crystal defects, and a defect map 1C obtained by combining them are shown. A particle defect whose scattering efficiency is improved as the illumination becomes shorter in wavelength is detected by the imaging element 111A using scattered light of illumination light having a shorter wavelength. At that time, by using the spatial distribution optical element 201A in accordance with the scattered light distribution of the particle defect, the particle defect can be detected with high sensitivity (1A is a defect map mainly including the detected particle defect existing on the sample). . The crystal defect can be detected with high sensitivity by detecting the image sensor 111B using the spatial distribution optical element 201 matched with the crystal defect detection with the other light having a longer wavelength than the short wavelength illumination (1B). Is a defect map mainly composed of detected crystal defects on the sample). By obtaining the defect map 1C of the particle defect and the crystal defect by the process of combining the defect maps 1A and 1B acquired by the image sensors 111A and 111B, it is possible to detect the particle defect and the crystal defect with high sensitivity. FIG. 31B shows an example in which illumination is performed with a single wavelength, one imaging element 111A uses a spatial distribution optical element 201 matched to a defect, and the other does not use a spatial distribution optical element. The defect map 1D is a defect map acquired using the spatial distribution optical element 201 matched to the defect to be detected. The defect map 1E is a defect map acquired without using the spatial distribution optical element 201. 1D ′ is obtained by multiplying the signal acquired by the defect map 1D by gain so that the amount of light scattered from the sample of the signal acquired by the defect map 1E is approximately the same. By subtracting the signal from 1D ′ to 1E, scattered light from the sample can be removed, and the defect to be obtained can be emphasized and detected (acquired defect map 1F).
The above is merely an example, and the present invention is not limited to this. Similarly, by using the acquired scattered light signals of a plurality of optical conditions, and appropriately performing addition / subtraction, etc., highly sensitive detection of a plurality of defect types and , Defect enhancement, defect classification, etc. are possible.

  Next, FIG. 32A shows an example of a GUI that can input the wavelength, polarization direction, elevation angle, azimuth angle, detection conditions of the optical defect detection apparatus, defects to be detected, and the like according to the first embodiment of the present invention. Shown in The user can also create a recipe. Needless to say, the GUI shown here is merely an example and can be changed as appropriate. For example, there may be a plurality of items of illumination conditions and detection conditions in one screen. Even if the user sets a numerical value each time, the illumination conditions performed in the past are stored as data, and this data is stored. You may make it read and utilize.

  Next, FIG. 32B shows an example of an input GUI for performing defect enhancement in the first embodiment of the present invention. For example, the same region on the sample 1 is illuminated and detected with two wavelengths, a short wavelength and a long wavelength, and an image 1G obtained by the short wavelength detection and an image 1H obtained by the long wavelength are used. A graph 140 is obtained in which the signal value at the detection and the signal value at the short wavelength detection are plotted on the vertical axis, and each bright spot is mapped. By selecting an area to be displayed on the graph 140 and displaying only the selected defect on the map 1I, it is possible to display only the defect to be viewed. Here, a straight line is selected as the region setting, and the non-display / non-alignment target and the display / alignment target are separated. The GUI shown here is only an example, and the defect map 1G, 1H, and 1I may be displayed on the same screen as the graph 140, and the region setting selection can be changed as appropriate while viewing the defect map 1I. It doesn't matter. In addition, the area can be arbitrarily set on the graph 140 by the user pressing a curve button.

(Second Embodiment)
A defect observation apparatus according to the second embodiment of the present invention will be described with reference to FIG. An inspection apparatus for inspecting a surface or a defect of the sample 601 includes a dark field illumination optical system configured by appropriately using a laser 607, an expander 609, an attenuator 610, a polarization control element 611, mirrors 612A and 612B, and a lens 613. A sample holder 602 for mounting a sample, a stage 603 having a Z stage and an XY stage for moving the sample holder, a sample height measuring device 605, an objective lens 614, a spatial distribution optical element 201, an imaging lens 615, and a dichroic A detection optical system configured by appropriately using imaging elements 606 and 619 in each of two optical paths branched by the mirror 621 and the dichroic mirror 621, a signal processing unit 616, and a monitor 617 are configured as appropriate. . Further, a detection system monitoring unit 620 that measures the state of a detection optical system configured using the dichroic mirror 621 and the image sensor 619, an illumination system monitoring unit that measures the state of the illumination optical system (not shown), and And a control unit that controls each operating unit.

