Defect inspection device and defect inspection method

Abstract

Provided is a defect inspection device comprising: a stage for placing and moving a sample; an illumination optical system that irradiates the sample surface with linear illumination light; a line sensor that detects scattered light generated from a linear irradiation region on the sample surface irradiated with the illumination light; a detection optical system that forms a linear illumination image, which is an image of the irradiation region, on the line sensor, the optical axis of the detection optical system being inclined with respect to the normal line of the sample surface; and an area sensor disposed at a position the optical distance if which from the sample surface is equal to the optical distance between the line sensor and the sample surface, the irradiation region being linear with the longitudinal direction thereof being the direction orthogonal to the main scanning direction of the sample, an image surface formed by the detection optical system and conjugate to the sample surface being inclined with respect to the optical axis of the detection optical system, a first light receiving surface of the area sensor being optically two-dimensionally parallel to the image surface, and a second light receiving surface of the line sensor being inclined with respect to the image surface and intersecting the image surface in a linear illumination image.

Description

Defect inspection device and defect inspection method

[0001] The present invention relates to a defect inspection apparatus and a defect inspection method.

[0002] For example, in a manufacturing process of a substrate for a semiconductor, a flat panel display, or the like, defects on the surface of the substrate are inspected. An example of an apparatus used for such inspection is a defect inspection apparatus (Patent Document 1) that includes "an illumination optical system that irradiates a linear illumination spot on a sample surface, a light-focusing detection unit that condenses light reflected from the illumination spot on the sample surface, and a sensor unit that forms an optical image of the illumination spot by the light-focusing detection unit on a light-receiving surface and outputs the optical image as an electrical signal, wherein a first angle formed between an optical axis of the light-focusing detection unit and a longitudinal direction of the linear illumination spot irradiated on the sample is 10 degrees or more and less than 80 degrees, and the sensor unit is a line sensor having an array of light-receiving units at a position conjugate with the linear illumination spot formed on the sample surface, and a second angle formed between an arrangement direction of the array of light-receiving units of the sensor unit and the optical axis of the light-focusing detection unit is 10 degrees or more and less than 80 degrees, and the second angle differs from the first angle by 5 degrees or more."

[0003] International Publication No. 2021 / 024319

[0004] In order to obtain stable, high inspection sensitivity in a dark-field optical defect inspection device employing an imaging-type detection optical system as described in Patent Document 1, fine adjustment of the focus of the optical system is required. In a detection optical system in which the optical axis is tilted with respect to the normal to the sample surface, when an image of a linear area on the sample surface is formed on a line sensor, the position of the linear illumination spot must be aligned with the linear area.

[0005] However, the linear area on the sample surface that you want to image on the line sensor will shift in the height direction from the focal plane where the line sensor is focused due to fluctuations in the height of the sample surface. Furthermore, the position of the object point that is focused on the line sensor may change on the focal plane and shift from the illumination spot due to, for example, a focus position shift or optical axis shift in the detection optical system, which also causes focus shift. However, when focus shift occurs, it is difficult to determine from the output signal of the line sensor whether the cause is a fluctuation in the height of the sample surface or a shift in the illumination spot.

[0006] An object of the present invention is to provide a defect inspection apparatus and a defect inspection method that can accurately adjust the focus of an optical system and maintain a stable, high inspection sensitivity.

[0007] In order to achieve the above object, the present invention provides a defect inspection device comprising: a stage for mounting a sample and moving it; an illumination optical system for irradiating a linear illumination light onto the sample surface; a line sensor for detecting scattered light generated from a linear illumination area on the sample surface irradiated with the illumination light; a detection optical system whose optical axis is inclined with respect to a normal to the sample surface and which guides the scattered light to the line sensor and forms a linear illumination image, which is an image of the illumination area, on the line sensor; and a signal processing device for processing signals output by the line sensor to identify defects, the device comprising: an area sensor disposed at an optical distance from the sample surface equal to the optical distance between the line sensor and the sample surface, the illumination area being a line whose major axis direction is a direction perpendicular to a main scanning direction of the sample, an image plane formed by the detection optical system and conjugate to the sample surface is inclined with respect to the optical axis of the detection optical system, a first light-receiving surface of the area sensor is optically two-dimensionally parallel to the image plane, and a second light-receiving surface of the line sensor is inclined with respect to the image plane and intersects with the image plane with the linear illumination image.

[0008] According to the present invention, the focus of the optical system can be adjusted with high precision, and high inspection sensitivity can be stably maintained.

[0009] FIG. 1 is a schematic diagram of a defect inspection apparatus according to an embodiment of the present invention. FIG. 1 is a schematic diagram showing an example of a scanning trajectory of a sample. FIG. 2 is a schematic diagram showing another example of a scanning trajectory of a sample. FIG. 3 is a schematic diagram showing the relationship between the optical axis of illumination light guided obliquely to the sample surface by the illumination optical system and the illumination intensity distribution shape. FIG. 4 is a schematic diagram showing the relationship between the optical axis of illumination light guided obliquely to the sample surface by the illumination optical system and the illumination intensity distribution shape. FIG. 5 is a schematic diagram showing the configuration of a detection optical system (tilt optical system) provided in a defect inspection apparatus. FIG. 6 is a schematic diagram showing the configuration of a detection optical system (tilt optical system) provided in a defect inspection apparatus. FIG. 7 is a schematic diagram showing the configuration of a detection optical system (vertical optical system) provided in a defect inspection apparatus. FIG. 8 is an arrow view of the detection optical system (vertical optical system) taken along the X-X line. FIG. 9 is an explanatory diagram of the principle of measuring height variations of a sample surface with a height measurement unit. FIG. 10 is a schematic diagram of the three-dimensional arrangement of a sample and a tilt optical system. FIG. 11 is a diagram showing the relationship between the position in the normal direction of the sample surface, the position of the illumination spot in the main scanning direction, the output signal of a line sensor, and the output signal of an area sensor. FIG. 12 is a diagram for explaining changes in the optical image of an area sensor due to a Z-shift of the sample surface and a Y-shift of the illumination. 10 is an explanatory diagram of a method for calculating an illumination spot position in the first embodiment; FIG. 11 is an explanatory diagram of a method for calculating an illumination spot position in the second embodiment; FIG. 12 is a flowchart showing a procedure related to the operation of a defect inspection apparatus;

[0010] Hereinafter, an embodiment of the present invention will be described with reference to the drawings.

[0011] First Embodiment - Defect Inspection Apparatus - FIG. 1 is a schematic diagram of a defect inspection apparatus 100 according to one embodiment of the present invention.

[0012] The defect inspection apparatus 100 according to this embodiment inspects a sample W and detects defects such as foreign matter and dents on the surface of the sample W (hereinafter referred to as the sample surface), particularly defects of a type appropriate for the inspection purpose. A typical example of the sample W is a disk-shaped semiconductor silicon wafer with a flat surface on which no pattern is formed. The defect inspection apparatus 100 includes a stage ST, an illumination optical system 3, detection optical systems 4-0, 4-N (N=1, 2, ...), a height measurement unit 8, a signal processing device 6, a control device 71, a user interface 72 such as a mouse and keyboard, and a monitor 73 serving as a display unit.

[0013] The signal processing device 6 and the control device 71 are computers equipped with a CPU (Central Processing Unit) and memory, and control a focus actuator (an actuator 9, described later in this embodiment) according to the output of the height measurement unit 8. The signal processing device 6 and the control device 71 may be configured by different computers or may be configured by a single computer. Furthermore, the signal processing device 6 and the control device 71 may each be configured by multiple computers connected by a network.

[0014] -Stage- The stage ST is a device that moves the sample W and is composed of a sample stage ST1 and a scanning device ST2. The sample stage ST1 is a stage that supports the sample W. The scanning device ST2 is a device that drives the sample stage ST1 to change the relative position of the sample W and the illumination optical system 3 and is composed of a translation stage, a rotation stage, and a Z stage. The translation stage supports the rotation stage via the Z stage, and the sample stage ST1 is supported on the rotation stage. The translation stage translates horizontally together with the rotation stage, and the rotation stage rotates around an axis extending up and down. The Z stage ST has the function of adjusting the height of the sample surface. Note that in Figure 1, the Z direction is upward in the plane of the paper, the X direction is to the right in the plane of the paper, and the Y direction is toward the back of the paper.

[0015] FIG. 2 is a schematic diagram showing an example of a scanning trajectory of the sample W by the scanning device ST2. As will be described later, the illumination spot 40, which is a linear illumination area on the sample surface irradiated with linear illumination light emitted from the illumination optical system 3, has an illumination intensity distribution that is long in one direction, as shown in the figure. The major axis direction of the illumination spot 40 (the radial direction of the sample stage ST1) is designated as direction S2, and the direction approximately perpendicular to the major axis (the circumferential direction around the rotation axis of the sample stage ST1) is designated as direction S1. The arrows in FIG. 2 indicate the scanning trajectory of the sample W. The illumination spot 40 is linear, with its major axis direction being perpendicular (direction S2) to the main scanning direction of the sample W (direction S1).

[0016] As the rotary stage rotates, the sample W rotates, and the illumination spot 40 scans in direction S1 relative to the sample surface. As the translation stage translates, the sample W moves horizontally, and the illumination spot 40 scans in direction S2 relative to the sample surface. As the scanning device ST2 operates to translate the sample W while rotating, the illumination spot 40 moves in a spiral trajectory from the center to the outer edge of the sample W, as shown in FIG. 2, and the entire sample surface is scanned. During one rotation of the sample W, the illumination spot 40 moves in direction S2 by a distance equal to or less than the length of the illumination spot 40 in direction S2.

[0017] The scanning device ST2 may be configured to include, instead of the rotation stage, another translation stage whose movement axis extends in a direction intersecting the movement axis of the translation stage in the horizontal plane. In this case, as shown in Figure 3, the illumination spot 40 scans the sample surface by overlapping linear trajectories rather than helical trajectories. The arrows in Figure 3 indicate the scanning trajectory of the sample W.

[0018] The arrows in Figure 3 indicate that the first translation stage is translated in direction S1 at a constant speed, the second translation stage is driven in direction S2 a predetermined distance (e.g., a distance equal to or less than the length of the illumination spot 40 in direction S2), and then the first translation stage is translated back in direction S1 again. In this way, the illumination spot 40 repeatedly scans linearly in direction S1 and moves in direction S2, thereby scanning the entire surface of the sample. Compared to this scanning method, the spiral scanning method shown in Figure 2 does not involve reciprocating motion, and is therefore advantageous for inspecting the sample W in a short time.

