Method and system for detecting nanodefects

Abstract

The application discloses a nano-defect detection method and a detection system. The nano-defect detection method comprises the following steps: generating a longitudinal spin-dipole light field, the longitudinal spin-dipole light field comprising two spin light fields with opposite spin directions, in the propagation direction of a test light beam, the longitudinal spin-dipole light field gradually evolves from a spin light field with one spin direction into a spin light field with the opposite spin direction, and the longitudinal spin-dipole light field is a uniform bright spot; obtaining a scattered light field of the longitudinal spin-dipole light field by a product to be detected; separating two spin light field components with different spin directions in the scattered light field; determining the proportion of the difference between the two spin light field components with different spin directions and the total intensity of the two spin light field components with different spin directions; and determining the longitudinal position of the nano-defect according to the proportion. In this way, the detection resolution of the longitudinal position of the nano-defect can break through the limitation of the diffraction limit of traditional optical measurement, and high-precision measurement of the longitudinal position of the nano-defect can be realized.

Description

Technical Field

[0001] This application relates to the field of nanodefect detection technology, and in particular to a method and system for detecting nanodefects. Background Technology

[0002] In semiconductor manufacturing and nanodevice fabrication, nanoscale defects (such as particle deposition, surface depressions, interface voids, and microcracks caused by material stress) significantly affect the electrical performance and optical response of devices, and are important factors determining chip yield and reliability.

[0003] Existing defect detection methods mainly include scanning electron microscopy (SEM), transmission electron microscopy (TEM), and atomic force microscopy (AFM). While these methods offer high resolution, they suffer from drawbacks such as slow scanning speed, limited testing range, and stringent requirements for sample environments, making it difficult to meet the online inspection needs of production lines. On the other hand, detection methods based on optical imaging or interferometry offer the advantages of speed and non-contact operation, but their spatial resolution is often limited by the diffraction limit, enabling high-precision positioning only in the transverse plane, with very limited ability to detect defects in the longitudinal (propagation direction).

[0004] In nanofabrication and device integration, the longitudinal information of defects is equally crucial. For example, in multilayered photonic chips or three-dimensional integrated circuits, the positional accuracy of interlayer defects directly affects alignment and connection quality. However, existing optical inspection methods lack effective means for ultra-high precision measurement of defect longitudinal positions, becoming a key bottleneck restricting the development of three-dimensional nanofabrication technology. Summary of the Invention

[0005] This application provides a method and system for detecting nanodefects to solve at least some of the problems in related technologies.

[0006] This application provides a method for detecting nanodefects, used to determine the longitudinal position of nanodefects in a product under test, wherein the longitudinal position is the position along the propagation direction of a test beam, and the method includes:

[0007] A longitudinal spin dipole field is generated, which includes two spin fields with opposite rotation directions. In the propagation direction of the test beam, the longitudinal spin dipole field gradually evolves from a spin field with one rotation direction to a spin field with the opposite rotation direction, and the longitudinal spin dipole field is a uniform bright spot.

[0008] Obtain the scattered light field of the product under test in response to the longitudinal spin dipole light field;

[0009] Separate the two spin light field components with different rotation directions in the scattered light field;

[0010] Determine the ratio of the difference between the two spin optical field components with different rotation directions to the total intensity of the two spin optical fields with different rotation directions;

[0011] The longitudinal position of the nanodefect is determined based on the specific gravity.

[0012] Optionally, generating the longitudinal spin dipole optical field includes:

[0013] The test beam is expanded and polarized.

[0014] The test beam, after beam expansion and polarization modulation, is irradiated onto a diffractive optical element, the modulation factor of which is:

[0015] ,in, The wavenumber of the test beam. To test the numerical aperture of the objective lens, The refractive index of the medium, The distance between the focal point of the oppositely rotating spin optical field and the center position of the oppositely rotating spin optical field; the diffractive optical element applies the modulation factor to the spin optical field of one spin direction and applies the conjugate of the modulation factor to the spin optical field of the other spin direction;

[0016] The output of the diffractive optical element is mapped onto the incident surface of the detection objective, and the detection objective focuses to generate the longitudinal spin dipole light field.

[0017] Optionally, obtaining the scattered light field of the product under test in response to the longitudinal spin dipole light field includes:

[0018] The scattered light field is collected through the incident surface of the detection objective lens;

[0019] The scattered light field and the test beam are separated by a beam splitter cube.

[0020] Optionally, separating the two spin components with different rotation directions in the scattered light field includes:

[0021] The scattered light field is focused by a focusing lens and the two spin light field components with different rotation directions are separated by a polarization grating.

[0022] Adjust the positions of the focusing lens and the polarization grating so that the two spin light field components with different rotation directions simultaneously illuminate the imaging plane of the CCD camera.

[0023] Optionally, determining the ratio of the difference between the two spin optical field components with different rotation directions to the total intensity of the two spin optical fields with different rotation directions includes:

[0024] The difference between the two spin light field components with different rotation directions and the total intensity of the two spin light fields with different rotation directions are determined based on the output of the CCD camera.

[0025] The specific gravity is determined based on the difference between the two spin light field components with different rotation directions and the total intensity of the two spin light fields with different rotation directions.

[0026] Optionally, determining the longitudinal position of the nanodefect based on the specific gravity includes:

[0027] Using a mapping table that establishes the correspondence between the calibrated longitudinal position of a nanodefect and its calibrated specific gravity, the longitudinal position of the nanodefect is determined based on the specific gravity.

[0028] Optionally, the mapping table is obtained through a mapping table calibration method, which includes:

[0029] After generating the longitudinal spin dipole field, the calibration product with nanodefects is moved along the propagation direction of the test beam, and the longitudinal calibration positions of multiple nanodefects are recorded.