  First, the configuration of the dark field illumination system will be described. The laser 607 emits illumination light 701 from a direction having an angle with respect to the normal direction of the sample, and forms a desired beam such as a spot or a line on the surface of the sample 601. Here, although not shown, the expander 609 is used to spread the illumination light 701 into a parallel light flux having a constant magnification. The attenuator 610 is an attenuator for controlling the light quantity / intensity of the illumination light 701 after passing through the expander 609. The polarization control element 611 is an element that controls the polarization state by changing the direction of liquid crystal molecules by rotating a polarizing plate or a wave plate, or by controlling the voltage ON / OFF to switch the polarization direction of light incident on the element. The mirrors 612A and 612B are a reflecting mirror group for adjusting the irradiation angle when the sample 601 is irradiated with the illumination light 701 after polarization control (electric field phase and amplitude control). Here, an example in which two mirrors are used is shown, but a configuration without using a mirror may be used, or one or three or more may be used. The lens 613 is a lens for converging the illumination light 701 at the irradiation position immediately before irradiating the sample 601. Further, as the dark field illumination system, a system capable of oscillating a plurality of wavelengths as shown in FIG. 2 may be appropriately selected and used.

  Next, the configuration of the detection optical system will be described. The objective lens 614 is an objective lens that collects light scattered and diffracted from foreign matters, defects, and patterns on the sample 601 by irradiation of the illumination light 701 by the laser 607 from the normal direction (upward) of the sample 601. Here, when the sample 601 which is a semiconductor device or the like to be inspected by the dark field defect inspection apparatus has a repetitive pattern, the diffracted light generated from the repetitive pattern is collected at regular intervals on the exit pupil of the objective lens 614. Shine. The spatial distribution optical element 201 is a filter that shields this repetitive pattern near the pupil plane, or a filter that controls and selects the polarization direction of all or a part of the light reflected from the object to be inspected or a specific polarization direction. Here, the various spatial distribution optical elements 201 described in the first embodiment may be appropriately used. The image forming lens 615 is a lens for forming an image on the image sensor 606 of scattered light and diffracted light from other than a repetitive pattern (for example, a place where a failure has occurred) that has passed through the spatial distribution optical element 201. The image sensor 606 is an optical sensor for sending an image focused and imaged by the imaging lens 615 to the signal processing unit 616 as electronic information. As a type of the optical sensor, a CCD, a CMOS, or the like is generally used, but the type is not limited here.

The signal processing unit 616 includes a circuit for converting the image data received from the image sensor 606 into a state that can be displayed on the monitor 617.
The XY stage of stage 3 is a stage for moving the sample holder 602 on which the sample 601 is placed. The sample 601 is scanned by moving the XY stage in the plane direction, and the Z stage is an inspection standard for the XY stage. This is a stage for moving the surface (the surface on which the sample 601 is placed) in the vertical direction (Z direction).
The sample height measuring unit 605 is a measuring instrument for measuring the inspection reference plane of the XY stage of the stage 3 and the height of the sample 601. An autofocus function can be provided in which the focus position is automatically adjusted by the Z stage of the stage 3 and the sample height measuring unit 605.

Next, the overall operation of the inspection apparatus according to the second embodiment will be described.
First, the illumination light 701 from the laser 607 illuminates the surface of the sample 601 from a direction having an angle with respect to the normal direction of the sample, thereby forming a desired beam on the sample 601. Light scattered and diffracted from foreign matters, defects, and patterns on the sample 601 by this beam is collected by the objective lens 614 above the sample. When the sample 601 has a repetitive pattern, the diffracted light generated from this repetitive pattern is collected at regular intervals on the exit pupil of the objective lens, so that the spatial distribution optics placed on the pupil plane or in the vicinity of the pupil plane. Light is shielded by the element 201. The spatial distribution optical element 201 may use the spatial distribution optical element 201 for the purpose of enhancing scattered light from a defect or suppressing scattered light from a sample.