[0019] --Illumination optical system-- Returning to the explanation of Figure 1, the illumination optical system 3 shown in Figure 1 is an optical unit including a plurality of optical elements for irradiating a sample W placed on a sample stage ST1 with linear illumination light, and forms a linear illumination spot 40 that is long in one direction (direction S2) on the sample surface.

[0020] As shown in FIG. 1, the illumination optical system 3 includes a laser light source 31, an attenuator 32, an emitted light adjustment unit 33, a beam expander 34, a polarization control unit 35, an anamorphic beam expander 39, a cylindrical lens 36, an anamorphic focusing mirror 310, and reflecting mirrors 371, 372, and 38.

[0021] The laser light source 31 is a unit that emits a laser beam as illumination light. The diameter of the laser beam emitted by the laser light source 31 is typically about 1 mm. When the defect inspection device 100 is used to detect minute defects near the sample surface, the laser light source 31 is a light source that emits a high-power laser beam with an output of 2 W or more in ultraviolet or vacuum ultraviolet light with a short wavelength (e.g., wavelength of 355 nm or less) that does not easily penetrate the interior of the sample W. In this embodiment, for example, a laser beam with a wavelength of 266 nm is used, but a laser beam with a wavelength of 200 to 380 nm or near ultraviolet light with a wavelength of 200 to 380 nm or vacuum ultraviolet light with a wavelength of 10 to 200 nm can be used depending on the purpose. Furthermore, when the defect inspection device 100 is used to detect defects inside the sample W, the laser light source 31 is a light source that emits a visible or infrared laser beam with a long wavelength that easily penetrates the interior of the sample W.

[0022] The emitted light adjustment unit 33 is a unit that adjusts the angle of the optical axis of the illumination light attenuated by the attenuator 32, and in this embodiment, is configured to include a plurality of reflecting mirrors 331 and 332. The reflecting mirrors 331 and 332 sequentially reflect the illumination light, and in this embodiment, the incident / exit plane of the illumination light with respect to the reflecting mirror 331 is configured to be perpendicular to the incident / exit plane of the illumination light with respect to the reflecting mirror 332.

[0023] The incident / exit plane is a plane that includes the optical axis that enters a certain reflecting mirror and the optical axis that exits from this reflecting mirror. For example, if a three-dimensional XYZ Cartesian coordinate system different from the XYZ coordinate system of Figure 1 is defined here, and illumination light is incident on reflecting mirror 331 in the +X direction, the illumination light is redirected, for example, in the +Y direction by reflecting mirror 331 and then in the +Z direction by reflecting mirror 332. In this example, the incident / exit plane of the illumination light for reflecting mirror 331 is the XY plane, and the incident / exit plane for reflecting mirror 332 is the YZ plane.

[0024] The reflecting mirrors 331 and 332 are each provided with a mechanism (not shown) for translating and a mechanism (not shown) for tilting the reflecting mirrors 331 and 332. The reflecting mirrors 331 and 332, for example, translate in the incident or outgoing direction of the illumination light relative to themselves, and tilt around the normal to the incident and outgoing surfaces. This allows, for example, the offset amount and angle in the XZ plane and the offset amount and angle in the YZ plane of the optical axis of the illumination light emitted in the +Z direction from the outgoing light adjustment unit 33 to be independently adjusted. In this embodiment, a configuration using two reflecting mirrors 331 and 332 is illustrated, but a configuration using three or more reflecting mirrors may also be used.

[0025] The beam expander 34 is a unit that expands the beam diameter of the incident illumination light, and includes a plurality of lenses 341 and 342. An example of the beam expander 34 is a Galilean type that uses a concave lens as the lens 341 and a convex lens as the lens 342. The beam expander 34 is provided with a mechanism (zoom mechanism) for adjusting the distance between the lenses 341 and 342, and the expansion rate of the beam diameter can be changed by adjusting the distance between the lenses 341 and 342.

[0026] The beam expander 34 expands the beam diameter by, for example, about 2 to 5 times, and in this case, if the beam diameter of the illumination light emitted from the laser light source 31 is 1 mm, the beam diameter of the illumination light is expanded to about 2 to 5 mm. If the illumination light incident on the beam expander 34 is not a parallel beam, it is possible to collimate the beam (make the beam quasi-parallel) in addition to adjusting the beam diameter by adjusting the distance between the lenses 341 and 342. However, the collimation of the beam may be achieved by installing a collimating lens separately from the beam expander 34 upstream of the beam expander 34.

[0027] The anamorphic beam expander 39 is a unit that expands the beam diameter of the illumination light in one direction to linearize the illumination light. Examples of the anamorphic beam expander 39 include an anamorphic prism pair that combines multiple prisms, or a cylindrical beam expander that combines multiple cylindrical lenses. The anamorphic beam expander 39 expands the beam diameter of the illumination light in the direction corresponding to the direction S2 on the sample W by approximately 3 to 15 times, to approximately 15 mm to 30 mm.

[0028] The polarization control unit 35 is an optical system that controls the polarization state of the illumination light and includes a half-wave plate 351 and a quarter-wave plate 352. For example, when performing oblique incidence illumination by inserting a reflecting mirror 371 (described later) into the optical path, the polarization control unit 35 polarizes the illumination light to P-polarize it, thereby increasing the amount of scattered light from defects on the sample surface compared to light polarized other than P-polarized light. If scattered light (called haze) from minute irregularities on the sample surface interferes with the detection of minute defects, the illumination light can be polarized to S-polarize it, thereby reducing the haze compared to light polarized other than S-polarized light. The polarization control unit 35 can also polarize the illumination light to circular polarization or 45-degree polarization, which is intermediate between P-polarization and S-polarization.

[0029] The cylindrical lens 36 is a unit that focuses the illumination light in a direction S2 on the sample W and adjusts the longitudinal dimension of the illumination spot 40.

[0030] The anamorphic focusing mirror 310 has a curvature in a direction corresponding to the direction S1 on the sample W, and is a unit that focuses the illumination light in the direction S1 on the sample W to adjust the short-side dimension of the illumination spot 40. This anamorphic focusing mirror 310 is a focusing element that focuses the laser irradiation light onto the irradiation position on the sample stage surface at an angle inclined with respect to the normal to the sample stage surface. It does not have a curvature in a direction perpendicular to the direction S1 on the sample W. The curved surface shape of the anamorphic focusing mirror 310 in the direction S1 on the sample W is aspherical to suppress spherical aberration, and specifically, is a parabolic surface that is optimal for focusing collimated incident light.

[0031] 1 , the reflecting mirror 371 can be translated by a drive mechanism (not shown) to move in and out of the optical path of the illumination light directed toward the sample W, thereby switching the incidence path of the illumination light with respect to the sample W. By inserting the reflecting mirror 371 into the optical path, the illumination light emitted from the polarization control unit 35 is reflected by the reflecting mirror 371 and passes through the anamorphic beam expander 39, the cylindrical lens 36, the reflecting mirror 38, and the anamorphic focusing mirror 310 to be incident obliquely on the sample W. On the other hand, by removing the reflecting mirror 371 from the optical path, the illumination light emitted from the polarization control unit 35 passes through the reflecting mirror 372, the polarizing beam splitter 48, the polarization control unit 47, the reflecting mirror 46, and the detection optical system 4-0 to be incident perpendicularly on the sample W.

[0032] 4 and 5 are schematic diagrams showing the relationship between the optical axis OA1 of the illumination light guided obliquely to the sample surface by the illumination optical system 3 and the illumination intensity distribution shape. FIG. 4 shows a schematic cross-section of the sample W cut along the plane of incidence of the illumination light incident on the sample W. FIG. 5 shows a schematic cross-section of the sample W cut along a plane that is perpendicular to the plane of incidence of the illumination light incident on the sample W and includes the normal to the sample surface. The plane of incidence is a plane that includes the optical axis OA1 of the illumination light incident on the sample W and the normal to the sample surface, and is a plane that is perpendicular to the sample surface with a linear illumination spot 40. Note that FIGS. 4 and 5 show only a portion of the illumination optical system 3, and for example, the exit light adjustment unit 33 and reflecting mirrors 371, 372, and 38 are not shown.

[0033] When the reflecting mirror 371 is inserted into the optical path, the illumination light emitted from the laser light source 31 is focused by the cylindrical lens 36 and adjusted to a desired intensity distribution, and is reflected by the reflecting mirror 38 to be obliquely incident on the sample W. In this manner, the illumination optical system 3 is configured to allow illumination light to be incident on the sample W from a direction oblique to the normal to the sample surface. This oblique incidence illumination is adjusted so that the illumination intensity distribution becomes a Gaussian distribution within the incident plane by adjusting the light intensity by the attenuator 32, the beam diameter by the beam expander 34, the polarization by the polarization control unit 35, and the intensity distribution by the cylindrical lens 36. As shown in FIG. 4 , the illumination spot 40 formed on the sample W has a Gaussian illumination intensity distribution in the direction S2. The beam width L1 in the direction S2, defined at 13.5% of the peak, is, for example, approximately 25 μm to 1600 μm.

[0034] In a plane perpendicular to the incident surface and the sample surface, the illumination spot 40 has a light intensity distribution in which the intensity is weaker at the periphery than at the center of the light beam, as shown in the illumination intensity distribution (illumination profile) LD2 shown in Fig. 5. Specifically, the intensity distribution is similar to a Gaussian distribution that reflects the intensity distribution of the light incident on the anamorphic collector mirror 310, or a first-order Bessel function of the first kind or a sinc function that reflects the aperture shape of the anamorphic collector mirror 310.

[0035] The width L2 of the illumination intensity distribution in the plane perpendicular to the incident surface and the sample surface is set to, for example, approximately 1.5 μm to 10 μm, narrower than the beam width L1 shown in FIG. 4 , in order to reduce haze generated from the sample surface. To make width L2 narrower than beam width L1, the NA (numerical aperture) of the light focused in direction S2 is larger than that in direction S1. This is because the anamorphic beam expander 39 expands the incident beam diameter in the focused direction S2 to a dimension larger than that in direction S1, and the focal length (optical path length of the focused beam) in direction S2 is shorter than that in direction S1. The width L2 of this illumination intensity distribution is the length of the region in the plane perpendicular to the incident surface and the sample surface where the illumination intensity is 13.5% or more of the maximum illumination intensity.

