A defect detection system and a defect detection method

Abstract

The application provides a defect detection system and a defect detection method. The defect detection system comprises a carrier for carrying a to-be-detected object, an illumination channel and a plurality of detection channels located on the same side of the carrier. An incident plane of incident light emitted by a light source is perpendicular to a detection plane, and the detection plane is a plane where optical axes of the plurality of detection channels are located. The plurality of detection channels comprise a first detection channel, a second detection channel and a third detection channel. The first detection channel comprises a first objective lens and a first detector, the second detection channel comprises a second objective lens and a second detector, and the third detection channel comprises a third objective lens and a third detector. A first numerical aperture of the first objective lens is greater than a second numerical aperture of the second objective lens. The defect detection system provided by the application has high detection sensitivity and a wide dynamic range.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of semiconductor, in particular to a defect detection system and a defect detection method. BACKGROUND

[0002] Defect detection is to detect and locate the physical and chemical defects such as particles and scratches on the surface of the object to be measured, which is directly related to the quality and yield of the chip and is an important link in the semiconductor manufacturing process. Defect detection can be dark field defect detection, that is, the surface of the object to be measured is illuminated, and the scattered light is collected to determine whether there is a defect. In addition, a high-precision moving platform can drive the object to be measured to move, thereby completing the defect detection of different regions on the surface of the object to be measured. Any contamination (such as foreign matter or particles generated during the process) or defect (such as scratches, dents, etc.) on the semiconductor chip with a physical size close to or larger than the feature size of the final product is harmful and must be strictly monitored throughout the manufacturing process. Therefore, patternless wafer surface defect detection has important applications in many fields of semiconductors, including: product quality inspection of wafer manufacturers; incoming inspection of chip manufacturers, contamination monitoring of process equipment and process; and equipment contamination condition inspection of semiconductor equipment manufacturers. Different sources of contamination are also different. Generally speaking, particle protrusion defects are mainly caused by the environment in the semiconductor equipment chamber, pit defects are mainly caused by the epitaxial and crystal pulling processes, and scratch defects are caused by errors in the polishing process. In defect detection, obtaining defect type information can help users in different fields to efficiently locate the defect source and effectively reduce the number of repeated detections required. Currently, it is not possible to accurately detect protrusion defects and pit defects. SUMMARY

[0003] Therefore, the purpose of the present application is to provide a defect detection system and a defect detection method with high detection sensitivity and wide dynamic range. The specific scheme is as follows:

[0004] In one aspect, the present application provides a defect detection system, comprising:

[0005] a stage; the stage is used to carry an object to be measured;

[0006] an illumination channel and a plurality of detection channels located on the same side of the stage; the illumination channel comprises a light source, the light source is used to provide a detection spot to a detection region of the object to be measured, the detection channel is used to collect scattered light generated by the detection spot in the detection region, and determine the signal intensity of the scattered light, so as to determine the defect information of the detection region based on the signal intensity;

[0007] the incident plane of the incident light emitted by the light source is perpendicular to the detection plane, and the detection plane is the plane where the optical axes of the plurality of detection channels are located.

[0008] The plurality of detection channels comprises a first detection channel, a second detection channel and a third detection channel; the optical axis of the first detection channel and the normal direction between the surface of the object to be detected have a first deviation angle, the optical axis of the second detection channel and the normal direction have a second deviation angle, the optical axis of the third detection channel and the normal direction have a third deviation angle, the first deviation angle and the second deviation angle are both greater than the third deviation angle, and the first deviation angle and the second deviation angle satisfy the angle difference condition with the third deviation angle.

[0009] The first detection channel comprises a first objective lens and a first detector, the second detection channel comprises a second objective lens and a second detector, and the third detection channel comprises a third objective lens and a third detector, and the first numerical aperture of the first objective lens is greater than the second numerical aperture of the second objective lens.

[0010] In a possible implementation, the detection spot is a line spot, the detection plane is perpendicular to a target plane, and the target plane is a plane in which the optical axis of the illumination channel and the long side of the detection spot are located.

[0011] In a possible implementation, the detector has a photosensitive surface, and the long side direction of the photosensitive surface is perpendicular to the target plane.

[0012] In a possible implementation, the second detection channel further comprises a light attenuation element, and the light attenuation element is configured to perform attenuation processing on the scattered light generated by the detection spot in the detection area.

[0013] In a possible implementation, the first deviation angle is equal to the second deviation angle.

[0014] In a possible implementation, the first deviation angle and the second deviation angle are both greater than or equal to 45°, and the third deviation angle is less than 45°.

[0015] In a possible implementation, the first numerical aperture is greater than or equal to 0.2 and less than or equal to 0.6; and / or, the second numerical aperture is greater than or equal to 0.1 and less than or equal to 0.3.

[0016] In a possible implementation, the illumination channel comprises a polarization adjustment module, and the polarization adjustment module is configured to adjust the polarization state of the light beam emitted by the light source.

[0017] In a possible implementation, the illumination channel comprises a shaping module, and the shaping module is configured to shape the light beam emitted by the light source to obtain the detection spot.

[0018] In another aspect, the application also provides a defect detection method applied to the defect detection system, the defect detection method comprising:

[0019] During the rotation of the object table, different regions of the object under test located on the object table are subjected to defect detection through the illumination channel and the plurality of detection channels;

[0020] During the defect detection of the detection region of the object under test, the light source provides a detection spot to the detection region through the illumination channel, the detection channels collect the scattered light generated by the detection spot in the detection region, and the signal intensity of the scattered light is determined;

[0021] Based on the first signal intensity of the first detection channel, the second signal intensity of the second detection channel, and the third signal intensity of the third detection channel, the defect information of the detection region is determined.

[0022] In a possible implementation, when the defect information includes a defect type, and the defect type includes a protrusion type defect and a depression type defect, the determination of the defect information of the detection region based on the first signal intensity of the first detection channel, the second signal intensity of the second detection channel, and the third signal intensity of the third detection channel comprises:

[0023] When the first signal intensity is greater than a first threshold, and the second signal intensity and the third signal intensity are not greater than the first threshold, it is determined that the detection region has a protrusion type defect; the first threshold is greater than or equal to the noise signal intensity corresponding to the object under test;

[0024] When the third signal intensity is greater than the first threshold, and the first signal intensity and the second signal intensity are not greater than the first threshold, it is determined that the detection region has a depression type defect.

[0025] In a possible implementation, when the defect information includes a defect type, and the defect type includes a protrusion type defect and a depression type defect, the determination of the defect information of the detection region based on the first signal intensity of the first detection channel, the second signal intensity of the second detection channel, and the third signal intensity of the third detection channel comprises:

[0026] When the first signal intensity and the second signal intensity are both greater than a first threshold, and the third signal intensity is not greater than the first threshold, it is determined that the detection region has a protrusion type defect; the first threshold is greater than or equal to the noise signal intensity corresponding to the object under test;

[0027] determining a signal strength ratio of the third signal strength and the first signal strength when the first signal strength and the third signal strength are both greater than the first threshold value and the second signal strength is not greater than the first threshold value; and determining that the detection area has a recessed type of defect if the signal strength ratio is greater than a second threshold value, and otherwise determining that the detection area has a protrusion type of defect.

[0028] In a possible implementation, when the defect information includes a defect type, and the defect type includes a protrusion type of defect and a recessed type of defect, the determining of the defect information of the detection area based on the first signal strength of the first detection channel, the second signal strength of the second detection channel, and the third signal strength of the third detection channel includes:

[0029] determining an effective signal strength based on the first signal strength and the second signal strength when the first signal strength, the second signal strength, and the third signal strength are all greater than a first threshold value; and the first threshold value is greater than or equal to a noise signal strength corresponding to the object to be detected;

[0030] determining a signal strength ratio of the third signal strength and the effective signal strength;

[0031] determining that the detection area has a recessed type of defect if the signal strength ratio is greater than a second threshold value, and otherwise determining that the detection area has a protrusion type of defect.

[0032] In a possible implementation, the determining of the effective signal strength based on the first signal strength and the second signal strength includes:

[0033] obtaining a first saturation intensity corresponding to a saturation power of the first detector, and a second saturation intensity corresponding to a saturation power of the second detector; and the first saturation intensity is less than or equal to the second saturation intensity;

[0034] determining the effective signal strength based on a difference between the first signal strength and the first saturation intensity, and a difference between the second signal strength and the second saturation intensity.

[0035] In a possible implementation, the determining of the effective signal strength based on the difference between the first signal strength and the first saturation intensity, and the difference between the second signal strength and the second saturation intensity includes:

[0036] determining the first signal strength as the effective signal strength when the first signal strength is less than or equal to a product of the first saturation intensity and a third threshold value; and the third threshold value is less than or equal to 1.

[0037] determining the effective signal intensity based on the first signal intensity and the second signal intensity when the first signal intensity is greater than the product of the first saturation intensity and the third threshold value, and is less than the first saturation intensity, and the second signal intensity is less than the second saturation intensity;

[0038] determining the second signal intensity as the effective signal intensity when the first signal intensity is greater than or equal to the first saturation intensity, and the second signal intensity is less than the second saturation intensity.

[0039] In a possible implementation, the determining the effective signal intensity based on the first signal intensity and the second signal intensity comprises:

[0040] the effective signal intensity is represented by one of the following two formulas:

[0041]

[0042] wherein, the first signal intensity, the second signal intensity, and T is the transmittance corresponding to the light attenuation element, T is 1 when no light attenuation element is added in the second detection channel;

[0043]

[0044] wherein, the first signal intensity, the second signal intensity, and T is the transmittance corresponding to the light attenuation element, T is 1 when no light attenuation element is added in the second detection channel, the first numerical aperture, the second numerical aperture.