[0030] Obtain the calibration scattered light field of the calibration product on the longitudinal spin dipole light field;

[0031] Separate the two spin light field components with different rotation directions in the calibrated scattered light field;

[0032] Determine multiple calibration weights at multiple calibration longitudinal positions.

[0033] Optionally, before generating the longitudinal spin dipole light field, the method further includes:

[0034] The range of the lateral position of the nanodefect is obtained, wherein the lateral position is a position perpendicular to the propagation direction of the test beam;

[0035] The product under test is moved in a direction perpendicular to the propagation direction of the test beam so that the nanodefect is located at the center of the test beam.

[0036] This application also provides a detection system for nano-defects, the detection system comprising:

[0037] A light field generation component is used to generate a longitudinal spin dipole light field, which includes two spin light fields with opposite rotation directions. In the propagation direction of the test beam, the longitudinal spin dipole light field gradually evolves from a spin light field with one rotation direction to a spin light field with the opposite rotation direction, and the longitudinal spin dipole light field is a uniform bright spot.

[0038] A scattered light field acquisition component is located on the side of the light field generation component away from the product under test, and is used to acquire the scattered light field of the product under test in response to the longitudinal spin dipole light field.

[0039] A separation component, located on the light-emitting side of the scattered light field acquisition component, is used to separate two spin light field components with different rotation directions in the scattered light field;

[0040] The processor, connected to the separation component, is used to determine the ratio of the difference between the two spin optical field components with different rotation directions to the total intensity of the two spin optical fields with different rotation directions, and to determine the longitudinal position of the nanodefect based on the ratio, wherein the longitudinal position is the position along the propagation direction of the test beam.

[0041] Optionally, the light field generating component includes:

[0042] A beam generator for generating the test beam;

[0043] A beam expander, located on the light-emitting side of the beam generator, is used to expand the test beam.

[0044] A polarizer, located on the light-emitting side of the beam expander, is used to adjust the polarization of the expanded test beam.

[0045] A diffractive optical element is disposed on the light-emitting side of the polarizer, and the tuning factor of the diffractive optical element is:

[0046] ,in, The wavenumber of the test beam. To test the numerical aperture of the objective lens, The refractive index of the medium, The distance between the focal point of the oppositely rotating spin optical field and the center position of the oppositely rotating spin optical field; the diffractive optical element applies the modulation factor to the spin optical field of one spin direction and applies the conjugate of the modulation factor to the spin optical field of the other spin direction;

[0047] A mapping component, located between the diffractive optical element and the detection objective, is used to map the output of the diffractive optical element onto the detection objective.

[0048] The detection objective lens, located on the light-emitting side of the mapping component, is used to focus and generate the longitudinal spin dipole light field.

[0049] Optionally, the scattered light field acquisition component includes:

[0050] A beam splitter cube is located between the mapping component and the detection objective, the detection objective being used to collect the scattered light field, and the beam splitter cube being used to separate the scattered light field from the test beam.

[0051] Optionally, the separation component includes:

[0052] A focusing lens is disposed on the light-emitting side of the beam-splitting cube to focus the scattered light field;

[0053] A polarization grating is disposed on the light-emitting side of the focusing lens to separate the focused scattered light field into two spin light field components with different rotation directions.

[0054] A CCD camera is used to simultaneously image the two spin light field components with different rotation directions.

[0055] Optionally, the processor is used to obtain the specific gravity and determine the longitudinal position of the nanodefect according to a mapping table of the correspondence between the calibrated longitudinal position of the nanodefect and the calibrated specific gravity.

[0056] The method and system for detecting nanodefects provided in this application generate a longitudinal spin dipole optical field. Along the propagation direction of the test beam, the longitudinal spin dipole optical field gradually evolves from a spin field with one rotation direction to a spin field with the opposite rotation direction. The method acquires the scattered light field from the product under test (PUT) in response to the longitudinal spin dipole optical field. It separates the two spin field components with different rotation directions within the scattered light field; determines the ratio of the difference between the two spin field components to the total intensity of the two spin field components; and determines the longitudinal position of the nanodefect based on the ratio. Thus, by acquiring the scattered light field from the product under test in response to the longitudinal spin dipole optical field and analyzing the two spin field components with different rotation directions, the longitudinal position of the nanodefect can be determined. The detection resolution of the longitudinal position of the nanodefect can overcome the diffraction limit limitation of traditional optical measurements, achieving high-precision measurement of the longitudinal position of nanodefects. This method ensures high resolution while also being fast, non-destructive, and easily integrated, meeting the future development needs of the nano-manufacturing and semiconductor industries. Attached Figure Description

[0057] The accompanying drawings, which are incorporated in and form part of this specification, illustrate embodiments consistent with this application and, together with the description, serve to explain the principles of this application.

[0058] Figure 1 This is a schematic flowchart illustrating a method for detecting nanodefects according to an embodiment of this application;

[0059] Figure 2 This is a schematic diagram illustrating the generation principle of a longitudinal spin dipole optical field according to an embodiment of this application;

[0060] Figure 3 This is a schematic diagram of the light field distribution of a longitudinal spin dipole light field according to an embodiment of this application;

[0061] Figure 4 for Figure 1 A schematic flowchart of one embodiment of step 11 of the method 10 for detecting nano-defects shown;

[0062] Figure 5 This is a schematic diagram of the phase structure of the control factor of a diffractive optical element according to an embodiment of this application;

[0063] Figure 6 This is a schematic diagram showing the focusing position separation of two oppositely rotating spin optical fields in a longitudinal spin dipole optical field according to an embodiment of this application;

[0064] Figure 7 This is a schematic diagram of the structure of a nanodefect detection system according to another embodiment of this application;

[0065] Figure 8 This is a flowchart illustrating a mapping table calibration method 30 according to an embodiment of this application;

[0066] Figure 9 This is a schematic diagram of the calibration result of the mapping table calibration method 30 shown in one embodiment. Detailed Implementation

[0067] This application provides a method 10 and a detection system for detecting nano-defects. The method 10 and detection system for detecting nano-defects of this application will be described in detail below with reference to the accompanying drawings. Unless otherwise specified, the features in the following embodiments and implementations can be combined with each other.