  The sample 601 is placed on the XY stage of the stage 3, and a two-dimensional image of scattered light from the sample 601 is obtained by scanning with the XY stage of the stage 3. At this time, the distance between the sample 601 and the objective lens 614 is measured by the sample height measuring unit 605 and adjusted by the Z stage of the stage 3.

  The two-dimensional image acquired by the image sensor 606 is classified for each foreign substance type and defect type by the signal processing unit 616, the size of the foreign substance or defect is obtained, and the result is displayed on the monitor 617.

  Next, a control unit 800 according to the second embodiment of the present invention will be described with reference to FIG. FIG. 34 is a block diagram showing an internal configuration of the control unit 800. The control unit 800 includes a recording unit 801, a comparison unit 802, a sensitivity prediction unit 803, and a feedback control unit 804 as appropriate.

  The recording unit 801 receives input of data from the monitored illumination system monitoring unit and detection system monitoring unit 620 and records these data. The comparison unit 802 receives the data recorded by the recording unit 801 and compares it with the ideal value in the database 805. Note that the characteristics of the light source and the element at the time of monitoring are calculated in advance before processing in the comparison unit 802. The sensitivity prediction unit 803 estimates / predicts the current device sensitivity from the difference between the recorded data and the ideal value. If the deviation between the recorded data and the ideal value is within an allowable range, each operation unit such as the illumination optical system and the detection optical system is controlled to start inspection. If it is outside the allowable range, the feedback control unit 804 applies feedback to each operation unit of the apparatus according to the predicted sensitivity predicted by the sensitivity prediction unit 803.

  Here, the database 805 is a database of ideal values used by the comparison unit 802, and ideal values are input to the database 805 in advance by theoretical calculation, optical simulation, or the like. At this time, the optical simulator models the inspection object, derives the scattered light intensity from the inspection object generated depending on the conditions of the illumination optical system, and calculates the light intensity detected by the inspection device. . The ideal value parameters in the database 805 include information such as the intensity distribution of the illumination optical system, the polarization state distribution, the focal length of the imaging lens 615, and the sensitivity of the image sensor 606. It is necessary to grasp the characteristics of these parameters in advance.

Next, a monitoring process procedure in the dark field defect detection apparatus according to the second embodiment of the present invention will be specifically described with reference to the flowchart of FIG.
First, the illumination system monitoring unit monitors the state of the illumination system (step S10). Further, the detection system monitoring unit 620 measures the state of the detection system (step S11). The measurement results obtained in step S10 and step S11 are sent to the comparison unit 802. The comparison unit 802 compares these measurement results with ideal values in the database 805, and further predicts detection sensitivity from “deviation” between the ideal values and the measurement results (step S12). Then, it is determined whether the predicted detection sensitivity is larger or smaller than the arbitrarily set threshold (step S13).

  If the predicted sensitivity is less than or equal to the threshold value, the optical system is calibrated (step S14), and the process returns to step S10 again. Here, if all the calibration required points can be automatically controlled, all calibration operations may be automatically performed. At this time, the calibration location may be determined in advance by theoretical calculation or optical system simulation. On the other hand, when the predicted sensitivity is equal to or higher than the threshold value, inspection of the illumination system and the detection system is started (step S15).

  As mentioned above, the invention made by the present inventor has been specifically described based on the embodiment. However, the present invention is not limited to the above embodiment, and various modifications can be made without departing from the scope of the invention. Not too long.