[0045] In a possible implementation, the determining the effective signal intensity based on the difference between the first signal intensity and the first saturation intensity, and the difference between the second signal intensity and the second saturation intensity comprises:

[0046] adding a light attenuation element in the second detection channel when the second signal intensity is greater than or equal to the second saturation intensity;

[0047] collecting scattered light generated by the detection light spot in the detection area through the second detection channel with the light attenuation element, and determining a fourth signal intensity of the scattered light;

[0048] When the first signal strength is greater than or equal to the first saturation strength and the fourth signal strength is less than the second saturation strength, the effective signal strength is determined based on the fourth signal strength.

[0049] In a possible implementation, the defect information further includes a defect size, and the determining of the defect information of the detection area based on the first signal strength of the first detection channel, the second signal strength of the second detection channel, and the third signal strength of the third detection channel includes:

[0050] When the detection area has the protrusion-type defect, a target signal strength is acquired; the target signal strength is the first signal strength or the second signal strength;

[0051] The target signal strength is matched with a calibration curve to obtain a defect size of a defect in the detection area; a detection channel corresponding to the target signal strength is consistent with a detection channel corresponding to the calibration curve.

[0052] In a possible implementation, the defect information further includes a defect size, and the determining of the defect information of the detection area based on the first signal strength of the first detection channel, the second signal strength of the second detection channel, and the third signal strength of the third detection channel includes:

[0053] When the detection area has the protrusion-type defect, a target signal strength is acquired; the target signal strength is the first signal strength or the second signal strength;

[0054] The target signal strength is matched with a calibration curve to obtain a defect size of a defect in the detection area; a detection channel corresponding to the target signal strength is consistent with a detection channel corresponding to the calibration curve.

[0055] The embodiment of the present application provides a kind of defect detection system and defect detection method, defect detection system includes the carrier for carrying the object to be measured, illumination channel and multiple detection channels located in the same side of carrier.The illumination channel includes light source, light source is used to provide detection spot to the detection area of object to be measured, detection channel is used to collect the scattering light generated in detection area of detection spot, and the signal intensity of scattering light is determined, so as to determine the defect information of detection area based on signal intensity;The incident plane of incident light emitted by light source is perpendicular to detection plane, and detection plane is the plane where the optical axis of multiple detection channels is located;Multiple detection channels include first detection channel, second detection channel and third detection channel;The normal direction between the optical axis of first detection channel and the surface of object to be measured has first deviation angle, the normal direction between the optical axis of second detection channel and the normal direction has second deviation angle, and the normal direction between the optical axis of third detection channel and the normal direction has third deviation angle, first deviation angle and second deviation angle are both greater than third deviation angle, and first deviation angle and second deviation angle meet angle difference condition with third deviation angle;First detection channel includes first objective and first detector, second detection channel includes second objective and second detector, and third detection channel includes third objective and third detector, and the first numerical aperture of first objective is greater than the second numerical aperture of second objective.

[0056] In this way, by setting at least three detection channels, the scattering light of each angle can be collected, the scattering light of different types of defects can be efficiently collected, so that the convex defect and the concave defect can be accurately identified, and the defect detection accuracy is improved.In addition, the first detection channel and the second detection channel have different numerical apertures, the first detection channel with larger numerical aperture has higher detection sensitivity, and can realize detection of small size defects, and the second detection channel with smaller numerical aperture can realize detection of larger size defects, and can widen the dynamic range of defect size detection.In summary, the defect detection system provided by the present application has higher detection sensitivity and wider dynamic range. BRIEF DESCRIPTION OF DRAWINGS

[0057] In order to more clearly illustrate the technical solutions in the embodiments of the present application or the prior art, the drawings needed in the embodiments or prior art description will be briefly introduced below. Obviously, the drawings in the following description are some embodiments of the present application, and those skilled in the art can also obtain other drawings according to these drawings without creative labor.

[0058] Figure 1 A schematic diagram of a defect detection system provided by an embodiment of the present application is shown;

[0059] Figure 2 A schematic diagram of scattering light provided by an embodiment of the present application is shown.

[0060] Figure 3 A perspective view of a defect detection system is shown according to an embodiment of the present application;

[0061] Figure 4 A schematic view of a defect detection system is shown according to another embodiment of the present application;

[0062] Figure 5 A schematic view of a defect detection system is shown according to another embodiment of the present application;

[0063] Figure 6 A side view of a defect detection system is shown according to an embodiment of the present application;

[0064] Figure 7 A signal distribution diagram of a defect is shown according to an embodiment of the present application;

[0065] Figure 8 A response curve diagram of a first detection channel and a second detection channel is shown according to an embodiment of the present application;

[0066] Figure 9 A response curve diagram of a first detection channel and a second detection channel is shown according to another embodiment of the present application;

[0067] Figure 10 A flow diagram of a defect detection method is shown according to an embodiment of the present application;

[0068] Figure 11 A diagram of signal intensity in detection channels in different directions for different types and sizes of defects is shown according to an embodiment of the present application.

[0069] Reference signs

[0070] Stage-101, object under test-102, illumination channel-103, light source-104, detection spot-105, polarization adjustment module-106, shaping module-107, first detection channel-201, second detection channel-202, third detection channel-203, first objective-2011, first detector-2012, second objective-2021, second detector-2022, third objective-2031, third detector-2032, incident plane-301, detection plane-302, target plane-303. DETAILED DESCRIPTION

[0071] In order to make the above objectives, features and advantages of the present application more apparent, the specific embodiments of the present application will be described in detail below with reference to the accompanying drawings.

[0072] In the following description, numerous specific details are set forth in order to provide a thorough understanding of the present application. However, the present application can be practiced without the specific details. In other instances, well-known methods, procedures, components, and circuits have not been described in detail so as not to unnecessarily obscure aspects of the present application.

[0073] As described in the background, at present, it is unable to accurately detect the protrusion type defects and the recess type defects.

[0074] Based on the above technical problems, the embodiments of the present application provide a defect detection system and a defect detection method, by setting at least three detection channels, the scattered light of each angle can be collected, the efficient collection of the scattered light of different types of defects can be realized, so that the protrusion type defects and the recess type defects can be accurately identified, and the defect detection accuracy is improved. In addition, the first detection channel and the second detection channel have different numerical apertures, the first detection channel with a larger numerical aperture has higher detection sensitivity, the detection of small size defects can be realized, the second detection channel with a smaller numerical aperture can realize the detection of larger size defects, the dynamic range of defect size detection can be widened, in general, the defect detection system provided by the present application has higher detection sensitivity and wider dynamic range.

[0075] In order to facilitate understanding, the defect detection system and the defect detection method provided by the embodiments of the present application will be described in detail below in combination with the drawings.

[0076] Reference Figure 1 As shown in the figure, it is a schematic diagram of a defect detection system provided by the embodiments of the present application, which comprises a stage 101, an illumination channel 103 and a plurality of detection channels.

[0077] The stage 101 is used for carrying a to-be-tested object 102, the to-be-tested object 102 is an object which needs to be tested for defects, for example, a wafer, etc., the to-be-tested object 102 can be placed above the stage 101.

[0078] The illumination channel 103 and the plurality of detection channels are located on the same side of the stage 101, so as to facilitate the realization of defect detection, for example, the to-be-tested object 102 is placed above the stage 101, then the illumination channel 103 and the plurality of detection channels are also located above the stage 101.

[0079] The illumination channel 103 can include a light source 104 for providing a detection light spot 105 to a detection area of the object 102, and the specific type of the light source 104 is not limited herein, for example, the light source 104 can be a laser light source, and the shape of the detection light spot 105 is not limited herein, for example, the detection light spot 105 can be a point light spot, a line light spot, an elliptical light spot, and the like, and the spot diameter of the point light spot is in the order of tens to hundreds of microns, and the point light spot can also be used for defect detection. The detection area is an area of the object 102 that needs to be detected for defects, and the detection area can be a partial area or the entire area of the surface of the object 102, and generally, when the detection area is detected for defects, the detection light spot 105 emitted by the light source 104 falls within the detection area.

[0080] In addition, the illumination channel 103 can be vertically arranged (not shown in the figure) or can be obliquely arranged, and the number of the illumination channel 103 can be one or more, and the present application does not limit this. When the illumination channel is multiple, the vertically arranged illumination channel 103 can achieve vertical incidence of light, and the obliquely arranged illumination channel 103 can achieve oblique incidence of light, thereby achieving the cooperative use of the two illumination modes of vertical incidence and oblique incidence, thereby enriching the scattering excitation conditions of the object and improving the full coverage ability of defect detection.

[0081] Specifically, the vertically arranged illumination channel 103 can emit a detection light beam along the normal direction of the surface of the object 102, and can form a symmetrical vertical illumination field on the surface of the object 102, which is suitable for exciting targets such as surface micro-particles, shallow concave defects, and the like, which have obvious positive scattering. The incidence direction has high incidence uniformity, can form regular and strong scattering signals, and is beneficial to signal extraction and imaging stability of subsequent detection channels.