[0068] Please refer to Figure 1 , Figure 1 This is a schematic flowchart of a nanodefect detection method 10 according to an embodiment of this application. The nanodefect detection method 10 is used to determine the longitudinal position of the nanodefect in the product under test, where the longitudinal position is the position along the propagation direction of the test beam. The method includes steps 11 to 15.

[0069] Step 11: Generate a longitudinal spin dipole field. The longitudinal spin dipole field includes two spin fields with opposite rotation directions. In the propagation direction of the test beam, the longitudinal spin dipole field gradually evolves from a spin field with one rotation direction to a spin field with the opposite rotation direction, and the longitudinal spin dipole field is a uniform bright spot. In the embodiments of this application, the two spin fields with opposite rotation directions are a left-handed spin field and a right-handed spin field, respectively.

[0070] Step 12: Obtain the scattered light field of the product under test in response to the longitudinal spin dipole light field;

[0071] Step 13: Separate the two spin light field components with different rotation directions in the scattered light field;

[0072] Step 14: Determine the ratio of the difference between the two spin optical field components with different rotation directions to the total intensity of the two spin optical fields with different rotation directions;

[0073] Step 15: Determine the longitudinal position of the nanodefects based on their specific gravity.

[0074] Please refer to Figure 2 and Figure 3 , Figure 2 This is a schematic diagram illustrating the generation principle of a longitudinal spin dipole optical field according to an embodiment of this application; Figure 3 This is a schematic diagram of the optical field distribution of a longitudinal spin dipole optical field according to an embodiment of this application. Figure 2 To illustrate the principle, the diagram shows that the longitudinal spin dipole optical field is generated by focusing through a lens. The circles and arrows in the diagram represent the spin structure of the optical field. It can be seen that the longitudinal spin dipole optical field includes two spin optical fields with opposite directions of rotation. The end closer to the lens is one direction of rotation, and the end farther from the lens is the other direction of rotation. There is a smooth transition between the two directions of rotation. Figure 3 This is a schematic diagram of the optical field distribution of a longitudinal spin dipole. The simulated spin dipole optical field distribution is shown in the xz plane. Brightness and darkness represent light intensity, and gray levels represent changes in spin structure. Here, R represents right-handed spin, and L represents left-handed spin. z is the vertical coordinate, and z=0 represents the focal plane position. It can be seen in the figure that the L field is not located at the focal plane position. Figure 2 Instead of being located at the center of the z-axis, the light field is shifted a certain distance in the positive z-direction. Correspondingly, the R-field, representing the right-handed spin field, is shifted a certain distance in the negative z-direction. The light field intensity remains constant between the two, while the spin structure changes significantly, resulting in the produced longitudinal spin dipole light field.

[0075] The longitudinal spin dipole optical field possesses both a gradually changing rotation direction and a uniform bright spot characteristic. The gradually changing rotation direction ensures that the optical field smoothly evolves from one rotation direction (e.g., left-handed) to the opposite rotation direction (e.g., right-handed) along the beam propagation direction, forming a unique correspondence between rotation direction and longitudinal position. The uniform bright spot characteristic avoids distortion of the scattered signal caused by uneven optical field intensity, ensuring the comparability of scattering signals from defects at different longitudinal positions. These two characteristics of the longitudinal spin dipole optical field are equivalent to assigning a rotation direction identifier to the longitudinal position, allowing the longitudinal position of the defect to be deduced from the rotation component of the scattered light.

[0076] By separating the two spin components of the scattered light and calculating the ratio of the difference between the two spin light field components with different spin directions to the total intensity of the two spin light fields with different spin directions, the essence is to extract the deviation signal of the spin gradient. The presence of nanodefects will change the spin evolution law of the local light field, resulting in an imbalance of the intensity of the two spin components. The normalized index of the ratio can effectively eliminate the background noise interference such as ambient light and sample surface reflection, highlighting the longitudinal position characteristics corresponding to the defect. Compared with directly using the intensity difference, the ratio has stronger anti-interference ability.

[0077] Thus, the nanodefect detection method 10 provided in this application can determine the longitudinal position of the nanodefect by acquiring the scattered light field of the product under test in response to the longitudinal spin dipole light field and analyzing the two spin light fields with different spin directions in the scattered light field. The detection resolution of the longitudinal position of the nanodefect can break through the diffraction limit of traditional optical measurement, realizing high-precision measurement of the longitudinal position of the nanodefect. It not only ensures resolution but also has the characteristics of being fast, non-destructive, and easy to integrate, meeting the development needs of future nano-manufacturing and semiconductor industries.

[0078] Figure 4 for Figure 1 The flowchart illustrates one embodiment of step 11 of the nanodefect detection method 10, as shown. Figure 4 As shown, step 11, which generates the longitudinal spin dipole optical field, includes steps 111 to 113.