[0036] Furthermore, the angle of incidence of the oblique incidence illumination on the sample W (the tilt angle of the incident optical axis relative to the normal to the sample surface) is adjusted to an angle suitable for detecting minute defects by adjusting the positions and angles of the reflecting mirrors 371 and 38. The angle of the reflecting mirror 38 is adjusted by the adjustment mechanism 381. For example, the larger the angle of incidence of the illumination light on the sample W (the smaller the illumination elevation angle, which is the angle between the sample surface and the incident optical axis), the weaker the haze that causes noise in the scattered light from minute foreign particles on the sample surface, making it more suitable for detecting minute defects. From the perspective of suppressing the effect of haze on the detection of minute defects, it is preferable to set the incident angle of the illumination light to, for example, 75 degrees or more (elevation angle 15 degrees or less). On the other hand, with oblique incidence illumination, the absolute amount of scattered light from minute foreign particles increases as the illumination incident angle decreases. Therefore, from the perspective of increasing the amount of scattered light from defects, it is preferable to set the incident angle of the illumination light to, for example, 60 degrees or more and 75 degrees or less (elevation angle 15 degrees or more and 30 degrees or less).

[0037] --Detection Optical System--Returning to the explanation of Figure 1, the detection optical systems 4-0, 4-N (N = 1, 2, ...) are optical units that detect scattered light generated by the illumination spot 40 on the sample surface by focusing it onto the line sensor 5-0, which is a vertical sensor for inspection, the line sensor 5-N (N = 1, 2, ...), which is an oblique sensor for inspection, the line sensor 5b, which is a vertical sensor for detecting large foreign objects, the area sensor 6-0 (Figure 8), which is a vertical sensor for adjustment, and the area sensor 6-N (N = 1, 2, ...), which is an oblique sensor for adjustment. These optical units include multiple optical elements, including a condenser lens (objective lens). Each objective lens of the detection optical systems 4-0, 4-N is arranged along the hemispherical surface of a sphere (celestial sphere) centered at the center of the illumination spot 40 on the sample W. The scattered light incident on the detection optical systems 4-0, 4-N is condensed and guided to the line sensors 5-0, 5-N and the area sensors 6-0, 6-N, respectively.

[0038] Oblique Detection Optical System Figures 6 and 7 are schematic diagrams of a detection optical system 4-N whose optical axis is tilted with respect to the sample surface. Figure 6 is a schematic cross-section (viewed in the -Y direction in Figure 1) of the detection optical system 4-N cut at the reflecting surface where scattered light incident on the detection optical system 4-N is reflected by the sample W. Figure 7 is a schematic diagram (viewed in the -Z direction in Figure 1) of the detection optical system 4-N viewed from a direction intersecting the optical axis of the detection optical system 4-N along the reflecting surface. The reflecting surface is a plane that includes the optical axis OA2 of the scattered light incident on the detection optical system 4-N and the normal to the sample surface. The image plane Pi (Figure 11) conjugate to the sample surface formed by the detection optical system 4-N is tilted with respect to the optical axis OA2 of the detection optical system 4-N.

[0039] As shown in Figures 6 and 7, the detection optical system 4-N has an optical axis OA2 inclined with respect to the normal n of the sample surface. The detection optical system 4-N guides scattered light from the illumination spot 40 to the corresponding line sensor 5-N, forming an optical image OI (a linear illumination image, which is an image of the irradiated area) of the illumination spot 40 on the line sensor 5-N. The detection optical system 4-N is a double-telecentric optical system configured so that the imaging magnification does not change even when the working distance changes. Along its optical axis OA2, the detection optical system 4-N includes a condenser lens (objective lens) 411, a half-wave plate 412, a polarizing beam splitter 42, a half-wave plate 431, cylindrical lenses 433 and 434, an imaging lens 435, and a line sensor 5-N. The scattered illumination light incident on the detection optical system 4-N is guided to the respective line sensors 5-N. The detection optical system 4-N also includes a beam diffuser 44 in the direction of propagation of the light split by the polarizing beam splitter 42.

[0040] In the detection optical system 4-N, the illumination scattered light is collected by a collecting lens 411, and the polarization direction of the light is controlled by a half-wave plate 412. The half-wave plate 412 can be rotated by an actuator (not shown).

[0041] The light that has passed through the half-wave plate 412 has its optical path split according to its polarization by the polarizing beam splitter 42. The combination of the half-wave plate 412 and the polarizing beam splitter 42 makes it easy to separate optical signals that indicate defects in the sample W from optical signals that hinder detection of defects in the sample W (roughness scattered light from the sample surface).

[0042] The light that passes through the polarizing beam splitter 42 is controlled by the half-wave plate 431 to have a polarization direction suitable for detection by the line sensor 5-N. On the other hand, the light whose optical path is split by the polarizing beam splitter 42 is attenuated by the beam diffuser 44 to prevent it from becoming stray light.

[0043] The cross-sectional shape of the light passing through the half-wave plate 431 is adjusted by cylindrical lenses 433 and 434. The cylindrical lenses 433 and 434 form a cylindrical beam expander, and the extent of the optical image OI formed on the light-receiving surface of the line sensor 5-N in the short-side direction γ is adjusted to be smaller than the extent of the optical image OI in the long-side direction δ. The long-side direction σ of the light-receiving surface of the line sensor 5-N (the direction in which the pixels on the light-receiving surface are arranged) coincides with the long-side direction of the optical image OI, and by reducing the image height (width) of the optical image OI in the short-side direction γ using the cylindrical lenses 433 and 434, the occurrence of defocus in the short-side direction γ is reduced. The light beam whose cross-sectional shape has been adjusted by the cylindrical lenses 433 and 434 is guided to the line sensor 5-N via an imaging lens 435, and the optical image OI of the illumination spot 40 is formed on multiple pixels of the line sensor 5-N. The detection signals of the optical image photoelectrically converted by each pixel of the line sensor 5 -N are output to the signal processing device 6 .

[0044] In this way, the detection optical system 4-N collects the scattered light from the illumination spot 40 irradiated onto the sample W by the illumination optical system 3, controls the polarization state of the incident scattered light, and forms an optical image OI of the illumination spot 40 on the light receiving surface of the corresponding line sensor 5-N.

[0045] The light-receiving surface of line sensor 5-N is entirely conjugate with the sample surface in the longitudinal direction δ, whereas in the lateral direction γ, only illumination spot 40, which is a linear illumination area, is conjugate with the sample surface. Specifically, as shown in Fig. 11, the light-receiving surface of line sensor 5-N is disposed so that image plane Pi, which is two-dimensionally conjugate with the sample surface, intersects with optical image OI, which is a linear illumination image (e.g., the major axis of optical image OI), and is inclined with respect to image plane Pi, with the major axis of the linear illumination image serving as the intersection line Li. The light-receiving surface of line sensor 5-N is parallel to the major axis of optical image OI and one-dimensionally conjugate with the sample surface.

[0046] Vertical Detection Optical System FIG. 8 is a diagram showing the configuration of a detection optical system 4-0 onto which scattered light emitted in the normal direction from the sample W is incident, and FIG. 9 is a view taken along the line XX in FIG.

[0047] The detection optical system 4-0 includes a condenser lens (objective lens) 451 and an imaging lens 452, and the scattered light condensed by the condenser lens 451 is guided by the imaging lens 452 to the line sensor 5-0.

[0048] In this detection optical system 4-0, a reflecting mirror 46 is disposed at the position of its own pupil between the condenser lens 451 and the imaging lens 452. In the case of epi-illumination in which the reflecting mirror 371 shown in FIG. 1 is removed from the optical path, the illumination light is incident on the sample W from the normal direction via the reflecting mirror 46. In this way, the condenser lens 451 of the detection optical system 4-0 also serves as a condenser lens that guides the epi-illumination to the sample W.

[0049] On the other hand, the reflecting mirror 46 also plays a role in branching the optical path of a portion of the scattered light that enters the detection optical system 4-0 from the illumination spot 40 by oblique incidence illumination or epi-illumination. As mentioned above, the illumination spot 40 has a linear intensity distribution that is long in the direction S2. As shown in FIG. 9 , the reflecting mirror 46 is longer than the illumination spot 40 in the direction S1, which is the minor axis direction of the linear illumination spot 40, as viewed from the line sensor 5-0, and shorter than the illumination spot 40 in the direction S2, which is the major axis direction of the illumination spot 40. As a result, scattered light that enters the detection optical system 4-0 from the sample W and does not interfere with the reflecting mirror 46 enters the line sensor 5-0 via the imaging lens 452, and scattered light that interferes with the reflecting mirror 46 is reflected by the reflecting mirror 46.

[0050] The scattered light incident from the sample W into the detection optical system 4-0 and reflected by the reflecting mirror 46 is guided to the line sensor 5b via the polarization control unit 47, the polarizing beam splitter 48, and the imaging lens 49. Similar to the polarization control unit 35, the polarization control unit 47 includes a quarter-wave plate 471 and a half-wave plate 472, and can adjust the polarization of the scattered illumination light incident from the reflecting mirror 46 to any desired polarization. The polarization control unit 47 corresponds to a polarizing element that can switch the polarization of the illumination light between S-polarized light (first polarization) and P-polarized light (second polarization orthogonal to the first polarization). During oblique incidence illumination, the polarization of the scattered illumination light incident on the polarizing beam splitter 48 is controlled to linear polarization by the quarter-wave plate 471 of the polarization control unit 47, so that the scattered illumination light reflected by the reflecting mirror 46 passes through the polarizing beam splitter 48 and enters the imaging lens 49. The line sensor 5b uses an imaging element, such as a silicon photomultiplier (SiPM), that is highly sensitive and faster than the other line sensors 5-0 and 5-N. Because its accumulation time is shorter than the other line sensors 5-0 and 5-N, it has a wide dynamic range and can measure the scattered light intensity of large foreign particles and defects (hereinafter referred to as large foreign particles) without saturating the light intensity, which would otherwise saturate other line sensors. Based on the detection signal from the line sensor 5b, the size of the large foreign particle is estimated and explosion / ablation of the large foreign particle is avoided. Explosion / ablation is avoided by detecting the large foreign particle in advance during a rotation before passing through the high-intensity region near the center of the illumination spot 40 and temporarily reducing the illumination intensity when the large foreign particle passes near the center of the illumination spot 40. The line sensor 5b can also be replaced with a point sensor that does not have spatial resolution. Using a point sensor makes it impossible to measure the precise position of the large foreign particle within the illumination spot 40's illumination range, but it is possible to detect the presence or absence of the large foreign particle within the illumination range. Compared to line sensors, this has the advantages that the sensor response and output speed can be made faster and the output signal processing system can be realized at low cost.