DESCRIPTION OF SYMBOLS 1 ... Sample 2 ... Sample holder 3 ... Stage 4 ... Electron microscope 5, 7 ... Optical height detection system 6 ... Optical microscope 8 ... Vacuum tank 9 ... Height control mechanism 10 ... Control system 11 ... User interface 12 ... Database 13 ... Network 101 ... Dark field illumination unit 102 ... Epi-illumination mirror 103 ... Vacuum sealed window 104 ... Mirror 105 ... Objective lens 110 ... Imaging optical system 111 ... Imaging element 112 ... Illumination light source 113 ... Optical filter 114 ... Wave plate 115 ... Flat Concave lens 116 ... Achromatic lens 117 ... Cylindrical lens 118 ... Lens group 119 ... Non-linear optical crystal 120 ... Half mirror 121 ... Mirror 122 ... Fiber 123, 132 ... Dichroic mirror 124 ... Non-linear optical crystal switching mechanism 125 ... Optical path cutting mechanism Replacement mechanism 126 ... Bright field illumination unit 127 ... Condensing lens 128 ... Bright field light source 129 ... Half mirror 130 ... Imaging lens 131 ... Lens 133 ... Image processing unit 134 ... Image display unit 135 ... Image storage unit 136 ... Spatial distribution optics Element holder 137 Rotation drive unit 138 Control unit 139 Signal processing unit 140 Graph 150 Light source 151 154 Condensing lens 152 Slit 153 Imaging lens 155 Detector 201 Spatial distribution optical element 202 Phase Shifter 203 ... 1/2 wavelength plate 204 ... 1/4 wavelength plate 205 ... Polarization direction control device 214, 216, 217 ... Alignment film 215 ... Liquid crystal 218 ... Electrode 219 ... Liquid crystal molecule 221 ... Power source 222 ... Transparent magnetic body 223, 224 ... Transparent substrate 225 ... Spatial filters 226, 227, 231, 234, 235, 236 ... polarizer 230 ... neutral density filter 233 ... light shielding plate 301 ... optical axis 302 ... pupil plane 303 ... slow axis 304 ... fast axis 318 ... X-polarized light 319 ... Y-polarized light 320 ... radial polarized light 327 ... Polarized transmission axis 601 ... Sample 602 ... Sample holder 603 ... Stage 605 ... Sample height measuring device 606, 619 ... Image sensor 607 ... Laser 609 ... Expander 610 ... Attenuator 611 ... Polarization control element 612 ... Mirror 613 ... Lens 614 ... Objective lens 615 ... Imaging lens 616 ... Signal processing unit 617 ... Monitor 620 ... Detection system monitoring unit 621 ... Dichroic mirror 800 ... Control unit 801 ... Recording unit 802 ... Comparison unit 803 ... Sensitivity prediction unit 804 ... Feedback control unit 805 ... Database

Claims (22)