[0082] The obliquely arranged illumination channel 103 can emit a detection light beam along an oblique angle, so that the detection light beam is incident to the surface of the object 102 at a certain oblique angle, and forms an inclined surface scattering with the structural edges or concave-convex interfaces of the object 102, which is helpful to excite asymmetric signals of defects (such as pit edges, protruding particle side walls, and the like) with large surface profile changes or more significant lateral scattering. The incidence direction significantly expands the distribution range of the signal light scattering direction, and improves the response ability of the detection system to different types of defects.

[0083] In actual application, when it is necessary to detect defects on the entire surface area or a partial surface area of the object 102, the detection light spot 105 can be moved or the object 102 can be moved, so that the detection light spot 105 traverses the entire surface area or the partial surface area, and the defect detection is realized.

[0084] The detection channel is used to collect scattered light generated by the detection spot 105 in the detection area and determine the signal intensity of the scattered light so as to determine the defect information of the detection area based on the signal intensity. Specifically, the detection channel may include an objective lens and a detector, with the detector located on the light-emitting side of the objective lens, and the detector being, for example, a camera.

[0085] In other words, the light beam in the illumination channel 103 is incident on the detection area to form a detection spot 105. The scattered light generated by the detection spot 105 is received by the detector through the objective lens. The detector can determine the defect information of the detection area based on the signal intensity of the scattered light. Specifically, the signal intensity of the scattered light is significantly different when there is a defect in the detection area and when there is no defect. Therefore, the defect information can be determined by collecting the scattered light. The defect information can include the size, type, and location of the defect. Defect types can be divided into dent defects and protrusion defects. Dent defects include pits and scratches, while protrusion defects include particles and bosses. The distribution of scattered light generated by different defects is also different.

[0086] Specifically, refer to Figure 2 The diagram shown is a schematic representation of scattered light provided in an embodiment of this application. When the detection spot 105 provided by the illumination channel 103 illuminates the object under test 102, the mechanical device can drive the object under test 102 to rotate. When the detection spot 105 illuminates a defect 108, the light signal interacts with the defect, and the resulting scattered light propagates in the hemispherical space above the object under test 102. Thus, multiple detection channels at different locations above the object under test 102 can collect light signals with different cone angles at different angles in space, so that defect detection can be performed based on the signal intensity determined by the light signal.

[0087] The incident plane 301, where the incident light emitted from the light source 104 is located, is perpendicular to the detection plane 302. The detection plane 302 is the plane containing the optical axes of multiple detection channels. Specifically, the incident plane 301 is the plane containing the optical axis of the incident light and the normal direction of the surface of the object to be tested 102.

[0088] Combination Figure 3 For explanation and reference Figure 3 The diagram shown is a three-dimensional schematic diagram of a defect detection system provided in an embodiment of this application. The detection channel and the illumination channel 103 are shown with their respective optical axes as representatives. The plane where the optical axis of the illumination channel 103 is located is the incident plane 301. All detection channels are coplanar and located within the detection plane 302. The detection plane 302 and the incident plane 301 are perpendicular to each other, thereby realizing the subsequent defect detection process.

[0089] The plurality of detection channels include a first detection channel 201, a second detection channel 202, and a third detection channel 203. The first detection channel 201 has a first deviation angle between an optical axis of the first detection channel 201 and a normal direction of the surface of the object 102 to be detected, the second detection channel 202 has a second deviation angle between an optical axis of the second detection channel 202 and the normal direction, and the third detection channel 203 has a third deviation angle between an optical axis of the third detection channel 203 and the normal direction. The first deviation angle and the second deviation angle are both greater than the third deviation angle, and the first deviation angle and the second deviation angle satisfy an angle difference condition with the third deviation angle.

[0090] That is, each detection channel has a deviation angle with the normal direction of the surface of the object 102 to be detected. The greater the deviation angle, the farther the detection channel is from the normal direction and the closer the detection channel is to the surface of the object 102 to be detected. The first deviation angle of the first detection channel 201 and the second deviation angle of the second detection channel 202 are both greater than the third deviation angle of the third detection channel 203. That is, the first detection channel 201 and the second detection channel 202 are wide (Wide) channels, referred to as W channels, and the third detection channel 203 is a narrow (Narrow) channel, referred to as an N channel.

[0091] In addition, the angle difference condition can be understood as the angle difference between two deviation angles being greater than a preset threshold, that is, the difference between the two deviation angles is large. That is, the angle difference between the first deviation angle and the third deviation angle is large, and the angle difference between the second deviation angle and the third deviation angle is also large. The angle difference between the first deviation angle and the second deviation angle is not limited here. In short, the angle difference between the wide channel and the narrow channel is large, so that the two types of channels can collect light signals at different angles. The wide channel collects scattered light with a large scattering angle, and the narrow channel collects scattered light with a small scattering angle, so as to realize defect detection using more comprehensive scattered light.

[0092] The first detection channel 201 comprises a first objective lens 2011 and a first detector 2012, the second detection channel 202 comprises a second objective lens 2021 and a second detector 2022, and the third detection channel 203 comprises a third objective lens 2031 and a third detector 2032. The first numerical aperture (NA) of the first objective lens 2011 is greater than the second numerical aperture of the second objective lens 2021. That is, the first objective lens 2011 in the first detection channel 201 has a larger numerical aperture than the second objective lens 2021 in the second detection channel 202, so that the resolution of the first objective lens 2011 is higher, and the first detection channel 201 can detect smaller defects, while the second detection channel 202 can detect larger defects. The defect detection system can ensure accurate detection of defects in the size range of tens of nanometers to hundreds of nanometers, and ensure traceability of defect sources during wafer production. For ease of description, the first detection channel 201 with a larger numerical aperture can be referred to as a W1 channel, and the second detection channel 202 with a smaller numerical aperture can be referred to as a W2 channel.

[0093] Next, the determination process of the numerical aperture is described. A standard sample can be prepared according to a preset dynamic range (i.e., a defect size detection range) and a defect type, for example, a dynamic range of 10 nm-100 nm, and the reliability and accuracy of the defect size and type are verified by other means such as electron microscopy measurement. Then, the numerical aperture of each objective lens is determined by simulation combined with the system parameters of the defect detection system, thereby improving the accuracy of the numerical aperture. Subsequently, an intensity-defect size curve can be plotted, so that the machine (i.e., the defect detection system) can directly determine the defect size according to the collected signal intensity, facilitating subsequent electron beam microscopy to find defects. Then, different sizes of defects on the surface of the test object 102 are detected to verify the stability of different machines and the accuracy of the curve.

[0094] In summary, by setting at least three detection channels, the scattered light of each angle can be collected, and the scattered light of different types of defects can be efficiently collected, thereby accurately identifying convex and concave defects and improving the accuracy of defect detection. In addition, the first detection channel 201 and the second detection channel 202 have different numerical apertures, the first detection channel 201 with a larger numerical aperture has higher detection sensitivity and can detect small size defects, and the second detection channel 202 with a smaller numerical aperture can detect larger size defects, which can widen the dynamic range of defect size detection. In summary, the defect detection system provided by the present application has higher detection sensitivity and wider dynamic range.

[0095] When the detection light spot is a line light spot, since multiple objectives simultaneously detect the same to-be-detected region, at least one objective optical axis is not perpendicular to the surface of the to-be-detected object 102, which can cause defocus to easily occur in some positions of the to-be-detected region, and this can affect the defect detection accuracy.

[0096] In a possible implementation, the first deviation angle is equal to the second deviation angle. That is, the first detection channel 201 and the second detection channel 202 can be symmetrically distributed about the normal direction of the surface of the to-be-detected object 102. In this way, the scattering light collected by the two W channels can be more consistent, and the scattering light at the same deviation angle is collected, thereby avoiding the difference between scattering light at different deviation angles and interfering with the defect detection result, and improving the accuracy of defect detection.

[0097] In a possible implementation, the first deviation angle and the second deviation angle are both greater than or equal to 45°, and the third deviation angle is less than 45°. That is, the first detection channel 201 and the second detection channel 202 are as far away from the normal direction of the surface of the to-be-detected object 102 as possible, and the third deviation angle is as close to the normal direction as possible. As an example, the first deviation angle and the second deviation angle can be 60°, and the third deviation angle can be 30°. In this way, the same detection region can be detected at different deviation angles, thereby improving the detection accuracy, and avoiding the case that all detection channels are excessively close to or far away from the normal direction of the surface of the to-be-detected object 102, which causes the detection result to be limited to a certain deviation angle range and the detection accuracy to be poor.

[0098] In a possible implementation, the first numerical aperture is greater than or equal to 0.2 and less than or equal to 0.6; and / or, the second numerical aperture is greater than or equal to 0.1 and less than or equal to 0.3.

[0099] That is, the first numerical aperture can be in the range of 0.2-0.6, and the second numerical aperture can be in the range of 0.1-0.3. However, it should be noted that the first numerical aperture is greater than the second numerical aperture when the two numerical apertures are determined. As an example, the first numerical aperture can be 0.3, and the second numerical aperture can be 0.1. In addition, the third numerical aperture of the third objective 2031 has no fixed size relationship with the first numerical aperture and the second numerical aperture, and the third numerical aperture can be in the range of 0.1-0.4. In summary, by determining the size of the first numerical aperture and the second numerical aperture, the accuracy of the signal strength of the first detection channel 201 and the second detection channel 202 can be further improved, thereby ensuring the accuracy of defect detection.