[0051] The polarization control unit 47 controls the polarization of the illumination light in response to a command signal from the control device 71 so that, under epi-illumination conditions with the reflecting mirror 371 removed from the optical path, the illumination light traveling toward the sample W is incident on the detection optical system 4-0 as polarized light in an arbitrary direction (for example, circularly polarized light). The control device 71 has a function of controlling the polarization control unit 47 so that the intensity within a set region of the intensity distribution of the detected optical image falls within a set range.

[0052] Line Sensors (Inspection Sensors) The line sensors 5-0 and 5-N are one-dimensional image sensors that detect scattered light generated from the linear illumination area on the sample surface irradiated with illumination light, i.e., the illumination spot 40. These line sensors 5-0 and 5-N are line sensors with an array-like light-receiving surface in which multiple pixels are arranged in a line, and correspond to the detection optical systems 4-0 and 4-N, respectively. CMOS image sensors or CCD image sensors are used for these line sensors 5-0 and 5-N. The detection optical systems 4-0 and 4-N form an optical image OI of the illumination spot 40 on the light-receiving surface of the corresponding line sensor 5-0 and 5-N. The line sensors 5-0 and 5-N photoelectrically convert the optical image OI formed on the light-receiving surface, perform predetermined sampling, convert the analog electrical signal into digital data, and output it to the signal processing device 6 as a data set of scattered light intensity of the optical image OI.

[0053] A portion of the scattered light collected by the detection optical system 4-0 is guided to the line sensor 5b in addition to the line sensor 5-0. The line sensor 5b also photoelectrically converts the optical image collected by the detection optical system 4-0, performs predetermined sampling, converts the electrical signal into digital data by analog-to-digital conversion, and outputs the digital data to the signal processing device 6.

[0054] In this embodiment, the light-receiving surface of the line sensor 5-N, which is an oblique sensor, is inclined with respect to the optical axis OA2 in accordance with the inclination of the optical axis OA2 of the corresponding detection optical system 4-N with respect to the sample surface, and the long axis of the light-receiving surface coincides with a position conjugate with the long axis of the illumination spot 40 irradiated on the sample surface. However, the line sensor 5-0, which faces the illumination spot 40 in the normal direction to the sample surface, is different from the line sensor 5-N. The light-receiving surface of the line sensor 5-0 is perpendicular to the optical axis OA2 of the detection optical system 4-0. The line sensors 5-N are each arranged so that the long axis (the center line extending in the longitudinal direction) of their light-receiving surfaces is parallel to the long axis of the optical image OI ( FIG. 6 ) of the illumination spot 40, and the entire illumination spot 40 fits within the one-dimensional light-receiving surface formed by arranging pixels in an array.

[0055] 1 is provided with an actuator 9 for each line sensor 5-N that moves the line sensor 5-N. The actuator 9 can be, for example, a piezo actuator, and can shift the corresponding line sensor 5-N with good response, thereby moving the corresponding line sensor 5-N three-dimensionally, for example, in parallel. In this embodiment, the actuator 9 serves as a focus actuator that moves the light receiving surface of the corresponding line sensor 5-N relative to the focal position of the optical image OI of the corresponding detection optical system 4-N.

[0056] Area Sensor (Adjustment Sensor) The area sensors 6-0 and 6-N are two-dimensional image sensors that capture a two-dimensional image of the illumination spot 40 on the sample surface and observe the enlarged image of the sample surface projected by the detection optical systems 4-0 and 4-N. The area sensors 6-0 and 6-N detect the position and intensity distribution of the optical image OI of the illumination spot 40, and adjust the position and focus of the illumination spot 40 based on the detected position and intensity distribution. In order to measure the position, width, and length of the illumination spot 40 with high precision, the area sensors 6-0 and 6-N are CMOS (Complementary Metal-Oxide-Semiconductor) area sensors or CCD (Charge Coupled Device) area sensors with a smaller pixel pitch than the line sensors 5-0 and 5-N.

[0057] These area sensors 6-0 and 6-N are disposed at positions where the optical distance from the sample surface mounted on the stage ST is equal to the optical distance between the corresponding line sensor (a line sensor to which scattered light is guided by the same detection optical system) and the sample surface. In other words, the optical image OI of the illumination spot 40 is simultaneously formed on both the corresponding line sensor and area sensor by the same detection optical system. The light-receiving surfaces of the area sensors 6-0 and 6-N are two-dimensionally conjugate with the sample surface (strictly speaking, the sample-side focal plane fp (FIG. 10) described later) and are optically parallel to the image plane Pi.

[0058] A part of the scattered light beam reflected by a half mirror 436 arranged between cylindrical lenses 433 and 434 of the detection optical system 4-N is guided to the area sensor 6-N of the oblique optical system via an imaging lens 437, and an optical image OI of the illumination spot 40 is formed on a plurality of pixels arranged two-dimensionally on the area sensor 6-N. A detection signal of the optical image OI photoelectrically converted by each pixel of the area sensor 6-N is output to the signal processing device 6.

[0059] An area sensor 6-0 (FIG. 8) of the vertical optical system is installed in place of the line sensor 5-0 at the position of the line sensor 5-0 or at a position conjugate with the line sensor 5-0 by a drive mechanism (not shown). When capturing an image using the area sensor 6-0, the reflecting mirror 46 is retracted from the optical path by a drive mechanism (not shown) so as not to reduce the resolution of the optical image OI.

[0060] - Height measurement unit - The height measurement unit 8 shown in Figure 1 is a unit for measuring the height of the illumination spot 40 on the sample surface, and is composed of a condenser lens 81, an imaging lens 82, and a two-dimensional sensor 83. The two-dimensional sensor 83 uses a two-dimensional CCD image sensor, a CMOS image sensor, or a PSD (position sensing detector). The two-dimensional sensor 83 photoelectrically converts the optical image OI formed on the light-receiving surface by the condenser lens 81 and the imaging lens 82, converts the electrical signal into digital data by analog-to-digital conversion, and outputs it to the control device 71. The output of the height measurement unit 8 may be input to the signal processing device 6, and then input to the control device 71 via the signal processing device 6.

[0061] 10 is an explanatory diagram of the principle of measuring the height fluctuation of the sample surface by the height measurement unit 8 shown in FIG. 1. The figure shows a case where the sample W has moved by Δn in the normal direction of the sample surface. In FIG. 10, the condenser lens 81 is omitted from the illustration.

[0062] Specularly reflected light of illumination light incident obliquely on the sample W at an elevation angle θ enters the two-dimensional sensor 83 via the imaging lens 82. Therefore, when the height of the sample W changes from the position indicated by the dashed line to the position indicated by the solid line, the trajectory of the specularly reflected light moves in parallel, and the incident position of the specularly reflected light on the two-dimensional sensor 83 changes. The two-dimensional sensor 83 outputs an electrical signal from the pixel on which the specularly reflected light is incident, i.e., an electrical signal corresponding to the incident position of the specularly reflected light. The incident position of the specularly reflected light when the illumination spot 40 on the sample W is at a reference height (when the height fluctuation Δn = 0) is defined as the reference position. As the height of the sample surface changes, the incident position of the illumination spot 40 on the sample surface shifts from the reference position. The amount of this shift can be used to measure the height fluctuation of the sample surface at the position of the illumination spot 40. The two-dimensional sensor 83 can also be replaced by a one-dimensional sensor (linear sensor) in which pixels are aligned in the direction in which the incident position of the specularly reflected light shifts due to height fluctuation.

[0063] -Signal Processing Device- Returning to the explanation of Figure 1, the signal processing device 6 is a computer that processes the detection signals output by the line sensors 5-0 and 5-N and discriminates defects on the sample surface, and is configured to include a ROM (Read Only Memory), a RAM (Random Access Memory), other memories, a CPU, an FPGA (Field-Programmable Gate Array), a timer, etc.

[0064] As an example, the signal processing device 6 is assumed to be configured as a single computer that forms a unit with the main body of the defect inspection device 100 (such as the stage, illumination optical system, detection optical system, line sensor, area sensor, etc.), but it may also be configured as multiple computers connected via a network. In this case, a server can be used as one of the multiple computers. This is an example in which a server is included as a component of the defect inspection device 100. For example, a configuration is possible in which a computer attached to the main body of the device acquires defect detection signals from the main body of the device, processes the detection data as necessary, and sends it to a server, and the server performs processes such as defect detection and classification.

[0065] The signal processing device 6 includes a subpixel interpolation processing unit 61, a storage unit (memory) 62, a signal integration unit 63, and a defect detection unit 64. The subpixel interpolation processing unit 61, the signal integration unit 63, and the defect detection unit 64 may be realized by hardware such as an FPGA circuit, or by a software program. Furthermore, at least some of these functions of the signal processing device 6 (particularly downstream processing) may be executed by a GPU (Graphics Processing Unit) or CPU installed in the server.

[0066] The subpixel interpolation processing unit 61 analyzes the position of the illumination spot 40 on the sample surface from the detection signals of the line sensors 5-0 and 5-N. The memory unit 62 stores the detection signals (digital data) input from the line sensors 5-0 and 5-N, the position data calculated by the subpixel interpolation processing unit 61, and other data, and accumulates them as scattered light data. The signal integration unit 63 integrates and calculates multiple scattered light data for different positions of the illumination spot 40 output from the same sensor based on the scattered light data accumulated in the memory unit 62, and also integrates and calculates scattered light data similarly integrated for different line sensors. The defect detection unit 64 extracts high-frequency, high-brightness areas on the sample surface as defects based on the scattered light data after the integration calculation (performs defect inspection).