An optical defect detection apparatus for inspecting a sample surface,
A stage for holding the sample;
An illumination optical system for irradiating the sample surface with light obliquely;
An objective lens that captures scattered light from the irradiated region on the sample surface, and scattered light from defects on the sample and scattered light from irregularities on the sample among the scattered light captured by the objective lens A detection optical system comprising: a spatial distribution optical element that controls the captured scattered light so as to increase the ratio; and an imaging element that detects the scattered light controlled by the spatial distribution optical element;
Have
The spatial distribution optical element is an element that controls a polarization direction of scattered light captured by the objective lens, an element that selects a polarization direction through which the scattered light captured by the objective lens transmits, or is captured by the objective lens. An optical defect detection apparatus characterized in that any one of elements having different transmittances of scattered light for each region or a combination of at least two of these elements is used. The optical defect detection device according to claim 1,
The detection optical system further includes branching means for branching scattered light captured by the objective lens into two or more optical paths,
The spatial distribution optical element is disposed in each of two or more optical paths branched by the branching unit,
The imaging device is arranged to detect each scattered light that is branched into each of the two or more optical paths and controlled by the spatial distribution optical element, respectively.
The optical defect detection apparatus further comprises a signal processing system for comparing or processing images detected by the arranged image pickup devices. The optical defect detection device according to claim 2,
The illumination optical system is an illumination optical system that irradiates a plurality of lights having different wavelengths,
The optical defect detection apparatus, wherein the branching means branches scattered light captured by the objective lens according to wavelength. The optical defect detection device according to any one of claims 1 to 3,
The spatial distribution optical element is a phase shifter, an optical rotator, a polarizer, a spatial filter, a neutral density filter, a phase shifter, or a defect on the sample surface with respect to light scattered by irregularities on the sample surface. Phase shifter that has the function of changing the phase of the scattered light so that the ratio of the scattered light becomes high and gives a spatially different phase difference, or the light scattered by the irregularities on the sample surface or the defects on the sample surface A distributed phase shifter having a function of converting the polarization direction of scattered light into linearly polarized light and having a spatially different fast axis and slow axis, or the sample surface with respect to light scattered by unevenness of the sample surface The function of aligning the polarization direction so that the ratio of the light scattered by the defects in the surface increases, and the rotation has a spatially different fast axis and slow axis. The ratio of the light scattered by the defects on the sample surface to the light scattered by the projections or irregularities on the sample surface increases the ratio of the light in the polarization direction, and the spatially different transmission axes Or a spatial filter having a function of selectively transmitting a region in which the ratio of light scattered by defects on the sample surface to light scattered by unevenness on the sample surface is high, or unevenness on the sample surface A neutral density filter having a function of selecting the transmittance so that the ratio of the light scattered by the defect on the sample surface to the light scattered by the sample has a spatially different transmittance, or irregularities on the sample surface The polarization direction of the scattered light so that the ratio of the light scattered by the defects on the sample surface to the light scattered by A photonic crystal having a function of controlling, a function of selecting a polarization direction through which the scattered light is transmitted, a function of selecting a transmittance of the scattered light, or a function of a combination of at least two of these, An optical defect detection apparatus comprising any one of or a combination thereof. The optical defect detection device according to any one of claims 1 to 3,
The spatial distribution optical element includes a phase shifter, an optical rotator, a polarizer, a spatial filter, an ND filter, or light scattered by a defect on the sample surface with respect to light scattered by irregularities on the sample surface. A phase shifter that has a function of changing the phase of scattered light so that the ratio becomes high and gives spatially different phase differences, or light scattered by irregularities on the sample surface or light scattered by defects on the sample surface 1/4 wavelength plate with the function of converting the polarization direction into linearly polarized light and having spatially different fast axis and slow axis, or scattered by defects on the sample surface against light scattered by the irregularities on the sample surface Half-wave plate or liquid crystal with the function of aligning the polarization direction so that the ratio of the emitted light is high and having a spatially different fast axis and slow axis Is an optical rotator composed of either a magneto-optic modulator or a light with a polarization direction that increases the ratio of light scattered by defects on the sample surface to light scattered by irregularities on the sample surface A polarizer having a function of transmitting light and having spatially different transmission axes, or a region where the ratio of the light scattered by the defects on the sample surface to the light scattered by the unevenness on the sample surface is selectively increased Either a mask having a transmitting function, a combination of a polarizer and a magneto-optical modulator, a combination of a polarizer and a liquid crystal, a combination of a polarizer and a half-wave plate, or a two-dimensional array shutter using MEMS A spatial filter having a light-shielding portion constituted by the surface of the sample surface with respect to light scattered by unevenness of the sample surface ND filter with a function to select the transmittance so that the ratio of the light scattered by the depression becomes high and spatially different transmittance, or a combination of a polarizer and a liquid crystal, or a polarizer and a magneto-optic modulator A polarization filter of the scattered light such that the ratio of the light scattered by the defects on the sample surface to the light scattered by the unevenness of the sample surface, or a light filter formed by any of the combinations A photonic having a function of controlling a direction, a function of selecting a polarization direction through which the scattered light is transmitted, a function of selecting a transmittance of the scattered light, or a function of a combination of at least two of these An optical defect detection device characterized by comprising any one of crystals or a combination thereof . The optical defect detection device according to any one of claims 1 to 3,