[0079] Step 111: Expand and polarize the test beam;

[0080] Step 112: Illuminate the expanded and polarized test beam onto the diffractive optical element. The modulation factor of the diffractive optical element is:

[0081] ,in, To test the wavenumber of the beam, To test the numerical aperture of the objective lens, The refractive index of the medium, The distance between the focal point of the spin optical field with opposite spin direction and the center position of the spin optical field with opposite spin direction; the conjugate of the control factor applied to the spin optical field with one spin direction and the control factor applied to the spin optical field with the other spin direction by the diffractive optical element.

[0082] Step 113: The output of the diffractive optical element is mapped onto the incident surface of the detection objective, and the detection objective is focused to generate a longitudinal spin dipole light field.

[0083] The purpose of expanding the test beam is to widen it from a narrow beam to a wide beam, so that the size of the beam cross-section matches the size of the mask of the diffractive optical element. This avoids the beam being too thin, which would cause uneven local light intensity during subsequent adjustment of the diffractive optical element. It ensures that the adjustment effect of the diffractive optical element on the beam can cover the entire beam cross-section, and the resulting longitudinal spin dipole light field can form a uniform bright spot, avoiding interference with the acquisition of defect scattering signals due to differences in the brightness of the spot.

[0084] The purpose of polarizing the expanded test beam is to unify the beam polarization state to a state that can be precisely controlled by the diffractive optical element, such as linear polarization. This eliminates the interference of chaotic polarization direction in natural or randomly polarized light on spin control, provides standardized initial conditions for the subsequent targeted application of spin control factors to the diffractive optical element, and ensures the stable generation of two spin optical fields with opposite spin directions.

[0085] The function of a diffractive optical element is to apply a conjugate phase difference to two oppositely rotating spin optical fields through a preset control factor. Among the control factors, The wavenumber of the test beam is related to its wavelength. The wavelength of the test beam is 532 nm, and k ≈ 1.1810 × 10⁷ rad / m. To measure the numerical aperture of the objective lens, a value of 0.65 was used in this example. The refractive index of the medium, The value is 1. NA and n are inherent system parameters, and a is the distance between the focal point and the center position of the two rotational optical fields. By adjusting a, the longitudinal separation distance of the two rotational optical fields can be precisely controlled, so that they form a gradual interval of two rotations along the beam propagation direction, rather than overlapping or discretely distributed.

[0086] In the embodiments of this application, A numerical aperture of 0.65 is considered relatively high. In optical focusing principles, the minimum focused spot size of an objective lens is directly proportional to the wavelength and inversely proportional to the numerical aperture. In this embodiment, the larger numerical aperture reduces the size of the minimum focused spot, allowing it to completely cover the area affected by the nanodefects. Furthermore, the larger numerical aperture makes the longitudinal separation distance between the two oppositely rotating spin optical fields more compact, shortening the length of the gradient interval from left-handed to right-handed dominance, ultimately achieving higher longitudinal positioning resolution.

[0087] Please also refer to Figure 5 , Figure 5This is a schematic diagram of the phase structure of the control factor of a diffractive optical element according to one embodiment of this application. The schematic diagram of the phase structure serves as a bridge between abstract complex formulas and concrete optical field control. Through the visualization design of a dual-region symmetrical linear phase gradient, the differentiated modulation process of two rotational optical fields is intuitively presented. At the same time, system parameters such as (k, a, NA, n) are correlated, revealing the physical root of the longitudinal spin dipole optical field's rotational gradient and uniform spot characteristics. It also provides an intuitive reference for the fabrication, system debugging, and multi-scenario adaptation of diffractive optical elements, and is a key technology carrier for achieving high-precision longitudinal detection of nano-defects in this application.

[0088] In this embodiment, applying a control factor to the spin optical field of one spin direction and a conjugate control factor to the spin optical field of the other spin direction ensures that the phase changes of the two spin optical fields are mirror-symmetric, thereby forming a smooth and gradual spin transition during longitudinal propagation. This makes the spin direction uniquely correspond to the longitudinal position. For example, a certain longitudinal position corresponds to left-handed light intensity dominance, and another position corresponds to right-handed light intensity dominance. This provides a principle basis for subsequently inferring the longitudinal position of the defect by the spin component of the scattered light.

[0089] In step 113, by mapping the output of the diffractive optical element to the incident surface of the detection objective, it can be ensured that the detection objective can fully receive and focus the modulation effect of the diffractive optical element, avoid the loss of part of the modulation optical field due to beam deflection, and ensure that the gradient range of the longitudinal spin dipole optical field fully covers the longitudinal range where the nanodefects of the product under test may exist.

[0090] By focusing the detection objective lens to generate a longitudinal spin dipole light field, the wide beam of light after being modulated by the diffractive optical element can be compressed into a nanoscale focused spot. This not only meets the requirements of spatial resolution of the light field for nanoscale defect detection and can be applied to nanoscale defects, but also enhances the interaction between the light field and the defect, improves the intensity of the scattered light signal, and makes the longitudinal gradient characteristics of the two spin light fields more significant after focusing (the gradient range is more concentrated, which is convenient for accurately locating the longitudinal position of the defect).

[0091] Please refer to Figure 6 , Figure 6 This diagram illustrates the focusing position separation of two oppositely rotating spin fields within a longitudinal spin dipole optical field, as shown in one embodiment of this application. The generated longitudinal spin dipole optical field is characterized by, at the wavelength or subwavelength scale, as the light field propagates, gradually evolving from a single-rotational spin field to a spin field with the opposite rotational direction. Furthermore, during this spin change, the light beam maintains its bright spot structure. Figure 6 As shown, the focusing positions of the opposite spin structures were separated, and it can be seen that the opposite spins are focused at different positions, with different peak values, which proves the generation of the longitudinal spin dipole light field.