[0067] To further explain the processing performed by the signal combining unit 63, the signal combining unit 63 adds up scattered light intensities for data obtained by scanning the same coordinates on the sample surface at different longitudinal positions of the linear illumination spot 40. In other words, when the sample W is spirally scanned, if coordinates on the sample surface are expressed in the rθ coordinate system, scattered light intensities for the same coordinates are added up for scattered light data having the same θ coordinate of the illumination spot 40 but different r coordinates. In the rθ coordinate system, the r coordinate is the radial coordinate on the sample surface, and the θ coordinate is the azimuthal coordinate on the sample surface, which is conceptually different from the elevation angle θ in FIG. 10 . Even for scattered light data output from the same sensor, for data with different r coordinates, the imaging position of scattered light from the same coordinates (i.e., pixels receiving scattered light from the same coordinates) changes depending on the movement of the sample stage ST1 during one rotation of the sample W. The movement of the sample stage ST1 per rotation is smaller than the beam width L1 shown in FIG. 4 . Therefore, based on the amount of movement of the sample stage ST1, the correspondence between the pixels that output the signals to be added as the scattered light intensity at the same coordinates is calculated in advance.

[0068] -Control Device- The control device 71 is a computer that controls the defect inspection device 100, such as the stage ST and the illumination optical system 3, and, like the signal processing device 6, is configured to include a CPU, FPGA, timer, etc. in addition to ROM, RAM, and other memories. The control device 71 is connected to the user interface 72, the monitor 73, and the signal processing device 6 by wire or wirelessly.

[0069] The configuration is not limited to this, and the functions of the signal processing device 6 may be incorporated into the control device 71, so that the control device 71 also functions as the signal processing device 6. The user interface 72 is a device through which the user inputs various operations, and various input devices such as a keyboard, mouse, touch panel, etc. may be appropriately adopted. The monitor 73 may be various display devices such as a liquid crystal display, etc.

[0070] The control device 71 receives the output of the height measurement unit 8, the encoders of the rotation stage and the translation stage, and inspection conditions input from a user interface 72 in response to an operator's operation. The inspection conditions include, for example, the type, size, shape, and material of the sample W, illumination conditions, detection conditions, and the like.

[0071] Furthermore, the control device 71 outputs signals to command the operation of the actuator 9, the detection optical systems 4-0 and 4-N, the stage ST, and the illumination optical system 3 in accordance with the height fluctuation of the sample W and the inspection conditions, and outputs coordinate data of the illumination spot 40 synchronized with the defect detection signal to the signal processing device 6. The control device 71 also drives the detection optical system 4-N, the stage ST, and the illumination optical system 3 in relation to the calibration of the illumination spot 40. The control device 71 also displays and outputs the defect inspection results obtained by the signal processing device 6 on a monitor 73.

[0072] --Arrangement of tilting optical system-- Figure 11 shows a schematic diagram of the three-dimensional arrangement of the sample W and tilting optical system. The optical axis OA2 of each detection optical system 4-N is tilted at an angle θ with respect to the normal to the sample W (the direction of normal n). The projection of the optical axis OA2 on the sample surface is tilted at an angle φ with respect to the major axis (direction S2) of the illumination spot 40. In this specification, θ, as used hereinafter, represents the tilt angle of the optical axis OA2 of the tilting optical system with respect to the normal n to the sample surface, and is conceptually different from the θ representing the elevation angle of the illumination light in Figure 10 and the θ in the rθ coordinate system described above.

[0073] When the optical axis OA2 of the detection optical system 4-N is tilted at an angle θ with respect to the normal n of the sample W and the projection of the optical axis OA2 on the sample surface is tilted at an angle φ with respect to the major axis of the illumination spot 40, the vector v0 of the optical axis OA2 in three-dimensional space is expressed by the following equation (1): v0=(sin θ·sin φ, sin θ·sin φ, cos φ·cos θ) (1)

[0074] The angle α between this vector v0 and the vector v2 (direction S2) of the major axis of the illumination spot 40 can be calculated by the following equation (2): α=arccos(sin θ·cos φ) (2)

[0075] Under the condition that the height of the sample surface is at the reference position (Δn=0), the optical image OI of the illumination spot 40 fits within the light receiving surface of the line sensor 5-N without correcting the positions of the detection optical system 4-N or the line sensor 5-N. If the length of the major axis of the illumination spot 40 is 2L, there is a difference of Δz shown in equation (3) in the working distance (the distance between the sample W and the detection optical system 4-N) between the center of the field of view on the sample surface and a point a distance x away from the center of the field of view. Δz=x(sin θ·cos φ, |x|<L...(3)

[0076] The imaging magnification M is determined by the condenser lens 411 and the imaging lens 435. Using this imaging magnification M, the position of the optical image OI on the light receiving surface of the line sensor 5-N at a point on the sample surface that is a distance x away from the center of the field of view is expressed by equation (4): ΔZ=M ２ x (sinθ・cosφ), |x|<L…(4)

[0077] Generally, a line sensor is arranged so that its light-receiving surface is perpendicular to the center line (optical axis) of the light beam emitted by the imaging lens. In contrast, in this embodiment, as shown in Figure 6, the light-receiving surface of the line sensor 5-N is tilted with respect to the optical axis OA2, thereby suppressing defocusing of the optical image OI due to the difference Δz in working distance that occurs within the field of view of the detection optical system 4-N, which is an inclined optical system.

[0078] The vector v1 of the long axis (center line extending in the longitudinal direction) of the light receiving surface shown in FIG. 6 is included in the same plane as the vector v2 (direction S2) of the long axis of the illumination spot 40 and the vector v0 of the optical axis OA2, and the angle β with respect to the vector v0 is set to satisfy the formula (5): tan α=M tan β (5)

[0079] In this case, the angle α between the vectors v0 and v2 satisfies the relationship in equation (6): cos α=sin θ·cos φ (6)

[0080] As the imaging magnification M increases, the angle β between vectors v0 and v1 decreases according to equation (5), and the angle of incidence of the reflected light on line sensor 5-N increases. When imaging magnification M is increased to 2, the scattered light enters line sensor 5-1 at an angle of incidence approaching 90 degrees at most. However, since the transmittance of the anti-reflection coating formed on the light-receiving surface of line sensor 5-1 depends on the angle of incidence of the light rays, and at angles of incidence approaching 90 degrees, the transmittance decreases and the light-receiving sensitivity of line sensor 5-N decreases, it is desirable to set imaging magnification M to 2 or less. Hereinafter, in this embodiment, imaging magnification M is assumed to be 1.

[0081] When the polarization of the illumination irradiating the defect and the substrate surface changes, the intensity and polarization of the scattered light from the defect and the substrate surface change, which changes the intensity ratio of the scattered light from the defect and the sample surface detected by the line sensor 5-N of the detection optical system 4-N, and therefore the defect detection performance changes.

[0082] Relationship Between Illumination Spot Position and Sensor Output Figure 12 shows the relationship between the sample surface position in the normal n direction, the position of the illumination spot 40 in the main scanning direction (direction S1), the output signal of the line sensor 5-N, and the output signal of the area sensor 6-N. Hereinafter, the deviation of the illumination spot 40 in the normal n direction from the linear object point area imaged on the line sensor 5-N will be referred to as the "sample surface Z deviation." A linear object point area is an area on the sample-side focal plane of the detection optical system 4-N where object points imaged on the line sensor 5-N exist in a line. Furthermore, the deviation of the illumination spot 40 in the main scanning direction (direction S1) from the object point area will be referred to as the "illumination Y deviation." The directions represented by Z and Y in the sample surface Z deviation and illumination Y deviation correspond to the Z and Y directions, respectively, of the XYZ coordinate system shown in Figure 1. The Z direction corresponds to the normal n direction, and the Y direction corresponds to direction S1 (the minor axis direction of the illumination spot 40).

[0083] For example, if the surface of the sample W (sample surface) indicated by the dashed line in Figure 10 is the sample-side focal plane fp of the detection optical system 4-N, then if either a sample surface Z shift or an illumination Y shift occurs with respect to the linear object point area fl imaged on the line sensor 5-N on this sample-side focal plane fp, the optical image OI of the illumination spot 40 captured by the line sensor 5-N will be defocused. If neither a sample surface Z shift nor an illumination Y shift occurs (if the linear illumination spot 40 coincides three-dimensionally with the object point area fl), the light incident on the detection optical system 4-N from the illumination spot 40 will be focused on the light-receiving surface of the line sensor 5-N, and the output signal from the line sensor 5-N will have a high intensity and a narrow intensity distribution, resulting in a clear optical image OI. In contrast, if a sample surface Z shift or an illumination Y shift occurs, the output signal from the line sensor 5-N related to the light from the illumination spot 40 will have a reduced peak intensity and a wide intensity distribution, resulting in a blurred optical image OI.

[0084] However, line sensor 5-N is a one-dimensional image sensor, and its pixels are aligned only in the direction corresponding to the long axis direction of illumination spot 40, so it cannot obtain positional deviation information in the short axis direction of illumination spot 40. Therefore, even when there is illumination Y deviation, a signal similar to that when there is sample surface Z deviation is output. Therefore, it is difficult to determine from the output signal of line sensor 5-N whether the cause of defocus is sample surface Z deviation or illumination Y deviation.

[0085] On the other hand, the area sensor 6-N is a two-dimensional image sensor, and has a plurality of pixels arranged not only in the direction corresponding to the major axis direction of the illumination spot 40 but also in the direction corresponding to the minor axis direction. Therefore, the data of the optical image OI2 obtained by the area sensor 6-N includes data on the intensity distribution of scattered light, as well as data on the position of the optical image OI2 on the light-receiving surface (the position in the direction corresponding to the major axis direction and the direction corresponding to the minor axis direction of the illumination spot 40). The displacement of the optical image OI2 on the light-receiving surface of the area sensor 6-N corresponds to the displacement of the illumination spot 40 on the sample surface.

[0086] If there is no sample surface Z deviation or illumination Y deviation (if the illumination spot 40 coincides with the object point region fl three-dimensionally), the light from the illumination spot 40 is focused on a position on the light-receiving surface of the area sensor 6-N corresponding to the object point region fl, and the optical image OI2 is positioned on the reference line fl' (e.g., the center line of the field of view of the area sensor 6-N) within the field of view of the area sensor 6-N, as shown in the upper part of Figure 12. In this case, the optical image OI2 output from the area sensor 6-N has a high intensity and a narrow intensity distribution, making the optical image OI2 clear. Therefore, if there is no defocus in the optical image OI2 or any positional deviation from the reference line fl', it can be seen that the illumination spot 40 coincides with the object point region fl in the direction of the normal n and in the direction S1 (or the amount of positional deviation between the illumination spot 40 and the object point region fl is within a predetermined tolerance).