[0100] In order to solve the defocus problem in line spot illumination, in a possible implementation, the detection spot 105 is a line spot, the detection plane 302 can be perpendicular to the target plane 303, and the target plane 303 is a plane in which the optical axis of the illumination channel 103 and the long side of the detection spot 105 are located.

[0101] The length of the long side of the line spot is different from the length of the short side, the length of the long side of the detection spot 105 can range from 0.5mm to 20mm, and the length of the short side can range from 1um to 100um. The target plane 303 can be a plane in which the optical axis of the illumination channel 103 and the long side of the detection spot 105 are located, which can be referred to as a plane in which the long side of the line spot is located.

[0102] By adjusting the parameters of the elements in the illumination channel 103, the line spot can be rotated, so that the position of the target plane 303 in which the long side of the line spot is located also changes. Referring to Figure 4 As shown in FIG. 3, it is a schematic diagram of another defect detection system provided by an embodiment of the present application, the long side direction of the line spot is rotated to the right, so the target plane 303 is also rotated to the right. Therefore, the target plane 303 and the incident plane 301 are different planes, and they can coincide or not coincide.

[0103] In the present application, the target plane 303 can be perpendicular to the detection plane 302, and since the incident plane 301 is also perpendicular to the detection plane 302, the target plane 303 is the incident plane 301 at this time, or the positions of the two planes coincide. Referring to Figure 5 As shown in FIG. 4, it is a schematic diagram of another defect detection system provided by an embodiment of the present application. At this time, the target plane 303 and the incident plane 301 are the same plane, and the detection plane 302 is perpendicular to the target plane 303.

[0104] In the present application, by setting the target plane 303 perpendicular to the detection plane 302, the long side direction of the detection spot 105 is perpendicular to the detection plane 302, and since the optical axes of the detection channels in the detection plane 302 all intersect at a point, and the point is located on the intersection line of the two planes, the long side direction of the detection spot 105 is perpendicular to the optical axes of the detection channels (i.e. Figure 5The long side of the detection light spot 105 is perpendicular to the three solid lines in the detection plane 302, and the optical axis of the detection channel is perpendicular to the light entrance surface of the objective lens, so the long side direction of the detection light spot 105 is parallel to the light entrance surface of the objective lens, and the distance between the two end points of the detection light spot 105 in the long side direction and the objective lens is basically equal, so the detection light spot 105 is basically located on the object focal plane of the objective lens as a whole, and the defocus phenomenon basically does not occur, the entire detection light spot 105 can be focused in the detector, so that each position of the line region irradiated by the line light spot is in a focused state, ensuring the accuracy of imaging reception at each position of the line region, thereby ensuring the accuracy of defect detection, and improving the spot utilization rate of the detection light spot 105. In short, the detection light spot 105 is parallel to the target plane 303, and the detector is perpendicular to the light-sensitive surface and parallel to the long side of the detection light spot 105, so as to realize focusing.

[0105] In a possible implementation, the short side length of the line light spot can be set as small as possible, for example, in the range of 10 microns to 30 microns.

[0106] Reference Figure 6 As shown in the figure, a side view of a defect detection system provided by an embodiment of the present application is shown, and a first detection channel 201 and a to-be-detected object 102 are shown. When the narrow side length (or line width) of the detection light spot 105 is the length of the AB segment, the distance L from the end point to the object focal plane is the length of A1. When the line width is the length of the CD segment, the distance L from the end point to the object focal plane is the length of C4. The line width CD is longer than the line width AB, and the distance C4 from the end point of the line width CD to the object focal plane is longer than the distance A1 from the end point of the line width AB to the object focal plane, so the virtual focus of the detection light spot 105 with the line width CD is more serious.

[0107] In other words, by setting the short side length of the detection light spot 105 to be small, even if the objective lens or the detector of the detection channel has a small angle of deflection in the narrow side direction, the distance L from the end point on the narrow side of the detection light spot 105 to the object focal plane of the objective lens will be smaller due to the small width of the short side, and even if there is a small angle of deflection in the narrow side direction, the deviation of the distance L caused by the deflection will not be too large, that is, after the deflection, the distance between the end point on the narrow side of the detection light spot 105 and the object focal plane of the objective lens is also small, and does not cause virtual focus. In short, the wider the line width, the more virtual focus, and vice versa, the shorter the line width, the less likely to be virtual focus, thereby improving the focusing accuracy, improving the imaging quality and stability.

[0108] In a possible implementation, the detector has a light-sensitive surface, and the long side direction of the light-sensitive surface is perpendicular to the target plane 303.

[0109] Specifically, the light-sensitive surface of the detector is used to collect the scattered light of the detection spot 105, when the length and width of the light-sensitive surface are different, in order to achieve clear imaging, the long side direction of the light-sensitive surface can be perpendicular to the detection plane 302, so that the long side direction of the light-sensitive surface is parallel to the target plane 303, and the long side direction of the light-sensitive surface is parallel to the long side direction of the detection spot 105, so that the scattered light of the detection spot 105 can be fully collected by the light-sensitive surface, the amount of scattered light entering the detector is improved, clear imaging of the detection spot 105 is achieved, and the detection efficiency is improved.

[0110] Reference Figure 7 As shown in FIG. 3, a signal distribution diagram of a defect provided by an embodiment of the present application is shown, (a) is a signal distribution diagram of a spherical defect, and (b) is a schematic diagram of three detection channels, and the color represents the signal intensity. In (b), the two W channels are located at the top and the bottom, and the N channel is located in the middle.

[0111] Next, taking a photodetector as an example, the signal intensity is described. The photodetector has a saturation power (also known as full well capacity), when the saturation power is reached, the light intensity received by the photodetector is the maximum light intensity, so the signal intensity also reaches the maximum value, and the signal intensity or light intensity will not change with the defect size. In addition, the saturation power of the photodetector and its minimum output noise often increase or decrease at the same time. The smaller the minimum output noise, the smaller the detectable defect size, and the smaller the saturation power.

[0112] Reference Figure 8 As shown in FIG. 3, a signal distribution diagram of a defect provided by an embodiment of the present application is shown, (a) is a signal distribution diagram of a spherical defect, and (b) is a schematic diagram of three detection channels, and the color represents the signal intensity. In (b), the two W channels are located at the top and the bottom, and the N channel is located in the middle. size ), in units of μm, and the ordinate is the relative intensity (Intensity), which is the ratio of the scattered light intensity received by the objective lens to the incident light intensity, so it has no unit. The relative intensity can also be called signal intensity or light intensity. When the defect size is greater than or equal to D1, for the first detection channel 201, the received light intensity reaches or exceeds the saturation power of the photodetector, at this time the output of the photodetector is the maximum value, and it no longer changes with the particle size, so D1 is the maximum defect size that can be defined by channel 1, and the dynamic range is 0-D1. The second detection channel 202 receives less light intensity, for the defect size in the interval of D1 to D2, at this time the signal light intensity does not reach saturation, and the maximum defect size that can be defined can be extended to D2, and the dynamic range is 0-D2.

[0113] It should be noted that the figure depicts the case where the saturation power of the photodetectors used in the two channels is the same. In practice, photodetectors with different saturation powers can also be selected. The maximum defect size that each channel can detect is related to the saturation power of the photodetector in that channel.

[0114] To further improve the dynamic range, in one possible implementation, the second detection channel 202 may also include an optical attenuation element, which is used to attenuate the scattered light generated by the detection spot 105 in the detection area.

[0115] Specifically, when the signal intensity of the scattered light is high, it easily reaches the saturation power of the photodetector. The signal intensity of the detection channel does not change with the defect size, resulting in inaccurate signal strength. Therefore, an optical attenuation element, such as an attenuator, can be added to the defect detection system. After passing through the attenuator, the light intensity decreases, ensuring it remains below the maximum intensity corresponding to the saturation power. This guarantees that the signal intensity accurately reflects the defect size, improving the accuracy of the signal strength readings.

[0116] refer to Figure 9 The diagram shown illustrates the response curves of a first detection channel 201 and a second detection channel 202 provided in an embodiment of this application. An optical attenuation element with an attenuation rate of OD1 is added to the second detection channel 202, enabling the second detection channel 202 to accurately detect defects ranging from D3 to D4 in size range (i.e., dynamic range). Figure 9 D3 in Figure 8 The D1 in the second detection channel 202 expands the dynamic range of the second detection channel 202, ensuring that the first detection channel 201 can efficiently and accurately detect ultra-small defects, while the second detection channel 202 can accurately measure defect sizes below D4.

[0117] In addition, the response curves for defect size and light intensity can be corrected to improve the accuracy of the response curves. The actual response curve of the instrument often deviates from the theoretical value, which is closely related to the operating conditions of the instrument. Therefore, corrections can be made according to the different operating conditions of different instruments. Specifically, standard defect sheets of different sizes are prepared based on the test object 102 with different refractive indices. The instrument's response to defects of different sizes is calibrated using these standard sheets. A correction coefficient is added to each defect size, ultimately obtaining a response curve for different film systems and different defect sizes. This allows the instrument to identify defects of different sizes, facilitating user tracing of defect origins. Within the Rayleigh scattering range, the defect size D... size D can be calculated using the following formula: size= C * x - b * k, where C is a constant for the conversion of light intensity and signal intensity, x is the signal intensity, k is the attenuation rate, b is a coefficient of the scattering signal intensity and particle diameter in the Rayleigh scattering range, and the value is usually between 4 and 6. The coefficient b is related to the collection angle, the defect material, and the polarization state of the light.