[0092] Meanwhile, steps 111-113 have a high degree of technical controllability. The light field parameters can be flexibly adjusted according to the defect detection requirements of different products under test (such as semiconductor silicon wafers, photonic chips, and nanofilms), while ensuring the consistency of light field generation. This solves the problem that traditional light field parameters are fixed and cannot be adapted to multiple scenarios.

[0093] By adjusting the beam expansion factor in step 111, the spot size of the final light field can be changed (e.g., expanding the beam to a larger spot for a large area of ​​the product under test to cover more detection areas; or reducing the spot size for a small local area to improve local resolution). By adjusting the 'a' value (the center distance between the two spin fields) in step 112, the length of the spin gradient interval of the longitudinal spin dipole field can be changed (e.g., increasing 'a' when detecting deep defects to make the gradient interval cover a deeper longitudinal range; or decreasing 'a' when detecting shallow defects to make the gradient interval more concentrated on the surface). By changing the detection objective with different NA values ​​in step 113, the focusing depth of the light field can be adjusted (high NA objectives focus more deeply and are suitable for thick samples; low NA objectives focus shallower and are suitable for thin samples).

[0094] Please also refer to Figure 7 , Figure 7 This is a schematic diagram of the structure of a nanodefect detection system 2 according to another embodiment of this application. Figure 7 In the illustrated embodiment, the nanodefect detection system 2 includes:

[0095] The light field generation component 21 is used to generate a longitudinal spin dipole light field, which includes two spin light fields with opposite rotation directions. In the propagation direction of the test beam, the longitudinal spin dipole light field gradually evolves from a spin light field with one rotation direction to a spin light field with the opposite rotation direction, and the longitudinal spin dipole light field is a uniform bright spot.

[0096] The scattered light field acquisition component 22 is located on the side of the light field generation component 21 away from the product under test, and is used to acquire the scattered light field of the product under test to the longitudinal spin dipole light field.

[0097] The separation component 23 is located on the light-emitting side of the scattered light field acquisition component 22 and is used to separate two spin light field components with different rotation directions in the scattered light field.

[0098] The processor 24, connected to the separation component 23, is used to determine the difference between two spin optical field components with different rotation directions and the ratio of the total intensity of the two spin optical fields with different rotation directions, and to determine the longitudinal position of the nanodefect based on the ratio, wherein the longitudinal position is the position along the propagation direction of the test beam.

[0099] exist Figure 7 In the illustrated embodiment, the light field generation component 21 includes:

[0100] Beam generator 211, used to generate a test beam;

[0101] Beam expander 212 is located on the light output side of beam generator 211 and is used to expand the test beam.

[0102] Polarizer 213 is located on the light-emitting side of beam expander 212 and is used to adjust the polarization of the expanded test beam.

[0103] Diffractive optical element 214 is disposed on the light-emitting side of polarizer 213, and the control factor of diffractive optical element 214 is:

[0104] ,in, To test the wavenumber of the beam, To determine the numerical aperture of objective lens 216, The refractive index of the medium, The distance between the focal point of the spin optical field with opposite spin direction and the center position of the spin optical field with opposite spin direction; the diffractive optical element 214 applies a control factor to the spin optical field with one spin direction and applies a conjugate of the control factor to the spin optical field with the other spin direction.

[0105] The mapping component 215 is located between the diffractive optical element 214 and the detection objective lens 216, and is used to map the output of the diffractive optical element 214 to the detection objective lens 216.

[0106] The detection objective lens 216, located on the light-emitting side of the mapping component 215, is used to focus and generate a longitudinal spin dipole light field.

[0107] Please refer to Figure 7 ,like Figure 7 As shown, the scattered light field acquisition component 22 includes:

[0108] A beam splitter cube 221 is located between the mapping assembly 215 and the detection objective 216. The detection objective 216 is used to collect the scattered light field, and the beam splitter cube 221 is used to separate the scattered light field and the test beam.

[0109] In this embodiment of the application, obtaining the scattered light field of the product under test in response to the longitudinal spin dipole light field includes:

[0110] The scattered light field is collected by detecting the incident surface of objective lens 216;

[0111] The scattered light field and the test beam are separated by the beam splitter cube 221.

[0112] The scattering of the longitudinal spin dipole light field by nanodefects is a small-angle scattering (the defect size is only on the nanometer scale, and the scattered light is mostly concentrated near the beam propagation direction). The high NA characteristic of the detection objective 216 can cover this small-angle scattering range, preventing the scattered light from escaping due to excessive angle, and significantly improving the collection efficiency of the scattered light. The incident surface and the light field generation path (diffractive optical element 214 → mapping component 215 → detection objective 216) are designed to be coaxial, ensuring that the propagation path of the scattered light from the defect point of action to the incident surface of the objective is not deviated, avoiding signal loss due to the deviation of the light collection angle.

[0113] The inspection objective 216 is both a focusing component for generating the longitudinal spin dipole light field and a light-collecting component for collecting the scattered light field, ensuring that the collection range of the scattered light completely coincides with the action range of the light field. After being focused by the inspection objective 216, the longitudinal spin dipole light field acts on the nano-defects of the product under test. The scattered light generated by the defect will return to the incident surface of the inspection objective 216 along the original optical path. There is no need to add an additional light-collecting lens, avoiding the misalignment of the action area of ​​the light field and the light-collecting area caused by the optical path offset of multiple components. This ensures that the collected scattered light accurately corresponds to the interaction between the defect and the target light field, and eliminates the interference of scattered light from irrelevant areas.

[0114] The beam-splitter cube 221 employs a polarization-independent or polarization-matched beam-splitting principle, ensuring that the rotational characteristics and intensity distribution of the scattered light remain unchanged during the separation process. Compared to other separation methods (such as filters), the beam-splitter cube 221 has no absorption loss, preventing distortion of scattered light intensity due to energy loss and guaranteeing the accuracy of subsequent calculations of rotational component differences and total intensity proportions.