[0087] The absence of defocus in the optical image OI2 can be determined, for example, when a standard specimen, the coordinates of which and the intensity distribution of scattered light from the defects of which are known, is inspected by the defect inspection apparatus 100, and the difference between the intensity distribution of scattered light measured by the area sensor 6-N from the known defect coordinates and the known intensity distribution is within a predetermined tolerance. The absence of positional deviation between the optical image OI2 and the reference line fl' can be determined, for example, when the distance between the optical image OI2 and the reference line fl' on the inspection image of the standard specimen is within a predetermined tolerance.

[0088] In contrast, if there is no sample surface Z shift but there is illumination Y shift, no defocus occurs in the optical image OI2, and the intensity distribution of the output signal from the area sensor 6-N related to the light from the illumination spot 40 is the same as when there is neither sample surface Z shift nor illumination Y shift. However, the imaging position of the illumination spot 40 on the light-receiving surface of the area sensor 6-N shifts in a direction corresponding to the illumination Y shift, and as shown in the middle of Figure 12, the optical image OI2 shifts from the reference line f1' within the field of view of the area sensor 6-N in a direction corresponding to the illumination Y shift. Therefore, for example, if there is no defocus but a position shift in the optical image OI2 obtained from a known defect in the standard sample described above, it can be seen that the sample surface coincides with the sample-side focal plane fp (or the difference in position between the sample surface and the sample-side focal plane fp in the direction of the normal n, i.e., the sample surface Z shift, is within a predetermined tolerance), and an illumination Y shift occurs by an amount corresponding to the shift of the optical image OI2 from the reference line f1'. The illumination Y shift can be calculated based on the shift of the optical image OI2 from the reference line f1'.

[0089] Furthermore, if there is no illumination Y misalignment but there is a sample plane Z misalignment, defocusing occurs in the optical image OI2, and the output signal of the area sensor 6-N related to the light from the illumination spot 40 changes in intensity distribution compared to when there is no sample plane Z misalignment (the peak value of the signal intensity decreases and the distribution width widens), as shown in the lower part of Figure 12, resulting in a blurred optical image OI2. In addition, the position on the light-receiving surface of the area sensor 6-N onto which the light from the illumination spot 40 is incident also shifts due to the sample plane Z misalignment, causing the optical image OI2 to shift from the reference line f1' in the direction corresponding to the minor axis direction of the illumination spot 40 within the field of view of the area sensor 6-N. Therefore, for example, if there is defocusing and positional shift in the optical image OI2 obtained from a known defect in the standard sample described above, it can be determined that at least a sample plane Z misalignment has occurred (the sample plane is shifted in the normal n direction relative to the sample-side focal plane fp beyond a preset tolerance). The amount of sample plane Z misalignment can be calculated based on the intensity distribution of the optical image OI2.

[0090] FIG. 13 is a diagram illustrating the change in the optical image of the area sensor 6-N due to a Z-axis shift of the sample surface and a Y-axis shift of the illumination light. The schematic diagram in FIG. 13 shows two detection optical systems 4-N (referred to here as detection optical systems 4-L and 4-R) and the corresponding area sensors 6-N (referred to here as area sensors 6-L and 6-R), which are positioned symmetrically with respect to the plane of incidence of the illumination light incident on the sample W. Although the optical axis OA2 of the detection optical systems 4-L and 4-R and the light-receiving surfaces of the area sensors 6-L and 6-R are actually tilted in three-dimensional space with respect to the sample surface, the schematic diagram in FIG. 13 is a two-dimensional conceptual diagram for simplicity. In the schematic diagram in FIG. 13, the sample surface is perpendicular to the plane of the paper, and the illumination spot 40 on the sample surface extends in the depth direction of the paper.

[0091] 13 illustrates an example of Z-shift on the sample surface when the illumination spot 40 deviates from the object point region f1 in the direction of the normal n of the sample surface (upward in the figure). When the sample surface is at the height of the sample-side focal plane fp and the illumination spot 40 coincides with the object point region f1, the image point 40' of the illumination spot 40 is focused on the light-receiving surfaces of the area sensors 6-L and 6-R and is located on a line on the light-receiving surfaces that corresponds to the reference line f1' (FIG. 12).

[0092] On the other hand, when the sample surface is displaced in the Z direction from the sample-side focal plane fp, the illumination spot 40 is displaced from the sample-side focal plane fp, which is conjugate with the light-receiving surfaces of the area sensors 6-L and 6-R. As a result, the image point 40" with the displaced illumination spot 40 as the object point is not focused on the light-receiving surfaces of the area sensors 6-L and 6-R, resulting in defocus. The schematic diagram in the upper part of Figure 13 illustrates an example in which the sample surface is displaced in the +Z direction from the sample-side focal plane fp, and the image point 40" moves toward the back of the light-receiving surface in the direction of light propagation. In the table in the upper part of Figure 13, this defocus is represented by +. If the sample surface is displaced in the -Z direction from the sample-side focal plane fp, the image point 40" moves toward the front of the light-receiving surface in the direction of light propagation.

[0093] Furthermore, if the sample surface is displaced in the Z direction from the sample-side focal plane fp, the relative position of the illumination spot 40 with respect to the detection optical systems 4-L and 4-R changes, changing the angle of the optical path of light from the illumination spot 40, and therefore the incident position of light on the light-receiving surfaces of the area sensors 6-L and 6-R (image position) also changes. The schematic diagram in the upper part of Figure 13 illustrates an example in which the illumination spot 40 is displaced in the +Z direction relative to the object point area fl, and the image positions of the area sensors 6-L and 6-R both move downward in the figure. In the table in the upper part of Figure 13, changes in image position are indicated by "-". If the illumination spot 40 were to be displaced in the -Z direction relative to the object point area fl, the image positions of the area sensors 6-L and 6-R both move upward in the figure.

[0094] The lower part of Fig. 13 illustrates an example of illumination Y deviation when the illumination spot 40 deviates from the object point region fl on the sample-side focal plane fp in the minor axis direction of the illumination spot 40 (to the right in the figure). In this case, since the illumination spot 40 is located on the sample-side focal plane fp that is conjugate with the light-receiving surfaces of the area sensors 6-L and 6-R, an image point 40" with the displaced illumination spot 40 as the object point is located on the light-receiving surfaces of the area sensors 6-L and 6-R, and no defocus occurs. In the table in the lower part of Fig. 13, the defocus in this case is represented as 0.

[0095] However, if the illumination spot 40 deviates from the object point region fl on the sample-side focal plane fp, the relative position of the illumination spot 40 with respect to the detection optical systems 4-L and 4-R changes, and the incident position of light on the light-receiving surfaces of the area sensors 6-L and 6-R (image position) also changes. The schematic diagram in the lower part of Figure 13 illustrates an example in which the illumination spot 40 deviates to the right with respect to the object point region fl, with the image position on the area sensor 6-L moving downward in the figure and the image position on the area sensor 6-R moving upward in the figure. In the table in the lower part of Figure 13, these changes in image position are denoted by - and +, respectively. If the illumination spot 40 deviates to the left with respect to the object point region fl, the image position on the area sensor 6-L will move upward in the figure and the image position on the area sensor 6-R will move downward in the figure.

[0096] - Method for Calculating the Illumination Spot Position - Figure 14 is an explanatory diagram of a method for calculating the illumination spot position in this embodiment. Figure 14 evaluates the output signal of the area sensor 6-N. In this figure, the horizontal axis represents the position of the optical image OI2 corresponding to the position of the illumination spot 40 in the minor axis direction, and the vertical axis represents the image blur of the optical image OI2 corresponding to the position of the illumination spot 40 in the normal direction to the sample surface. The image blur is determined from the signal intensity distribution of the optical image OI2. In this example, the relationship between the Z-axis deviation of the sample surface and the Y-axis deviation of the illumination spot 40 is determined from the output signal of the area sensor 6-N, and the deviations between the object point area fl and the illumination spot 40 (Z-axis deviation of the sample surface and Y-axis deviation of the illumination spot 40) are calculated. The output signal to be evaluated is the output signal of at least one sensor of the area sensor 6-N. When evaluating the position of the illumination spot 40 using the output signals of multiple sensors in the area sensor 6-N, statistics can be calculated for each evaluation result based on the output signal of each sensor (e.g., the average, median, mode, etc.). This point is explained below. The values ​​in parentheses in Fig. 14 represent the illumination Y position shift amount and the sample surface Z shift amount. For example, the notation (+, -) indicates a condition in which the illumination Y shift occurs in the + direction and the sample surface Z shift occurs in the - direction. The -, 0, and + in parentheses in Fig. 14 correspond to the -, 0, and + used in the table in Fig. 13.

[0097] In Figure 14, point F represents data on the optical image OI2 obtained by at least one area sensor (here, the area sensor 6-L shown in Figure 13) without any Z-axis deviation of the sample surface or Y-axis deviation of the illumination. The Z-axis position of the sample surface and the Y-axis position of the illumination spot 40 are adjusted by the stage ST and illumination optical system 3 so that the data on point F can be obtained by the area sensor 6-L. The curve R0 passing through point F represents the relationship between image position and image blur when there is no Y-axis deviation of the illumination. In the case of the area sensor 6-L, within a field of view centered on an arbitrary coordinate Cx (not shown) on the sample surface, if the sample surface is shifted in the +Z direction from the sample-side focal plane fp, a shift in the image position occurs in the +Y direction along the curve R0 (e.g., point G). Conversely, even within a field of view centered on the same coordinate Cx, if the sample surface is shifted in the -Z direction from the sample-side focal plane fp, a shift in the image position occurs in the -Y direction along the curve R0 (e.g., point E). The same coordinates refer to coordinates where the outputs of the encoders of the rotation stage and translation stage of the stage ST are the same.

[0098] Here, for example, suppose that with the illumination spot 40 shifted in the +Y direction (to one side in the short axis direction of the illumination spot 40), the stage ST is driven to change the height of the sample surface four times, and data for points A-D in Figure 14 is obtained for the same coordinate Cx on the sample surface. Then, a curve R1 that passes through these points A-D in the two-dimensional coordinate system of Figure 14 can be identified as the relationship between the sample surface Z shift amount and the illumination Y shift amount when the illumination spot 40 is shifted in the +Y direction. The image position (distance from the reference line fl') where the image blur amount becomes 0 on this curve R1 corresponds to the illumination Y shift amount.