The spatial distribution optical element includes a phase shifter, an optical rotator, a polarizer, a spatial filter, an ND filter, or light scattered by a defect on the sample surface with respect to light scattered by irregularities on the sample surface. A phase shifter that has a function of changing the phase of scattered light so that the ratio becomes high and gives spatially different phase differences, or light scattered by irregularities on the sample surface or light scattered by defects on the sample surface 1/4 wavelength plate with the function of converting the polarization direction into linearly polarized light and having spatially different fast axis and slow axis, or scattered by defects on the sample surface against light scattered by the irregularities on the sample surface Half-wave plate or liquid crystal with the function of aligning the polarization direction so that the ratio of the emitted light is high and having a spatially different fast axis and slow axis Has a function of selectively transmitting light in a polarization direction in which the ratio of the light scattered by the defect on the sample surface to the light scattered by the irregularities on the sample surface is increased. Polarizers having different transmission axes, or regions having a high ratio of light of multiple wavelengths scattered by defects on the sample surface to light of multiple wavelengths scattered by unevenness on the sample surface by the multiple illumination, Either a color filter having a function of selectively transmitting for each wavelength, or a mask, a combination of a polarizer and a magneto-optical modulator, a combination of a polarizer and a liquid crystal, or a combination of a polarizer and a half-wave plate A spatial filter having a light-shielding portion constituted by the surface of the sample surface with respect to light scattered by unevenness of the sample surface An ND filter, a color filter, a combination of a polarizer and a liquid crystal, or a polarizer and a magnetic device having a function of selecting a transmittance so that the ratio of light scattered by the depression is increased and having a spatially different transmittance A neutral density filter comprising any combination of optical modulators, or the scattered light so that the ratio of the light scattered by the sample surface defects to the light scattered by the sample surface irregularities is increased A function for controlling the polarization direction of the scattered light, a function for selecting a polarization direction for transmitting the scattered light, a function for selecting the transmittance of the scattered light, or a function combining at least two of them A photonic crystal having any one of these, or a combination thereof. Optical defect detection device. The optical defect detection device according to claim 1,
The optical defect detection apparatus, wherein the imaging device acquires a Fourier transform image of light scattered on the sample surface imaged by the detection optical system. The optical defect detection device according to any one of claims 1 to 6,
Based on the signal obtained from the image sensor so that the ratio of the light scattered by the defects on the sample surface to the light scattered by the irregularities on the sample surface is higher, the polarization direction transmitted by the spatial distribution optical element Or an optical defect detection apparatus that selects and changes any one or a combination of a direction in which a polarization direction is converted or a transmittance. The optical defect detection device according to claim 8,
The signal obtained from the imaging element is the same as the imaging element that forms an image of the sample surface using the spatial distribution optical element, and the sample surface to be observed using the spatial distribution optical element A signal obtained by irradiating the same region, a signal obtained by irradiating a region different from the sample surface to be observed using the spatial distribution optical element, or obtained from the image sensor during scanning on the stage An optical defect detection apparatus, wherein the signal is a signal obtained by statistically processing a signal obtained from the image sensor during scanning on the stage. The optical defect detection device according to claim 8,
An optical defect characterized in that the signal obtained from the image sensor is a signal obtained from the individual image sensor different from the image sensor that forms an image of the sample surface using the spatial distribution optical element. Detection device. The optical defect detection device according to any one of claims 1 to 10,
further,
The light scattered by the defects on the sample surface relative to the light scattered by the irregularities on the sample surface for each optical condition including at least one of defect shape, defect type, illumination wavelength, illumination polarization, detection polarization, and detection conditions A library storing either or a combination of a polarization direction transmitted through the spatial distribution optical element having a high ratio, a direction for converting the polarization direction, or a transmittance;
A GUI for inputting at least one of illumination condition or detection condition or defect shape information to be detected;
An optical defect detection apparatus comprising: The optical defect detection device according to claim 11,
Control means for selecting and changing either or a combination of a polarization direction transmitted through the spatial distribution optical element, a direction for converting the polarization direction, or a transmittance based on information input to the library and the GUI A defect inspection apparatus comprising: The optical defect detection device according to any one of claims 1 to 12,
The optical defect detection device, wherein the spatial distribution optical element is arranged in the pupil plane of the detection optical system, in the vicinity of the pupil plane, or between the objective lens and the sample. The optical defect inspection apparatus according to any one of claims 1 to 12,