[0115] exist Figure 7 In the illustrated embodiment, the separation component 23 includes:

[0116] A focusing lens 231 is disposed on the light-emitting side of the beam-splitting cube 221 and is used to focus the scattered light field;

[0117] A polarization grating 232 is disposed on the light-emitting side of the focusing lens 231 and is used to separate the focused scattered light field into two spin light field components with different rotation directions.

[0118] CCD camera 233 is used to simultaneously image two spin light field components with different rotation directions.

[0119] In this embodiment of the application, step 13, separating the two spin light field components with different rotation directions in the scattered light field, includes:

[0120] The scattered light field is focused by the focusing lens 231 and the two spin light field components with different rotation directions are separated by the polarization grating 232.

[0121] Adjust the positions of the focusing lens 231 and the polarization grating 232 so that two spin light field components with different rotation directions simultaneously illuminate the imaging plane of the CCD camera 233.

[0122] Scattered light propagates divergently from the point of effect of the defect (primarily scattering at small angles, but still exhibiting spatial diffusion). The focusing lens 231, with a preset focal length (precisely matched with the light outlet of the beam splitter cube 221 and the spacing of the polarization grating 232), compresses the divergent scattered light into a quasi-parallel beam or a focused spot, concentrating the scattered light energy within the effective area of ​​the subsequent polarization grating 232. Without focusing, the scattered light will only reach the polarization grating 232 due to spatial diffusion, with the remaining energy escaping. After focusing by the focusing lens 231, the energy utilization rate can be increased to over 80%, significantly enhancing the signal strength after rotational separation and preventing the effective information from being overwhelmed by CCD imaging noise due to a weak signal.

[0123] In this embodiment, the polarization grating 232 is based on the principle of geometric phase modulation to physically separate the left-handed and right-handed spin light fields in the scattered light, thus overcoming the limitation of traditional polarizers that can only distinguish linear polarization but cannot separate circular polarization.

[0124] By adjusting the position of the focusing lens 231 (along the optical axis), the size of the focused spot and the imaging position can be controlled. If the focusing lens 231 is too close, the focused spot will be too small, causing the imaging areas of the two rotational components to overlap and become indistinguishable; if the position is too far, the spot will be too large and may exceed the imaging plane of the CCD camera 233, resulting in signal loss.

[0125] After adjustment, it is necessary to ensure that the diffraction spot of the left-handed component and the diffraction spot of the right-handed component are located on the left and right sides of the imaging plane of the CCD camera 233, respectively, and both are within the effective imaging range, so as to provide spatial conditions for independently measuring the intensity of the two directions of rotation.

[0126] In this embodiment of the application, determining the ratio of the difference between two spin optical field components with different rotation directions to the total intensity of the two spin optical fields with different rotation directions includes:

[0127] The difference between two spin light field components with different rotation directions and the total intensity of the two spin light fields with different rotation directions are determined based on the output of the CCD camera 233; the specific gravity is determined based on the difference between two spin light field components with different rotation directions and the total intensity of the two spin light fields with different rotation directions.

[0128] The difference between two spin light field components with different rotation directions is S3 in the Stokes parameter of the spin dipole light field, and the total intensity of the two spin light fields with different rotation directions is S0 in the Stokes parameter of the spin dipole light field. Determining S3 and S0 in the Stokes parameter of the spin dipole light field through the output of the CCD camera 233 is prior art, and will not be elaborated further in this application.

[0129] In this embodiment of the application, determining the longitudinal position of the nanodefect based on its specific gravity includes:

[0130] Using a mapping table that corresponds to the calibrated longitudinal position of nanodefects and their calibrated specific gravity, the longitudinal position of nanodefects is determined based on their specific gravity.

[0131] In this embodiment of the application, the processor 24 is used to obtain the specific gravity and determine the longitudinal position of the nanodefect according to the mapping table of the correspondence between the calibrated longitudinal position of the nanodefect and the calibrated specific gravity.

[0132] In the nanodefect detection method 10 of this application, the design of determining the longitudinal position based on the specific gravity using a mapping table between the calibrated longitudinal position and the calibrated specific gravity of the nanodefect is a key step in connecting the quantitative feature (specific gravity) and the physical position (longitudinal coordinate). By establishing a precise correlation through pre-calibration and using normalized features to infer the position, the pain point of no clear mapping between longitudinal position and signal in traditional optical detection is solved, while ensuring detection accuracy and efficiency.

[0133] The mapping table is not a simple list of data, but a quantitative carrier of physical laws based on the gradual change in the rotation direction of the longitudinal spin dipole light field. Its core logic stems from the fact that the longitudinal spin dipole light field exhibits a continuous gradual change from "left-handed to right-handed" along the propagation direction (longitudinal). When the defect interacts with the light field at different longitudinal positions, the proportion of the rotation component of the scattered light will change monotonically with the position (e.g., from close to 1 → 0 → close to 1). This monotonic correlation makes the proportion uniquely mapped to the longitudinal position, that is, one proportion value corresponds to only one longitudinal position, and vice versa.

[0134] Please refer to Figure 8 , Figure 8 This is a flowchart illustrating a mapping table calibration method 30 according to an embodiment of this application. Figure 8 As shown, the mapping table is obtained through mapping table calibration method 30, which includes steps 31 to 34.

[0135] Step 31: After generating the longitudinal spin dipole optical field, move the calibration product with nanodefects along the propagation direction of the test beam and record the longitudinal calibration positions of multiple nanodefects.