[0099] 14 illustrates an example in which the curve R1 is determined from four points, but considering that the amount of image blur becomes 0 when the sample surface coincides with the sample-side focal plane fp regardless of the deviation of the illumination Y, the curve R1 can be determined from at least two points. However, in practice, considering the influence of variations and errors in the measurement values, it is preferable to determine the curve R1 from three or more points.

[0100] Based on the above, when using the defect inspection device 100 configured as described above to process the signal output by the line sensor 5-N and identify defects, an area sensor 6-N corresponding to the line sensor 5-N is provided, and the amount of deviation between the object point area fl and the illumination spot 40 is calculated based on the optical image OI2 obtained by the area sensor 6-N.The amount of deviation is used as a correction value to adjust the three-dimensional position (height and Y position) of at least one of the illumination spot 40 and the object point area fl, and the illumination spot 40 can be aligned with the object point area fl.

[0101] If the sample surface Z shift amount and illumination Y shift amount are determined using the above method, the alignment of this illumination spot 40 can be achieved by manually adjusting, for example, the illumination optical system 3, the stage ST, the detection optical system 4-N, etc., or in this embodiment, automatic adjustment is also possible.

[0102] With regard to the automatic adjustment of the position of the illumination spot 40, the defect inspection apparatus 100 is configured such that, for example, the signal processing device 6 calculates the Z-axis deviation amount of the sample surface and the Y-axis deviation amount of the illumination, and these deviation amounts are output as correction values ​​to the control device 71, which then controls, for example, at least one of the illumination optical system 3, the stage ST, and the detection optical system 4-N. That is, the signal processing device 6 calculates the deviation amount between the object point area f1 imaged on the line sensor 5-N and the illumination spot 40 based on the position and intensity distribution on the two-dimensional light-receiving surface of the optical image OI2 obtained by the area sensor 6-N, as described in FIG. 14 . Specifically, the signal processing device cooperates with the control device 71 to identify the relationship (curve R1) between the Z-axis deviation amount of the sample surface and the Y-axis deviation amount of the illumination, based on the position and intensity distribution on the two-dimensional light-receiving surface of multiple optical images OI2 obtained by the area sensor 6-N for the same coordinate Cx on the sample surface by changing the height of the stage, as described in FIG. 14 . The control device 71 controls at least one of the stage ST, the illumination optical system 3, and the detection optical system 4-N using the sample surface Z shift amount and the illumination Y shift amount calculated from the curve R1 by the signal processing device 6 as correction values, corrects the position of at least one of the illumination spot 40 or the object point area fl, and three-dimensionally aligns the illumination spot 40 with the linear object point area fl.

[0103] -Effect- If the illumination Y deviation amount is to be determined based on the output signal of the line sensor 5-N of the oblique detection optical system without using the area sensor 6-N of the defect inspection device 100, it is necessary to move the illumination spot 40 in the short axis direction while moving the stage ST in the Z direction, and repeatedly obtain the optical image OI2 for the same coordinate Cx under multiple conditions in which the sample surface height is changed for each position of the illumination spot 40.

[0104] In contrast, by using at least one area sensor 6-N provided in each oblique detection optical system, it is possible to measure the position change in addition to the defocus of the optical image OI2, and as described above, it is possible to measure the illumination Y deviation amount in addition to the sample surface Z deviation amount from multiple pieces of data with different positions in the Z direction on the sample surface without moving the illumination spot 40 in the minor axis direction. This makes it possible to align the illumination spot 40 with the object point area fl, adjust the focus of the optical system of the defect inspection device 100 accurately in a short time, and maintain stable, high inspection sensitivity.

[0105] Furthermore, since the Z deviation amount of the sample surface and the Y deviation amount of the illumination can be measured, the stage ST and the illumination optical system 3 can be automatically adjusted using these values ​​as correction values.

[0106] In addition, since each detection optical system 4-N, which is the oblique optical system to be calibrated, is provided with an area sensor 6-N corresponding thereto, these multiple area sensors 6-N can be used appropriately to calibrate the optical system, which has the advantage of ensuring redundancy.

[0107] Second Embodiment The hardware configuration of the defect inspection apparatus 100 according to this embodiment is the same as that of the first embodiment. The defect inspection apparatus 100 is equipped with a plurality (N units) of oblique detection optical units, each including a line sensor 5, a detection optical system 4, and an area sensor 6. The detection optical systems 4-N, which are oblique detection optical systems included in each of these detection optical units, are arranged symmetrically with respect to the plane of incidence of the illumination light. For example, the optical axes OA2 of the detection optical systems 4-L and 4-R used in the description of FIG. 13 are symmetric with respect to a plane (the plane of incidence of the illumination light without illumination Y misalignment) that is orthogonal to the sample-side focal plane fp in the linear object point areas fl that are imaged on each line sensor. Ignoring the illumination Y deviation and the sample surface Z deviation, if the optical axis OA2 of the detection optical system 4-R is inclined by an angle θ with respect to the normal n of the sample W and by an azimuthal angle φ with respect to the major axis of the illumination spot 40, as shown in FIG. 11 , the optical axis OA2 of the detection optical system 4-L, which is positioned symmetrically to the detection optical system 4-R, is inclined by an angle θ with respect to the normal n and by an angle −φ in the azimuthal direction with respect to the major axis of the illumination spot 40. In this embodiment, the illumination Y deviation is calculated using a pair of detection optical systems symmetrical to each other, such as the detection optical systems 4-L and 4-R. The output signals evaluated in this embodiment are output signals from at least one pair of area sensors whose corresponding detection optical systems are symmetrically positioned. When the position of the illumination spot 40 is evaluated using the output signals of multiple pairs of area sensors, statistics can be calculated for each evaluation result based on the output signals of each pair (e.g., using the average, median, mode, etc.).

[0108] FIG. 15 is an explanatory diagram of a method for calculating the illumination spot position in the second embodiment. The values ​​in parentheses in FIG. 15 represent the illumination Y position shift and sample surface Z shift, as in FIG. 14 . The upper diagram in FIG. 15 represents data obtained, for example, by area sensor 6-L corresponding to detection optical system 4-L, and the lower diagram in FIG. 15 represents data obtained, for example, by area sensor 6-R corresponding to detection optical system 4-R. Points A-G shown in the upper diagram in FIG. 15 are the same as points A-G in FIG. 14 . Points A-G and A'-G' in FIG. 15 represent data on optical image OI2 obtained at the same coordinates, and the conditions associated with points A-G (illumination Y shift and sample surface Z shift) correspond to the conditions associated with points A'-G'. For example, the data at points A and A' represent data measured simultaneously by area sensors 6-L and 6-R.

[0109] As shown in the upper diagram of Figure 15, for example, if the illumination spot 40 is displaced in the +Y direction when there is no Z deviation on the sample surface, the image position on the area sensor 6-L is displaced in the +Y direction relative to the reference line fl'. If the sample surface is displaced in the +Z direction when there is no Z deviation on the sample surface, the image position on the area sensor 6-L is displaced in the +Y direction relative to the reference line fl', as shown by the curve R0. On the other hand, as shown in the lower diagram of Figure 15, if the illumination spot 40 is displaced in the +Y direction when there is no Z deviation on the sample surface, the image position on the area sensor 6-R is displaced in the -Y direction relative to the reference line fl'. If the sample surface is displaced in the +Z direction when there is no Y deviation on the sample surface, the image position on the area sensor 6-R is displaced in the +Y direction relative to the reference line fl', as shown by the curve R0'.

[0110] As can be seen from Figure 15, the difference in the Y direction between the image positions of optical images OI2 obtained at the same coordinates by a pair of area sensors 6-L and 6-R, whose detection optical systems are positioned symmetrically, corresponds to the illumination Y shift, regardless of the sample surface Z shift. For example, by calculating the difference a-a' between the Y position a of point A and the Y position a' of point A' based on the identity of the shapes of curves R1 and R1' in Figure 15, the Y position (illumination Y shift) of the illumination spot 40 on the sample-side focal plane fp can be calculated based on this difference a-a'. The sample surface Z shift can be determined separately from the illumination Y shift, for example, from the intensity distribution. The illumination Y deviation amount thus found is corrected by, for example, correcting the Y position of the illumination spot 40, and then adjusting the height of the stage ST or the focus of at least one of the illumination optical system 3 and the detection optical system 4-N so that the illumination spot 40 is imaged on the area sensors 6-L and 6-R in that state (the image blur amount becomes 0), thereby correcting the sample surface Z deviation amount, and thereby aligning the illumination spot 40 with the object point area fl.

[0111] 15 shows an example in which measurements are taken four times (points A-D / points A'-D') at the same coordinates by changing the height of the sample surface, but according to this embodiment, the illumination Y shift amount and sample surface Z shift amount can be calculated with at least one measurement. However, in practice, when considering the effects of variations and errors in the measurement values, it is desirable to calculate the illumination Y shift amount and sample surface Z shift amount from multiple measurements.

[0112] Based on the above, when using the defect inspection device 100 configured as described above to process the signal output by the line sensor 5-N and identify defects, multiple area sensors 6-N corresponding to the line sensor 5-N are provided, and the detection optical system calculates the amount of deviation between the object point area fl and the illumination spot 40 based on the optical image OI2 obtained by a pair of area sensors with a symmetrical positional relationship, and the three-dimensional position (height and Y position) of at least one of the illumination spot 40 and the object point area fl is adjusted using the deviation amount as a correction value, thereby aligning the illumination spot 40 with the object point area fl.

[0113] If the sample surface Z shift amount and illumination Y shift amount are determined using the above method, the alignment of this illumination spot 40 can be achieved by manually adjusting, for example, the illumination optical system 3, the stage ST, the detection optical system 4-N, etc., or automatic adjustment is also possible as in the first embodiment.

[0114] In this embodiment, the same effects as in the first embodiment can be obtained. Furthermore, the amount of deviation between the illumination spot 40 and the object point area fl can be obtained by performing at least one measurement.