The optical defect inspection apparatus, wherein the spatial distribution optical element is incorporated in the objective lens. The optical defect detection device according to any one of claims 1 to 14,
2. The optical defect detection apparatus according to claim 1, wherein the spatial distribution optical element has a mechanism exchangeable with another spatial distribution optical element. The optical defect detection device according to any one of claims 1 to 15,
The optical defect detection apparatus characterized in that the light used in the illumination optical system uses any one of a visible light laser, an ultraviolet light laser, a vacuum ultraviolet light laser, a lamp, and a light emitting diode. The optical defect detection device according to claim 3,
The light of a plurality of wavelengths used in the illumination optical system is emitted from a plurality of lasers, or a laser capable of emitting a plurality of wavelengths oscillates light of different wavelengths depending on time, or the light emitted from the lamp is separated by a wavelength band, Or, light oscillated from a single laser oscillates light of different wavelengths depending on the time by inserting / removing a nonlinear optical crystal or electro-optical element, or the optical path is switched by inserting / removing a mirror for light oscillated from a single laser Each optical path that diverges by splitting light oscillated from a single laser by optical path splitting means such as fiber, half mirror or dichroic mirror by changing the presence or absence of crystal or electro-optical element passing Or, it can be modulated with a nonlinear optical crystal or an electrical engineering element in one optical path. Oscillates light of wavelengths, optical defect detection apparatus, characterized in that is obtained by either. The optical defect detection device according to any one of claims 1 to 17,
The optical defect detection apparatus further comprises an optical height detection means for obtaining height information necessary for focusing of the objective lens. A defect observation apparatus comprising the optical defect detection apparatus according to any one of claims 1 to 18,
A defect observation apparatus comprising an electron microscope for observing defects of the sample aligned based on positional information of defects on the sample surface obtained by the optical defect detection apparatus. A defect observation apparatus for inspecting a sample surface and observing defects,
A stage for holding the sample;
An illumination optical system for irradiating the sample surface with light obliquely;
An objective lens that captures scattered light from the irradiated region on the sample surface, and scattered light from defects on the sample and scattered light from irregularities on the sample among the scattered light captured by the objective lens A detection optical system comprising: a spatial distribution optical element that controls the captured scattered light so as to increase the ratio; and an imaging element that detects the scattered light controlled by the spatial distribution optical element;
An optical defect detection device comprising:
An electron microscope that performs alignment based on positional information of defects or foreign matter on the sample surface obtained by the optical defect detection device, and observes the defects;
Have
The spatial distribution optical element is an element that controls a polarization direction of scattered light captured by the objective lens, an element that selects a polarization direction through which the scattered light captured by the objective lens transmits, or is captured by the objective lens. A defect observing apparatus characterized in that any one of the elements having different transmittances of scattered light for each region or a combination of at least two of them is used. An optical defect detection method for detecting defects on a sample surface,
The sample surface is illuminated with illumination light from an oblique direction, the scattered light from the illuminated region on the sample surface is captured by an objective lens, and the scattered light captured by the objective lens is a half mirror or dichroic The light is branched into two optical paths by one of the mirrors, and a Fourier-transformed image of the light scattered on the sample surface using the branched first light is detected by the first imaging element, and the first imaging element is detected. A spatial distribution optical element having a function of controlling a polarization direction arranged on the branched second optical path based on a signal obtained from the image sensor, selecting a transmission polarization direction, and selecting a transmittance. The second branched light is imaged on the second image sensor by the imaging optical system, and a defect on the sample surface is detected from the image obtained by the second image sensor. Optical defect detection method characterized by . An optical defect detection method for inspecting defects on a sample surface,
Illuminate the sample surface obliquely with illumination light, capture scattered light from the illuminated area on the sample surface with an objective lens, and control the polarization direction of the captured light with a spatial distribution optical element Then, the direction of polarized light to be transmitted is selected, the transmittance is selected, the light transmitted through the spatial distribution optical element is imaged on the image sensor by the imaging optical system, and the surface of the sample is obtained from the image obtained by the image sensor. To detect defects on the top,
Input optical conditions such as defect shape, defect type, illumination wavelength, illumination polarization, detection polarization, and detection conditions on the GUI, for each optical condition such as defect shape, defect type, illumination wavelength, illumination polarization, detection polarization, and detection conditions. The ratio of the light scattered by the defects on the sample surface to the light scattered by the irregularities on the sample surface is increased, the polarization direction transmitted by the spatial distribution optical element, the direction for changing the polarization direction, or the transmittance The ratio of the light scattered by the defect on the sample surface to the light scattered by the unevenness on the sample surface under the conditions input by the GUI from a library storing any one or a combination thereof is increased. Select the polarization direction to transmit, the direction to change the polarization direction, or the transmittance, or a combination of these , The polarization direction or a direction which converts the polarization direction, or transmittance, either or optical defect detecting method characterized by having a function of controlling the combination of passes of the spatial distribution optical element.

Images