[0118] The above analysis is mainly for the case that the incident plane 301 and the detection plane 302 are perpendicular to each other. For different angles between the incident plane 301 and the detection plane 302, the collected spatial angle will change according to the incident angle and the angle between the incident plane 301 and the detection plane 302. For several preset incident angles and angles of the machine itself, the corresponding curves are calculated, and a mathematical model is established to correct the curve data according to the actual machine test results. For different working modes, configuration is performed, and details are not repeated.

[0119] In a possible implementation, the illumination channel 103 includes a polarization adjustment module 106, which is configured to adjust the polarization state of the light beam emitted by the light source 104.

[0120] Specifically, the light beam emitted by the light source 104 passes through the polarization conversion module to realize the adjustment of the polarization state, and can be modulated into different polarization forms such as linear polarization, circular polarization, etc. The linear polarization can be divided into p-polarized light, s-polarized light, and specific angle linear polarization, and the circular polarization can be divided into left-handed circular polarization and right-handed circular polarization. The polarization adjustment module 106 can be a wave plate or a combination of wave plates, or an optical element formed by a metasurface. The light beam modulated by polarization passes through the shaping module 107 and is focused on the object 102 to form a linear light spot.

[0121] In this way, defect detection can be realized by using light beams with various polarization states, and the accuracy of defect detection can be improved. In addition, the polarization adjustment module 106 can also enhance the particle signal. Some particles are more sensitive to light of a specific polarization direction, and rotating the polarization direction can enhance the scattering signal of the particles and improve the detection sensitivity. That is, the signal intensity is improved, and the signal-to-noise ratio (SNR) is improved. The polarization adjustment module 106 can also suppress background noise. The surface roughness of the object 102 and thin film interference can produce background noise, and rotating the polarization direction can suppress the background noise in a specific direction and improve the signal-to-noise ratio. That is, the noise intensity is reduced, and the SNR is improved.

[0122] In a possible implementation, the illumination channel 103 includes a shaping module 107, which is configured to shape the light beam emitted by the light source 104 to obtain the detection light spot 105.

[0123] Specifically, the light emitted by the light source 104 can pass through the polarization conversion module and the beam shaping module 107 in turn. The shaping module 107 can be a collimating flat-top shaping element. The shaping module 107 can also be a free-form surface shaping element, which can provide a single degree of freedom line width compression in the narrow edge direction. The shaping module 107 can be implemented in the form of a diffractive optical element (DOE), a microlens array, a prism, etc. The shaping module 107 can also include a one-dimensional beam expanding element, a collimating flat-top shaping element, a free-form surface shaper, or any other suitable shaping module 107 known in the art. As an example, the shaping module 107 can shape the circular detection light spot 105 into a linear detection light spot 105.

[0124] Reference Figure 10 As shown in the figure, a flowchart of a defect detection method provided by the embodiment of the application is provided. The method is applied to a defect detection system. The method includes the following steps.

[0125] S101, during the rotation of the object table 101, the different regions of the object to be tested 102 located on the object table 101 are detected for defects by the illumination channel 103 and the plurality of detection channels.

[0126] Specifically, the mechanical structure device can be used to transport the object to be tested 102 to the object table 101. The object table 101 can drive the object to be tested 102 to rotate, so that the detection light spot 105 is irradiated to each region on the surface of the object to be tested 102, thereby providing incident light for the detection region. At this time, the detection light spot 105 emitted by the illumination channel 103 is incident on the surface of the object to be tested 102, and the plurality of detection channels collect scattered light for defect detection.

[0127] In actual application, the light source 104 can be a laser. The laser emits a light beam of a specific wavelength. The wavelength range of the light can be 193nm-405nm. The p-polarized light can be formed through the polarization adjustment module 106. The detection light spot 105, such as a line light spot, can be formed through the beam shaping module 107. The length range of the line light spot can be 0.5mm to 20mm, and the width range can be 1μm to 100μm. The illumination channel 103 can be one or more. For example, one illumination channel 103 can provide 0-degree incident light, and another illumination channel 103 can provide 70-degree incident light.

[0128] S102, when detecting the detection region of the object to be tested 102 for defects, the light source 104 provides the detection light spot 105 to the detection region through the illumination channel 103. The scattered light generated by the detection light spot 105 in the detection region is collected by using each detection channel, and the signal strength of the scattered light is determined.

[0129] Specifically, when the detection light spot 105 is incident on the defect, the defect scatters the scattered light under the action of the detection light spot 105, and the scattered light is collected by using three detection channels in the same detection plane 302. Each detection channel can determine the signal strength of the collected scattered light.

[0130] S103, based on the first signal strength of the first detection channel 201, the second signal strength of the second detection channel 202, and the third signal strength of the third detection channel 203, determine the defect information of the detection area.

[0131] Since each signal strength can reflect the condition of the collected scattered light, based on the three signal strengths, the defect information of the detection area can be determined. In this way, by setting at least three detection channels, the scattered light of each angle can be collected, and the scattered light of different types of defects can be efficiently collected, so that the protrusion type defect and the recess type defect can be accurately identified, and the defect detection accuracy is improved. In addition, the first detection channel 201 and the second detection channel 202 have different numerical apertures. The first detection channel 201 with a larger numerical aperture has higher detection sensitivity and can detect small size defects. The second detection channel 202 with a smaller numerical aperture can detect larger size defects, which can widen the dynamic range of defect size detection. In summary, the defect detection system provided by the present application has higher detection sensitivity and wider dynamic range.

[0132] Next, the process of detecting the defect type is described, which mainly includes the following three cases. Case one, the signal strength of only one channel of the three detection channels is greater than the noise signal strength, that is, only one channel collects valid signals. Case two, the signal strength of two channels of the three detection channels is greater than the noise signal strength. Case three, the signal strength of the three detection channels is greater than the noise signal strength.

[0133] When the signal strength of only one channel of the three detection channels is greater than the noise signal strength, in one possible implementation, when the defect information includes the defect type, and the defect type includes the protrusion type defect and the recess type defect, S103 determines the defect information of the detection area based on the first signal strength of the first detection channel 201, the second signal strength of the second detection channel 202, and the third signal strength of the third detection channel 203, which can include S1031-S1032.

[0134] S1031, when the first signal strength is greater than the first threshold, and the second signal strength and the third signal strength are not greater than the first threshold, it is determined that the detection area has a protrusion type defect; the first threshold is greater than or equal to the noise signal strength corresponding to the to-be-measured object 102.

[0135] S1032, when the third signal strength is greater than the first threshold value and the first signal strength and the second signal strength are not greater than the first threshold value, it is determined that the detection area has a recessed type defect.

[0136] Specifically, the defect type can be divided into a recessed type defect and a protruding type defect. The recessed type defect is, for example, a pit, a scratch, etc. The protruding type defect is, for example, a particle, a boss, etc. The distribution of scattered light generated by different defects is also different. The signal strengths collected by the first detector 2012, the second detector 2022 and the third detector 2032 are respectively denoted as a first signal strength I W1 , a second signal strength I W2 and a third signal strength I N . Since three signal strengths are involved in the present application, how to determine the defect type, the defect size and the like based on the three signal strengths will be mainly introduced in the following.

[0137] Specifically, when the signal strength of the scattered light collected by the detection channel from the test object 102, the signal strength scattered by the defect to the light will be collected, and the signal strength scattered by the rough area of the substrate of the test object 102 to the light will also be collected. The scattered light of the rough area of the substrate has little effect on the defect detection, so the useless signal strength can be removed.

[0138] The noise signal strength of the test object 102 is the signal strength generated by the rough part of the substrate in the detection area of the test object 102. This part of the signal strength will not help to divide the defect type. Therefore, the signal strength collected by the detector needs to be greater than the noise signal strength corresponding to the test object 102, so as to ensure that the signal strength collected by the detector can be used to accurately determine the defect type.

[0139] When the signal strength collected by the W channel is greater than the noise signal strength, since the first numerical aperture of the first objective lens 2011 of the W1 channel is greater than the second numerical aperture of the second objective lens 2021 of the W2 channel, the signal collection capability of the W1 channel is greater than that of the W2 channel, and therefore the first signal strength is greater than the second signal strength. Therefore, the signal strength greater than the noise signal strength is the first signal strength, that is, the first signal strength is greater than the first threshold value, and the second signal strength and the third signal strength are not greater than the first threshold value. At this time, the first detection channel 201 is an effective channel, so as to determine the defect size based on the effective channel in the following.

[0140] For this kind of situation, it is indicated that the scattered light of the defect is mainly concentrated near the deviation angle corresponding to the wide channel, so that the defect located in the detection area is a protruding type defect, for example, a particle, etc.

[0141] If only the signal strength collected by the N channel is greater than the noise signal strength in the three detection channels, it indicates that the scattering light of the defect is mainly concentrated near the deviation angle corresponding to the narrow channel, thereby indicating that the defect is a recessed defect, such as a pit, etc. At this time, the third detection channel 203 is an effective channel.

[0142] In this way, when the signal strength collected by only one of the three detection channels is greater than the defect signal strength, the defect type can be accurately determined based on whether the detection channel is the N channel or the W channel, thereby achieving fast and accurate detection of the defect type.

[0143] When the signal strength collected by two of the three detection channels is greater than the noise signal strength, in one possible implementation, when the defect information includes the defect type, and the defect type includes the protruding defect and the recessed defect, S103, the defect information of the detection area is determined based on the first signal strength of the first detection channel 201, the second signal strength of the second detection channel 202, and the third signal strength of the third detection channel 203, which can include S1033-S1034.