[0115] 16 is a flowchart showing the procedure for operating the defect inspection apparatus 100. The flowchart in FIG. 16 can be configured to be executed by, for example, the signal processing device 6 or the control device 71. By automatically calibrating the position of the illumination spot 40 described in the first or second embodiment at an appropriate timing between defect inspections of the sample W during operation of the defect inspection apparatus 100, the high inspection accuracy of the defect inspection apparatus 100 can be maintained, and highly reliable defect inspection can be continued.

[0116] 16 , after the most recent defect inspection has been completed or while the control device 71 is on standby, in step S101, for example, the control device 71 determines whether a periodic calibration (full) is currently required. This determination is made, for example, based on whether the time elapsed since the previous periodic calibration (full) is equal to or greater than a predetermined time Ts1. Alternatively, the determination in step S101 may be performed by determining whether the amount of change in the operating environment (e.g., humidity or temperature) of the defect inspection device 100 is equal to or greater than a predetermined value Rs1, raising concerns about the impact on the accuracy of the optical system. The humidity and temperature can be determined using the output of a thermometer or hygrometer provided in or communicable with the defect inspection device 100.

[0117] If it is determined that a full periodic calibration is necessary, the control device 71, for example, cooperates with the signal processing device 6 to perform a full periodic calibration in step S102. Here, data acquisition (acquisition of points A-D and A'-D', etc., in FIGS. 14 and 15 ) is performed by inspecting a standard sample with known defect positions, and analysis (calculation of the Z-axis deviation of the sample surface and the Y-axis deviation of the illumination) is performed. If the result shows that the illumination spot 40 deviates from the object point region fl beyond a preset tolerance, correction values ​​related to adjustment of the optical axes and focus of the illumination optical system 3 and the detection optical system 4-N and correction values ​​related to adjustment of the height of the sample surface are updated, and at least one of the focus and angle of the optical axis OA1 of the illumination optical system 3, the focus and angle of the optical axis OA2 of each detection optical system 4-N, and the height of the sample surface is corrected, thereby aligning the object point region fl with the illumination spot 40.

[0118] If it is determined that regular calibration (full) is not necessary, the control device 71 determines, for example, in step S103, whether a situation currently requires regular calibration (simplified). This determination is made, for example, by determining whether the time elapsed since the previous regular calibration (full) was performed is equal to or greater than a predetermined time Ts2 (Ts2<Ts1). Alternatively, the determination in step S103 may be made by determining whether the amount of change in the operating environment (e.g., humidity or temperature) of the defect inspection device 100 is equal to or greater than a predetermined range Rs2 (Rs2<Rs1).

[0119] If it is determined that periodic calibration (simple) needs to be performed, for example, the control device 71 performs periodic calibration (simple) in step S104 in cooperation with the signal processing device 6. Here, periodic calibration (simple) is a configuration operation that is simpler and more frequently performed than periodic calibration (full), and, for example, updates the offset correction value for the sample surface height based on the position and width (blur amount) of the optical image OI2 on the area sensors 6-0 and 6-N and the detection signal intensity and width (blur amount) by the line sensors 5-0 and 5-N, based on the inspection of a standard sample.

[0120] After performing the periodic calibration (simple), if the error between the results of the reinspection of the defects on the standard specimen and the known data exceeds a predetermined tolerance in step S105, the control device 71, for example, proceeds to step S102 and performs periodic calibration (full). If the error between the results of the reinspection of the defects on the standard specimen and the known data is equal to or less than the predetermined tolerance, or if it is determined in step S103 that periodic calibration (simple) is unnecessary, the control device 71, for example, performs pre-inspection measurements and adjustments in step S106, such as adjusting the position of the illumination spot 40 observed by the spot monitor (area sensor 6-0 of the vertical optical system shown in FIG. 8) and the height of the specimen surface measured by the AF sensor (two-dimensional sensor 83 shown in FIG. 10). After performing the pre-inspection measurements and adjustments, the control device 71, for example, inspects the specimen W in step S107. After inspecting the specimen W, the control device 71, for example, returns the procedure to step S101.

[0121] 16 , regular calibration (full) is automatically performed in conjunction with the operation of the defect inspection apparatus 100, thereby making it possible to continue inspection of the sample W while suppressing the increase in the error between the illumination spot 40 and the object point area fl. Furthermore, regular calibration (simple) has the advantage of being simpler and being able to be performed in a shorter time than regular calibration (full), and of being able to maintain a certain level of accuracy, but it does not resolve any positional deviation of the illumination spot 40 in the minor axis direction. In contrast, in the example of FIG. 16 , by frequently performing regular calibration (simple), it is possible to efficiently inspect the sample W while maintaining a certain level of accuracy while suppressing the time required for calibration, and by performing regular calibration (full) infrequently, it is possible to maintain the ideal optical state of the illumination optical system 3 and the detection optical system 4-N.

[0122] (Modifications) The present invention is not limited to the above-described embodiments and may include various modifications. For example, the above-described embodiments have been described in detail to clearly explain the present invention, and are not necessarily limited to those including all of the described configurations. For example, it is possible to replace part of the configuration with another configuration. It is also possible to delete part of the configuration of the embodiment or add another configuration. It is also possible to combine multiple embodiments.

[0123] 3...Illumination optical system, 4-N...Detection optical system, 5-N...Line sensor, 6...Signal processing device, 6-N...Area sensor, 40...Illumination spot (irradiation area), 71...Control device, 100...Defect inspection device, fl...Object point area, fp...Specimen-side focal plane, n...Normal, OA2...Optical axis, OI, OI2...Optical image (linear illumination image), Pi...Image plane, S1...Main scanning direction, S2...Direction perpendicular to the main scanning direction, ST...Stage, W...Specimen

Claims

1. A defect inspection device comprising: a stage that carries a sample and moves; an illumination optical system that irradiates the sample surface with linear illumination light; a line sensor that detects scattered light generated from a linear illumination area on the sample surface irradiated with the illumination light; a detection optical system whose optical axis is inclined with respect to the normal to the sample surface and that guides the scattered light to the line sensor and forms a linear illumination image that is an image of the illumination area on the line sensor; and a signal processing device that processes signals output by the line sensor to identify defects; and an area sensor located at a position whose optical distance from the sample surface is equal to the optical distance between the line sensor and the sample surface, wherein the illumination area is linear with its major axis direction perpendicular to the main scanning direction of the sample, and an image plane formed by the detection optical system that is conjugate to the sample surface is inclined with respect to the optical axis of the detection optical system, a first light-receiving surface of the area sensor is optically two-dimensionally parallel to the image plane, and a second light-receiving surface of the line sensor is inclined with respect to the image plane and intersects with the image plane with the linear illumination image.

2. A defect inspection device according to claim 1, comprising a plurality of detection optical units each including the line sensor, the detection optical system, and the area sensor, wherein the optical axes of the detection optical systems of the first detection optical unit and the second detection optical unit are symmetrical with respect to a plane perpendicular to the focal plane in the linear object point areas imaged on each line sensor.

3. A defect inspection device according to claim 1, characterized in that the signal processing device calculates the amount of deviation between the linear object point area imaged on the line sensor and the illumination area based on the linear illumination image obtained by the area sensor.

4. A defect inspection device as described in claim 3, characterized in that the signal processing device calculates the amount of deviation between the linear object point area imaged on the line sensor and the irradiation area based on the position and intensity distribution on the first light receiving surface of the linear illumination image obtained by the area sensor.

5. A defect inspection device according to claim 4, characterized in that the signal processing device calculates a first amount of deviation in the normal direction of the sample surface and a second amount of deviation in the main scanning direction as the amount of deviation between the object point area and the irradiation area.

6. A defect inspection device according to claim 5, characterized in that the signal processing device determines the relationship between the first amount of deviation and the second amount of deviation based on the positions and intensity distributions on the first light receiving surface of multiple linear illumination images obtained by the area sensor for the same coordinates on the sample surface by changing the height of the stage.

7. A defect inspection device according to claim 5, further comprising a control device for controlling the stage, the illumination optical system, and the detection optical system, wherein the control device controls at least one of the stage, the illumination optical system, and the detection optical system using the first amount of deviation and the second amount of deviation as correction values, and three-dimensionally aligns the positions of the object point area and the irradiation area.

8. A defect inspection device according to claim 2, characterized in that the signal processing device calculates the amount of deviation between the object point area and the irradiation area based on the position of the linear illumination image obtained by the area sensor of the first detection optical unit and the position of the linear illumination image obtained by the area sensor of the second detection optical unit.

9. A defect inspection device according to claim 8, wherein the signal processing device calculates the amount of deviation in the main scanning direction as the amount of deviation between the object point area and the irradiation area.

10. A defect inspection device as described in claim 9, characterized in that the signal processing device calculates the amount of deviation in the main scanning direction based on the difference in position on the first light receiving surface between the linear illumination images obtained at the same coordinates on the sample surface by the area sensors of the first detection optical unit and the second detection optical unit.

11. A defect inspection device according to claim 9, further comprising a control device for controlling the stage, the illumination optical system, and the detection optical system, wherein the control device controls at least one of the stage, the illumination optical system, and the detection optical system using the amount of deviation as a correction value, and three-dimensionally aligns the positions of the object point area and the irradiation area.

12. A defect inspection method for detecting defects by processing a signal output by the line sensor, the method comprising: a stage for placing a sample on the stage and moving it; an illumination optical system for irradiating linear illumination light onto the sample surface; a line sensor for detecting scattered light generated from a linear illumination area on the sample surface irradiated with the illumination light; and a detection optical system whose optical axis is inclined with respect to the normal to the sample surface and which guides the scattered light to the line sensor and forms a linear illumination image, which is an image of the illumination area, on the line sensor, the method comprising: providing an area sensor disposed at a position whose optical distance from the sample surface is equal to the optical distance between the line sensor and the sample surface; making the illumination area linear with its major axis direction perpendicular to the main scanning direction of the sample; tilting an image plane conjugate to the sample surface by the detection optical system with respect to the optical axis of the detection optical system; making a first light receiving surface of the area sensor optically two-dimensionally parallel to the image plane; and tilting a second light receiving surface of the line sensor with respect to the image plane so that the linear illumination image intersects with the image plane; a defect inspection method comprising: calculating a deviation amount between the linear object point area imaged on the line sensor and the illumination area based on the linear illumination image obtained by the area sensor; and three-dimensionally aligning the positions of the object point area and the illumination area using the deviation amount as a correction value.

Images