[0136] Step 32: Obtain the calibration scattered light field of the calibration product for the longitudinal spin dipole light field;

[0137] Step 33: Separate the two spin light field components with different rotation directions in the calibrated scattered light field;

[0138] Step 34: Determine multiple calibration weights at multiple calibration longitudinal positions.

[0139] Please also refer to Figure 7 ,exist Figure 7 In the detection system 2, a stage 217 is included. During calibration, a calibration product with nano-defects is placed on the stage 217; during measurement, the product to be measured is placed on the stage 217. The calibration product and the product to be measured can be moved by moving the stage 217, and their positions can be recorded.

[0140] In this patent application, the mapping table calibration is not an independent operation, but rather a complete replication of the detection process of the defect to be tested. The corresponding specific gravity is determined by the known longitudinal position, ultimately forming a unique mapping relationship between the calibrated longitudinal position and the calibrated specific gravity. The detection system 2 used in the calibration process and the detection system 2 used in the testing process are the same system, calibrated and tested under the same longitudinal spin dipole light field, ensuring consistency between calibration and testing and improving detection accuracy.

[0141] Furthermore, in the detection system 2, optical components (such as the processing error of diffractive optical element 214 and the aberration of detection objective lens 216) and environmental factors (such as the change in the refractive index of the medium caused by temperature) will introduce systematic errors. If the specific gravity-position relationship is directly calculated through theoretical models, the error may be relatively large.

[0142] The mapping table calibration is a calibration in a real-world scenario. System errors will be reflected simultaneously in the calibration process and the detection process (such as the error of the diffractive optical element 214 causing the separation distance of the two rotational optical fields to shift, which has been incorporated into the position-gravity correlation during calibration). This is equivalent to canceling the system error through calibration, which can reduce the system error of the detection system 2 and improve the accuracy of the longitudinal position measurement of nanodefects.

[0143] Please refer to Figure 9 , Figure 9 This is a schematic diagram of the calibration results of the mapping table calibration method 30 shown in one embodiment. By moving the calibration product with nanodefects along the propagation direction of the test beam and recording the longitudinal positions of multiple nanodefects, a specific gravity correspondence curve is obtained. R 2 The correlation coefficient represents the linear fit, proving the accuracy of the fit. In the figure, 'e' is the labeled error of the linear fit slope, representing the measurement accuracy, and 'Step' represents the step size of each nanometer displacement stage movement.

[0144] In this embodiment of the application, before generating the longitudinal spin dipole light field, the method further includes:

[0145] The range of the lateral position of the nanodefect is obtained, where the lateral position is the position perpendicular to the propagation direction of the test beam.

[0146] In some specific embodiments, obtaining the range of the lateral position of the nanodefect includes: turning on the laser and placing the product under test or the calibration product on the stage 217, shrinking the size of the incident light beam so that the focal field forms a relatively uniform illumination field, thereby using the imaging function of the CCD camera 233 to obtain the approximate position of the nanodefect.

[0147] The test product is moved perpendicular to the propagation direction of the test beam to position the nanodefect at the center of the beam. Specifically, the stage 217 is adjusted to move the nanodefect to a more central position within the beam, ensuring that the particle is within the measurement region of the longitudinal spin dipole field, i.e., near the optical axis. This ensures the nanodefect is within the measurement range of the longitudinal spin dipole field, improving detection accuracy.

[0148] The nanodefect identification technology provided in this application designs and utilizes a novel optical field—a longitudinal spin dipole optical field. The polarization of this field changes along the longitudinal direction, while the light intensity remains a bright spot structure throughout this process. Furthermore, by analyzing the backscattering of particles and using specific gravity, a relationship between specific gravity and the position of the nanodefect is established. This relationship is then used for measurement, ultimately enabling the identification of the longitudinal position of nanodefects with an accuracy approaching 200 pm. This high resolution for detecting the longitudinal position of nanodefects overcomes the diffraction limit of traditional optical measurements, achieving high-precision measurement of the longitudinal position of nanodefects. It maintains high resolution while also being fast, non-destructive, and easily integrated, meeting the future development needs of the nano-manufacturing and semiconductor industries.

[0149] Other embodiments of this application will readily occur to those skilled in the art upon consideration of the specification and practice of the application disclosed herein. This application is intended to cover any variations, uses, or adaptations of this application that follow the general principles of this application and include common knowledge or customary techniques in the art not disclosed herein.

[0150] It should be understood that this application is not limited to the precise structure described above and shown in the accompanying drawings, and various modifications and changes can be made without departing from its scope. The scope of this application is limited only by the appended claims.

Claims

1. A method for detecting nanodefects, characterized in that, The method for determining the longitudinal position of nanodefects in a product under test, wherein the longitudinal position is the position along the propagation direction of the test beam, includes: A longitudinal spin dipole field is generated, which includes two spin fields with opposite rotation directions. In the propagation direction of the test beam, the longitudinal spin dipole field gradually evolves from a spin field with one rotation direction to a spin field with the opposite rotation direction, and the longitudinal spin dipole field is a uniform bright spot. Obtain the scattered light field of the product under test in response to the longitudinal spin dipole light field; Separate the two spin light field components with different rotation directions in the scattered light field; Determine the ratio of the difference between the two spin optical field components with different rotation directions to the total intensity of the two spin optical fields with different rotation directions; The longitudinal position of the nanodefect is determined based on the specific gravity.