[0144] S1033, when the first signal strength and the second signal strength are both greater than the first threshold, and the third signal strength is not greater than the first threshold, it is determined that the detection area has a protruding defect; the first threshold is greater than or equal to the noise signal strength corresponding to the object to be measured 102.

[0145] If the signal strength of the two W channels is greater than the noise signal strength, i.e., the first signal strength and the second signal strength are both greater than the first threshold, and the third signal strength corresponding to the N channel is not greater than the first threshold, it indicates that the scattering light of the defect is mainly concentrated in a larger deviation angle range, thereby indicating that the type of the defect is a protruding defect.

[0146] In addition, if the W1 channel with a larger numerical aperture has reached the saturation power in the two W channels, and the W2 channel with a lower numerical aperture has not reached the saturation power, the second detection channel 202 is an effective channel, and the defect size can be determined based on the second signal strength subsequently. If the W1 channel with a larger numerical aperture has not reached the saturation power, the first detection channel 201 is an effective channel, and the defect size can be determined based on the first signal strength subsequently.

[0147] S1034, when the first signal strength and the third signal strength are both greater than the first threshold, and the second signal strength is not greater than the first threshold, the signal strength ratio of the third signal strength to the first signal strength is determined; if the signal strength ratio is greater than the second threshold, it is determined that the detection area has a recessed defect, otherwise, it is determined that the detection area has a protruding defect.

[0148] If the signal strength collected by the N channel and the W channel are both greater than the noise signal strength, since the collection ability of the W1 channel is higher than the collection ability of the W2 channel, at this time the first signal strength is greater than the noise signal strength, and the second signal strength is not greater than the noise signal strength. Then, the first signal strength and the third signal strength can be compared to determine whether the defect type is a protrusion type or a recess type.

[0149] Specifically, the type determination can be performed by the signal strength ratio K of the third signal strength I N and the first signal strength I w1 . The signal strength ratio K can be expressed as I N / I w1 . The second threshold is a pre-set threshold value, which is used as a basis for dividing the recess type defect and the protrusion type defect. If the signal strength ratio is greater than the second threshold, it means that the signal strength comparison is large, and it can be determined that the detected defect type is a recess type defect. If the signal strength ratio is not greater than the second threshold, it means that the signal strength comparison is small, and the detected defect type is a protrusion type defect.

[0150] In short, if the signal strength ratio K is greater than the second threshold, it is a recess type defect, and if the signal strength ratio K is not greater than the second threshold, it is a protrusion type defect. The second threshold range can vary according to different system layouts, for example, between 1-5.

[0151] In this way, the signal strength ratio determined by the signal strength of the wide and narrow channels can determine the type of defect, thereby realizing accurate differentiation between protrusion type defects and recess type defects, and improving the accuracy of defect type detection. In addition, if it is determined to be a protrusion type defect, the W1 channel can be set as an effective channel, and if it is determined to be a recess type defect, the N channel can be indicated as an effective channel.

[0152] Next, the principle of determining the defect type based on the signal strength ratio of the wide channel and the narrow channel will be described. The wide channel can be the first detection channel 201 or the second detection channel 202.

[0153] Referring to Figure 11 , the defect size change will cause the signal strength collected by each detection channel to change accordingly. Two different sizes of protrusion and recess type defects are described in the vertical collection channel light intensity distribution, where θ (theta) represents the angle between the collection channel optical axis and the wafer normal direction.

[0154] It can be found from the figure that the scattered light is unevenly distributed above the wafer, and the distribution characteristics are related to the defect type. The scattered light of the protrusion type defect is mainly distributed in the large-angle channel, and the scattered light of the pit type defect is mainly distributed in the small-angle channel. Therefore, only when the signal intensity collected by the W1 channel (as described above, the signal collection capability of the W1 channel is greater than that of the W2 channel) or the two W channels is greater than the noise signal intensity, it indicates that the scattered light of the defect is mainly concentrated near the deviation angle corresponding to the wide channel, thereby indicating that the defect located in the detection area is a protrusion type defect; only when the signal intensity collected by the N channel is greater than the noise signal intensity, it indicates that the scattered light of the defect is mainly concentrated near the deviation angle corresponding to the narrow channel, thereby indicating that the defect is a pit type defect, for example, a pit and the like.

[0155] The signals of the wide channel and the narrow channel are both from the scattered light of the same defect, and the signal intensity ratio of the two channels , specifically I N / I w1 or I N / I w2 or or , exhibits different rules for different defect types: the protrusion type defect I N / I W is smaller, and the pit type defect I N / I W is larger. In Figure 11 , the central angle of the W1 channel represents the angle between the optical axis direction of the W1 channel and the normal direction of the surface of the object to be measured 102, the central angle of the W2 channel represents the angle between the optical axis direction of the W2 channel and the normal direction of the surface of the object to be measured 102, and the central angle of the N channel represents the angle between the optical axis direction of the N channel and the normal direction of the surface of the object to be measured 102. Among them, the dashed line corresponds to a defect size of 41 nm, and the solid line corresponds to a defect size of 31 nm. The red line is used to illustrate the signal light intensity of the pit type defect at different angles, and the blue line is used to illustrate the signal light intensity of the protrusion type defect at different angles.

[0156] For the letter marks on Figure 11 , a brief description is given, for example, I W1,大 represents the signal intensity obtained by detecting the large-size protrusion type defect by using the W1 channel, I W1,大 ’ represents the signal intensity obtained by detecting the large-size pit defect by using the W1 channel, I W1,小 represents the signal intensity obtained by detecting the small-size protrusion type defect by using the W1 channel, I W1,小 ’ represents the signal intensity obtained by detecting the small-size pit type defect by using the W1 channel, I N,大 ’ represents the signal intensity detected by using the N channel to detect the large-size pit type defect, and IN,大 represents the signal intensity detected by the Nth channel for large-size protrusion-type defects, I N,小 represents the signal intensity detected by the Nth channel for small-size recess-type defects, I N,小 represents the signal intensity detected by the Nth channel for small-size protrusion-type defects, I W2,大 represents the signal intensity detected by the W2th channel for large-size protrusion-type defects, I W2,大 represents the signal intensity detected by the W2th channel for large-size recess-type defects, I W2,小 represents the signal intensity detected by the W2th channel for small-size protrusion-type defects, I W2,小 represents the signal intensity detected by the W2th channel for small-size recess-type defects, I

[0157] Next, the signal intensity ratios of the large-size protrusion-type defects and the large-size recess-type defects are compared and explained. The signal intensity ratio of the large-size protrusion-type defects is I N / I w1 As an example, when the first detection channel 201 (W1 channel) is an effective W channel, it can be found through calculation that I N,大 / I W1,大 <I N,大 ’ / I W1,大 ’, that is, the signal intensity ratio of the recess-type defects is larger, and the signal intensity ratio of the protrusion-type defects is smaller. Therefore, if the signal intensity ratio is larger, the defect type is more likely to be a recess-type defect, and if the signal intensity ratio is smaller, the defect type is more likely to be a protrusion-type defect.

[0158] When the signal intensities of the three detection channels are all greater than the noise signal intensity, in one possible implementation, when the defect information includes a defect type, and the defect type includes a protrusion-type defect and a recess-type defect, S103 determines the defect information of the detection area based on the first signal intensity of the first detection channel 201, the second signal intensity of the second detection channel 202, and the third signal intensity of the third detection channel 203, which can include S1035-S1037.

[0159] S1035, when the first signal intensity, the second signal intensity, and the third signal intensity are all greater than a first threshold, determining an effective signal intensity based on the first signal intensity and the second signal intensity; the first threshold is greater than or equal to the noise signal intensity corresponding to the object to be detected 102.

[0160] S1036, determining a signal intensity ratio of the third signal intensity and the effective signal intensity.

[0161] S1037, if the signal intensity ratio is greater than a second threshold, determining that the detection area has a recess-type defect, otherwise, determining that the detection area has a protrusion-type defect.

[0162] Specifically, when the signal intensity collected by the three detection channels is all greater than the noise signal intensity, the effective signal intensity can be determined according to the signal intensity corresponding to the two W channels (i.e., the first signal intensity and the second signal intensity). As an example, any one of the first signal intensity and the second signal intensity can be taken as the effective signal intensity, and as another example, the effective signal intensity can also be determined according to whether the signal intensity reaches the maximum signal intensity corresponding to the saturation power. In short, the effective signal intensity can reflect the signal intensity of the scattered light collected by the W channel, which can be used for determining the signal intensity ratio in the subsequent to determine the defect type, and in addition, the effective signal intensity can also be used to determine the defect size.

[0163] Then, the third signal intensity I N The signal intensity ratio of the effective signal intensity The signal intensity ratio K can be expressed as If the signal intensity ratio is greater than the second threshold value, it means that the signal intensity is relatively large, and it can be determined that the detected defect type is a recess type defect. If the signal intensity ratio is not greater than the second threshold value, it means that the signal intensity is relatively small, and the detected defect type is a protrusion type defect.

[0164] In this way, when the signal intensity collected by the three detection channels is all higher than the noise signal intensity, the accurate effective signal intensity can be determined from the two wide channels, and based on the third signal intensity of the narrow channel and the effective signal intensity of the wide channel, the signal intensity ratio can be determined to realize accurate detection of the defect type.

[0165] In order to determine the accurate effective signal intensity from the first signal intensity and the second signal intensity, in a possible implementation, S1035 determines the effective signal intensity based on the first signal intensity and the second signal intensity, which can include S10351-S10352.

[0166] S10351, obtaining a first saturation intensity corresponding to the saturation power of the first detector 2012 and a second saturation intensity corresponding to the saturation power of the second detector 2022; the first saturation intensity is less than or equal to the second saturation intensity.

[0167] S10352, determining the effective signal intensity based on the difference between the first signal intensity and the first saturation intensity, and the difference between the second signal intensity and the second saturation intensity.

[0168] Specifically, the first saturation intensity is a maximum signal intensity corresponding to when the first detector 2012 reaches a saturation power, and the second saturation intensity is a maximum signal intensity corresponding to when the second detector 2022 reaches a saturation power. By setting the first saturation intensity to be no greater than the second saturation intensity, since the objective lens of the first detection channel 201 has a greater numerical aperture than the objective lens of the second detection channel 202, that is, the first signal intensity I W1 is greater than the second signal intensity I W2 , so that the case where the first signal intensity I W1 does not reach the first saturation intensity I1, and the second signal intensity I W2 reaches the second saturation intensity I2, that is, I W1 <I1, and I W2 >I2, can be avoided.

[0169] Since the closer the signal intensity is to the saturation intensity, the closer the detector is to the saturation power at this time, the signal intensity output by the detector is more likely to be inaccurate, so the difference between the first signal intensity and the first saturation intensity can reflect the reliability and accuracy of the first signal intensity, and the difference between the second signal intensity and the second saturation power can reflect the reliability and accuracy of the second signal intensity. Based on this, according to the difference between the two signal intensities and the respective corresponding saturation intensities, a more accurate effective signal intensity can be determined, greatly improving the accuracy of the effective signal intensity.

[0170] In one possible implementation, S10352, determining the effective signal intensity based on the difference between the first signal intensity and the first saturation intensity, and the difference between the second signal intensity and the second saturation intensity, can include S201-S203.

[0171] S201, when the first signal intensity is less than or equal to the product of the first saturation intensity and a third threshold value, determining that the first signal intensity is the effective signal intensity; the third threshold value is less than or equal to 1.

[0172] Specifically, the first signal intensity is represented by I w1 , the first saturation intensity is represented by I1, and the third threshold value is represented by a. If the following condition is met , it indicates that the difference between the first signal intensity and the first saturation intensity is still relatively large, and the first signal intensity is accurate enough, so the first signal intensity can be taken as the effective signal intensity, and the first detection channel 201 is the effective channel.

[0173] In addition, the third threshold a can be not greater than 1, further, 0.8≤a≤1 can be set, a is a fixed constant for each optical system, and the specific value is related to the linear interval distribution of the response curves of the two photodetectors in the system at the detection wavelength. In addition, in the present application, the third threshold a is introduced to participate in the signal strength comparison, so that the signal strength comparison can be compared in the linear range of the detector, and the inaccuracy caused by comparison in the nonlinear range is avoided. The third threshold a is used to calibrate the linear range.

[0174] S202, when the first signal strength is greater than the product of the first saturation intensity and the third threshold, and less than the first saturation intensity, and the second signal strength is less than the second saturation intensity, determining an effective signal strength based on the first signal strength and the second signal strength.

[0175] Specifically, the second signal strength is represented by I w2 The second saturation intensity is represented by I2, if a*I1 w1 <I1, and I w2 <I2, it indicates that the first signal strength does not exceed the first saturation intensity, and the first signal strength is accurate enough. And the second signal strength also does not exceed the second saturation intensity, which indicates that the second signal strength is also relatively accurate, both of the signal strengths can be used, and the effective signal strength can be determined based on the first signal strength and the second signal strength. In addition, the first detection channel 201 and the second detection channel 202 are both effective channels.

[0176] In a possible implementation, when the first detection channel 201 and the second detection channel 202 are both effective channels, that is, in the process of determining the effective signal strength based on the first signal strength and the second signal strength, the effective signal strength is represented by one of the following two formulas:

[0177]

[0178] wherein, is the first signal strength, is the second signal strength, and T is the transmittance corresponding to the light attenuation element, when the light attenuation element is not added in the second detection channel 202, T is 1;

[0179]

[0180] wherein, is the first signal strength, is the second signal strength, and T is the transmittance corresponding to the light attenuation element, when the light attenuation element is not added in the second detection channel 202, T is 1, is the first numerical aperture, is the second numerical aperture.

[0181] In practical applications, which formula is used to determine the effective signal intensity can be selected based on system settings. Both of the above formulas can obtain accurate effective signal intensity.

[0182] S203, when the first signal intensity is greater than or equal to the first saturation intensity and the second signal intensity is less than the second saturation intensity, determining the second signal intensity as the effective signal intensity.

[0183] Specifically, if I w1 >I1and I w2 <I2, it indicates that the first detector 2012 has reached the saturation power, and the first signal intensity is not accurate enough. At this time, the second detection channel 202 has not reached the saturation power, and the second signal intensity can be determined as the effective signal intensity. In addition, the second detection channel 202 can be used as an effective channel.

[0184] In one possible implementation, S10352, based on the difference between the first signal intensity and the first saturation intensity, the difference between the second signal intensity and the second saturation intensity, determining the effective signal intensity can include S204-S206.

[0185] S204, when the second signal intensity is greater than or equal to the second saturation intensity, increasing the light attenuation element in the second detection channel 202.

[0186] S205, collecting the scattered light generated by the detection light spot 105 in the detection area through the second detection channel 202 with the light attenuation element, and determining the fourth signal intensity of the scattered light.

[0187] S206, when the first signal intensity is greater than or equal to the first saturation intensity and the fourth signal intensity is less than the second saturation intensity, determining the effective signal intensity based on the fourth signal intensity.

[0188] That is, when I w2 >I2, it indicates that the second detection channel 202 has also reached the saturation power of the second detector 2022, and the second signal intensity is also not accurate enough. Therefore, it indicates that there is no effective channel at this time, which is a large size defect. If it is necessary to accurately define the size of the large defect, it is necessary to increase the light attenuation element in the second detection channel 202. The second detection channel 202 with the added light attenuation element is used to continue detecting the defect until the fourth signal intensity collected by the second detection channel 202 is not greater than the second saturation intensity, indicating that the second detector 2022 does not reach the saturation power after the light attenuation element is added, and the fourth signal intensity is accurate enough. The fourth signal intensity can be used as the effective signal intensity.

[0189] In this way, by adding the light attenuation element in the second detection channel 202, the dynamic range of the system can be expanded, and accurate detection of large-size defects can be realized.

[0190] Next, how to determine the defect size is described. For the protrusion type defect, the effective channel is specifically the W channel, and for the recess type defect, the effective channel is specifically the N channel. The defect size can be determined based on the signal strength of the effective channel.

[0191] For the protrusion type defect, in one possible implementation, the defect information further includes the defect size, S103, and the defect information of the detection area is determined based on the first signal strength of the first detection channel 201, the second signal strength of the second detection channel 202, and the third signal strength of the third detection channel 203, which can include S301-S302.

[0192] S301, when the detection area has a protrusion type defect, the target signal strength is obtained; the target signal strength is the first signal strength or the second signal strength.

[0193] S302, matching the target signal strength with the calibration curve to obtain the defect size of the defect in the detection area; the detection channel corresponding to the target signal strength is consistent with the detection channel corresponding to the calibration curve.

[0194] Specifically, when the detection area has a protrusion type defect, the first signal strength or the second signal strength can be obtained as the target signal strength. If the first signal strength does not reach the first saturation strength corresponding to the saturation power of the first detector 2012, the first signal strength can be used as the target signal strength, otherwise, the second signal strength is used as the target signal strength.

[0195] The calibration curve is used to reflect the corresponding relationship between the signal strength and the defect size. Different detection channels can correspond to different calibration curves. If the target signal strength is the first signal strength, the calibration curve corresponding to the first detection channel 201 is compared with the first signal strength to determine the defect size. If the target signal strength is the second signal strength, the calibration curve corresponding to the second detection channel 202 is compared with the second signal strength to determine the defect size. In this way, the calibration curve strictly corresponds to the detection channel, and the defect size of the protrusion type defect can be accurately determined, greatly improving the accuracy of size detection.

[0196] For the recess type defect, in one possible implementation, the defect information further includes the defect size, S103, and the defect information of the detection area is determined based on the first signal strength of the first detection channel 201, the second signal strength of the second detection channel 202, and the third signal strength of the third detection channel 203, which can include S401-S402.

[0197] S401, when the detection area has a recess type defect, acquiring a third signal intensity.

[0198] S402, matching the third signal intensity with a calibration curve corresponding to the third detection channel 203 to obtain the defect size of the defect in the detection area.

[0199] Specifically, for the recess type defect, only one N channel is included in the defect detection system, so the calibration curve corresponding to the third detection channel 203 (i.e. the N channel) can be determined, and the third signal intensity collected by the third detector 2032 is compared with the calibration curve, so that the defect size corresponding to the third signal intensity can be determined, and the size of the recess type defect can be accurately detected.

[0200] Each of the embodiments in the specification is described in a progressive manner, and the same or similar parts between the embodiments can be referred to each other. Each embodiment focuses on the difference from other embodiments. In particular, the device embodiments are described more simply because they are basically similar to the method embodiments, and the relevant parts can be referred to the part of the method embodiments.