2. The method for detecting nanodefects according to claim 1, characterized in that, The generation of the longitudinal spin dipole optical field includes: The test beam is expanded and polarized; The test beam, after beam expansion and polarization modulation, is irradiated onto a diffractive optical element, the modulation factor of which is: ,in, The wavenumber of the test beam. To test the numerical aperture of the objective lens, The refractive index of the medium, The distance between the focal point of the oppositely rotating spin optical field and the center position of the oppositely rotating spin optical field; the diffractive optical element applies the modulation factor to the spin optical field of one spin direction and applies the conjugate of the modulation factor to the spin optical field of the other spin direction; The output of the diffractive optical element is mapped onto the incident surface of the detection objective, and the detection objective focuses to generate the longitudinal spin dipole light field.

3. The method for detecting nanodefects according to claim 2, characterized in that, The step of obtaining the scattered light field of the product under test in response to the longitudinal spin dipole light field includes: The scattered light field is collected through the incident surface of the detection objective lens; The scattered light field and the test beam are separated by a beam splitter cube.

4. The method for detecting nanodefects according to claim 1, characterized in that, The separation of the two spin light field components with different rotation directions in the scattered light field includes: The scattered light field is focused by a focusing lens and the two spin light field components with different rotation directions are separated by a polarization grating. Adjust the positions of the focusing lens and the polarization grating so that the two spin light field components with different rotation directions simultaneously illuminate the imaging plane of the CCD camera.

5. The method for detecting nanodefects according to claim 4, characterized in that, The determination of the ratio of the difference between the two spin optical field components with different rotation directions to the total intensity of the two spin optical fields with different rotation directions includes: The difference between the two spin light field components with different rotation directions and the total intensity of the two spin light fields with different rotation directions are determined based on the output of the CCD camera. The specific gravity is determined based on the difference between the two spin light field components with different rotation directions and the total intensity of the two spin light fields with different rotation directions.

6. The method for detecting nanodefects according to claim 1, characterized in that, Determining the longitudinal position of the nanodefect based on the specific gravity includes: Using a mapping table that establishes the correspondence between the calibrated longitudinal position of a nanodefect and its calibrated specific gravity, the longitudinal position of the nanodefect is determined based on the specific gravity.

7. The method for detecting nanodefects according to claim 6, characterized in that, The mapping table is obtained through a mapping table calibration method, which includes: After generating the longitudinal spin dipole field, the calibration product with nanodefects is moved along the propagation direction of the test beam, and the longitudinal calibration positions of multiple nanodefects are recorded. Obtain the calibration scattered light field of the calibration product on the longitudinal spin dipole light field; Separate the two spin light field components with different rotation directions in the calibrated scattered light field; Determine multiple calibration weights at multiple calibration longitudinal positions.

8. The method for detecting nanodefects according to claim 1, characterized in that, Before generating the longitudinal spin dipole optical field, the method further includes: The range of the lateral position of the nanodefect is obtained, wherein the lateral position is a position perpendicular to the propagation direction of the test beam; The product under test is moved in a direction perpendicular to the propagation direction of the test beam so that the nanodefect is located at the center of the test beam.

9. A system for detecting nanodefects, characterized in that, The detection system includes: A light field generation component is used to generate a longitudinal spin dipole light field, which includes two spin light fields with opposite rotation directions. In the propagation direction of the test beam, the longitudinal spin dipole light field gradually evolves from a spin light field with one rotation direction to a spin light field with the opposite rotation direction, and the longitudinal spin dipole light field is a uniform bright spot. A scattered light field acquisition component is located on the side of the light field generation component away from the product under test, and is used to acquire the scattered light field of the product under test in response to the longitudinal spin dipole light field. A separation component, located on the light-emitting side of the scattered light field acquisition component, is used to separate two spin light field components with different rotation directions in the scattered light field; The processor, connected to the separation component, is used to determine the ratio of the difference between the two spin optical field components with different rotation directions to the total intensity of the two spin optical fields with different rotation directions, and to determine the longitudinal position of the nanodefect based on the ratio, wherein the longitudinal position is the position along the propagation direction of the test beam.

10. The nanodefect detection system according to claim 9, characterized in that, The light field generating component includes: A beam generator for generating the test beam; A beam expander, located on the light-emitting side of the beam generator, is used to expand the test beam. A polarizer, located on the output side of the beam expander, is used to adjust the polarization of the expanded test beam. A diffractive optical element is disposed on the light-emitting side of the polarizer, and the tuning factor of the diffractive optical element is: ,in, The wavenumber of the test beam. To test the numerical aperture of the objective lens, The refractive index of the medium, The distance between the focal point of the oppositely rotating spin optical field and the center position of the oppositely rotating spin optical field; the diffractive optical element applies the modulation factor to the spin optical field of one spin direction and applies the conjugate of the modulation factor to the spin optical field of the other spin direction; A mapping component, located between the diffractive optical element and the detection objective, is used to map the output of the diffractive optical element onto the detection objective. The detection objective lens, located on the light-emitting side of the mapping component, is used to focus and generate the longitudinal spin dipole light field.

11. The nanodefect detection system according to claim 10, characterized in that, The scattered light field acquisition component includes: A beam splitter cube is located between the mapping component and the detection objective, the detection objective being used to collect the scattered light field, and the beam splitter cube being used to separate the scattered light field from the test beam.

12. The nanodefect detection system according to claim 11, characterized in that, The separation component includes: A focusing lens is disposed on the light-emitting side of the beam-splitting cube to focus the scattered light field; A polarization grating is disposed on the light-emitting side of the focusing lens to separate the focused scattered light field into two spin light field components with different rotation directions. A CCD camera is used to simultaneously image the two spin light field components with different rotation directions.

13. The nanodefect detection system according to claim 9, characterized in that, The processor is used to obtain the specific gravity and determine the longitudinal position of the nanodefect according to the mapping table of the correspondence between the calibrated longitudinal position of the nanodefect and the calibrated specific gravity.

Images