[0201] The above is only the preferred embodiment of the present application. Although the preferred embodiment of the present application has been disclosed as above, it is not intended to limit the present application. Any skilled person in the art can make many possible changes and modifications to the technical solution of the present application, or modify it into equivalent embodiments with equivalent changes without departing from the scope of the technical solution of the present application. Therefore, any simple modification, equivalent change and modification made to the above embodiments according to the technical essence of the present application, without departing from the content of the technical solution of the present application, still falls within the scope of protection of the technical solution of the present application.

Claims

1. A defect detection system, characterized by, The application relates to a device for detecting defects of a sample, comprising: a sample stage; the sample stage is used for carrying a sample to be detected; an illumination channel and a plurality of detection channels are located on the same side of the sample stage; the illumination channel comprises a light source, which is used for providing a detection spot to a detection area of the sample to be detected, and the detection channels are used for collecting scattered light generated by the detection spot in the detection area and determining signal intensity of the scattered light, so as to determine defect information of the detection area based on the signal intensity; an incident plane where incident light emitted by the light source is located is perpendicular to a detection plane where optical axes of the plurality of detection channels are located; the plurality of detection channels comprise a first detection channel, a second detection channel and a third detection channel; an optical axis of the first detection channel has a first deviation angle with a normal direction between a surface of the sample to be detected, an optical axis of the second detection channel has a second deviation angle with the normal direction, and an optical axis of the third detection channel has a third deviation angle with the normal direction; the first deviation angle and the second deviation angle are both greater than the third deviation angle, and the first deviation angle and the second deviation angle satisfy an angle difference condition with the third deviation angle; the first detection channel comprises a first objective lens and a first detector, the second detection channel comprises a second objective lens and a second detector, and the third detection channel comprises a third objective lens and a third detector; a first numerical aperture of the first objective lens is greater than a second numerical aperture of the second objective lens; different areas of the sample to be detected located on the sample stage are subjected to defect detection through the illumination channel and the plurality of detection channels; when the detection area of the sample to be detected is subjected to defect detection, the light source provides a detection spot to the detection area through the illumination channel, and each detection channel is used for collecting scattered light generated by the detection spot in the detection area and determining signal intensity of the scattered light; defect information of the detection area is determined based on first signal intensity of the first detection channel, second signal intensity of the second detection channel and third signal intensity of the third detection channel; when the defect information comprises a defect type, the defect type comprises a protrusion type defect and a depression type defect, and the defect information of the detection area is determined based on the first signal intensity, the second signal intensity and the third signal intensity, comprising: when the first signal intensity, the second signal intensity and the third signal intensity are all greater than a first threshold value, effective signal intensity is determined based on the first signal intensity and the second signal intensity; the first threshold value is greater than or equal to a noise signal intensity corresponding to the sample to be detected; a signal intensity ratio of the third signal intensity and the effective signal intensity is determined; if the signal intensity ratio is greater than a second threshold value, it is determined that the detection area has the depression type defect, otherwise, it is determined that the detection area has the protrusion type defect.

2. The defect detection system of claim 1, wherein The detection light spot is a line light spot, and the detection plane is perpendicular to a target plane, which is a plane where an optical axis of the illumination channel and a long side of the detection light spot are located.

3. The defect detection system of claim 2, wherein The detector has a photosensitive surface, and a long side direction of the photosensitive surface is perpendicular to the target plane.

4. The defect detection system of claim 1, wherein The second detection channel further comprises a light attenuation element configured to attenuate scattered light generated by the detection light spot in the detection area.

5. The defect detection system of claim 1, wherein The first deviation angle is equal to the second deviation angle.

6. The defect detection system of claim 1, wherein The first deviation angle and the second deviation angle are both greater than or equal to 45°, and the third deviation angle is less than 45°.

7. The defect detection system of claim 1, wherein The first numerical aperture is greater than or equal to 0.2 and less than or equal to 0.6, and / or the second numerical aperture is greater than or equal to 0.1 and less than or equal to 0.

3.

8. The defect detection system of claim 1, wherein, The illumination channel comprises a polarization adjustment module configured to adjust a polarization state of the light beam emitted by the light source.

9. The defect detection system of claim 1, wherein, The illumination channel comprises a shaping module configured to shape the light beam emitted by the light source to obtain the detection light spot.

10. The defect detection system of claim 1, wherein, When the defect information comprises a defect type, and the defect type comprises a protrusion type defect and a depression type defect, the determining of the defect information of the detection area based on the first signal intensity of the first detection channel, the second signal intensity of the second detection channel, and the third signal intensity of the third detection channel comprises: When the first signal intensity is greater than a first threshold, and the second signal intensity and the third signal intensity are both not greater than the first threshold, it is determined that the detection area has a protrusion type defect; the first threshold is greater than or equal to a noise signal intensity corresponding to the object to be measured. When the third signal intensity is greater than the first threshold, and the first signal intensity and the second signal intensity are both not greater than the first threshold, it is determined that the detection area has a depression type defect.

11. The defect detection system of claim 1, wherein, When the defect information comprises a defect type, and the defect type comprises a protrusion type defect and a depression type defect, the determining of the defect information of the detection area based on the first signal intensity of the first detection channel, the second signal intensity of the second detection channel, and the third signal intensity of the third detection channel comprises: When the first signal intensity and the second signal intensity are both greater than a first threshold, and the third signal intensity is not greater than the first threshold, it is determined that the detection area has a protrusion type defect; the first threshold is greater than or equal to a noise signal intensity corresponding to the object to be measured. When the first signal intensity and the third signal intensity are both greater than the first threshold, and the second signal intensity is not greater than the first threshold, a signal intensity ratio of the third signal intensity to the first signal intensity is determined; if the signal intensity ratio is greater than a second threshold, it is determined that the detection area has a depression type defect, otherwise, it is determined that the detection area has a protrusion type defect.

12. The defect detection system of claim 1, wherein, The determining of the effective signal intensity based on the first signal intensity and the second signal intensity comprises: acquire a first saturation intensity corresponding to a saturation power of the first detector and a second saturation intensity corresponding to a saturation power of the second detector; the first saturation intensity is less than or equal to the second saturation intensity; determine an effective signal intensity based on a difference between the first signal intensity and the first saturation intensity, and a difference between the second signal intensity and the second saturation intensity.

13. The defect detection system of claim 12, wherein, The determination of the effective signal intensity based on the difference between the first signal intensity and the first saturation intensity, and the difference between the second signal intensity and the second saturation intensity, comprises: when the first signal intensity is less than or equal to a product of the first saturation intensity and a third threshold value, determine the first signal intensity as the effective signal intensity; the third threshold value is less than or equal to 1; when the first signal intensity is greater than a product of the first saturation intensity and the third threshold value, and less than the first saturation intensity, and the second signal intensity is less than the second saturation intensity, determine the effective signal intensity based on the first signal intensity and the second signal intensity; when the first signal intensity is greater than or equal to the first saturation intensity, and the second signal intensity is less than the second saturation intensity, determine the second signal intensity as the effective signal intensity.

14. The defect detection system of claim 13, wherein, The determination of the effective signal intensity based on the first signal intensity and the second signal intensity, comprises: said effective signal strength is represented by one of the following two equations: ； wherein, is the first signal intensity, is the second signal intensity, T is the transmittance corresponding to the light attenuation element, and when no light attenuation element is added in the second detection channel, T is 1. ； wherein, is the first signal intensity, is the second signal intensity, T is the transmittance corresponding to the light attenuation element, when no light attenuation element is added in the second detection channel, T is 1, is the first numerical aperture, is the second numerical aperture.

15. The defect detection system of claim 12, wherein, The determination of the effective signal intensity based on the difference between the first signal intensity and the first saturation intensity, and the difference between the second signal intensity and the second saturation intensity, comprises: when the second signal intensity is greater than or equal to the second saturation intensity, increase a light attenuation element in the second detection channel; collect scattered light generated by the detection light spot in the detection area through the second detection channel with the light attenuation element, and determine a fourth signal intensity of the scattered light; when the first signal intensity is greater than or equal to the first saturation intensity, and the fourth signal intensity is less than the second saturation intensity, determine the effective signal intensity based on the fourth signal intensity.

16. The defect detection system of any of claims 10-11, wherein, The defect information further comprises a defect size, and the determination of the defect information of the detection area based on the first signal intensity of the first detection channel, the second signal intensity of the second detection channel and the third signal intensity of the third detection channel, comprises: when the detection area has the protrusion type defect, acquire a target signal intensity; the target signal intensity is the first signal intensity or the second signal intensity; match the target signal intensity with a calibration curve to obtain a defect size of the defect in the detection area; the detection channel corresponding to the target signal intensity is consistent with the detection channel corresponding to the calibration curve.

17. The defect detection system of any of claims 10-11, wherein, The defect information further comprises a defect size, and the determination of the defect information of the detection area based on the first signal intensity of the first detection channel, the second signal intensity of the second detection channel and the third signal intensity of the third detection channel, comprises: when the detection area has the protrusion type defect, acquire a target signal intensity; the target signal intensity is the first signal intensity or the second signal intensity; match the target signal intensity with a calibration curve to obtain a defect size of the defect in the detection area; the detection channel corresponding to the target signal intensity is consistent with the detection channel corresponding to the calibration curve. Matching the third signal strength with a calibration curve corresponding to the third detection channel to obtain a defect size of the defect in the detection area.

Images