Wafer quality detection method in ultra-clean environment

Abstract

The invention relates to the technical field of wafer detection, in particular to a wafer quality detection method in an ultra-clean environment, which comprises the following steps: performing laser scanning detection on the surface of a wafer on a laser detection table, performing gas flushing on the wafer with an abnormal surface, and performing laser scanning detection again; when the laser detection table detects that the surface of the wafer has defects, performing thermal radiation heating on the surface of the wafer, and detecting the specific defects of the wafer through an infrared thermal imager; performing ultrasonic detection on the interior of the wafer, judging whether the wafer has an internal defect according to an ultrasonic reflection signal, generating an ultrasonic detection image of the wafer with the internal defect, and extracting defect data of the wafer with the internal defect by constructing a neural network; and generating a wafer quality detection report, and displaying a wafer detection result. According to the invention, multiple detection is carried out on the surface and the outside of the wafer, the detection efficiency is improved while the detection accuracy is improved, and the occupied resources during detection are saved.

Description

Technical Field

[0001] The present invention relates to the technical field of wafer detection, and in particular to a wafer quality detection method in an ultra-clean environment. Background Art

[0002] Wafer technology is the core of the semiconductor industry. In the early 1970s, with the introduction of Moore's Law, the diameter of wafers gradually increased from the initial 2 inches to the current 300 mm, and the integration of wafers increased rapidly. The requirements for wafer quality and process are also constantly increasing. Therefore, continuously improving wafer manufacturing technology and testing levels and ensuring wafer quality will be the key to the sustainable development of the semiconductor industry.

[0003] For today's large-area wafers, the problems of surface and internal defect detection deserve in-depth exploration. Large-area wafers not only need to detect surface defects, but also need to detect internal defects of the wafers. At the same time, the detection efficiency and detection cost also need to be considered. These are technical problems that need to be solved urgently.

[0004] The patent with the publication number CN116645362A discloses a quality intelligent detection method and system for silicon carbide wafers, and provides a quality intelligent detection method and system for silicon carbide wafers, the method includes: connecting a positioning stage, positioning and completing the acquisition of image acquisition results; constructing a basic comparison feature set, performing feature comparison on the image acquisition results, and determining the abnormal area; scanning the abnormal area, obtaining the abnormal area image acquisition results; collecting abnormal features of the same type of wafers, classifying and identifying and constructing an abnormal feature recognition model; inputting the abnormal area image acquisition results into the abnormal feature recognition model, initial recognition, calling the abnormal feature set, performing abnormal feature matching on the abnormal area image acquisition results, and outputting the abnormal detection results, which solves the technical problem of low defect detection efficiency of silicon carbide wafers, realizes automatic defect detection of silicon carbide wafers, and improves the technical effect of defect detection efficiency. This method only detects surface defects of wafers from a visual perspective, and does not deeply detect internal defects of wafers.

[0005] To this end, the present invention proposes a wafer quality detection method in an ultra-clean environment. Summary of the invention

[0006] To achieve the above object, the present invention provides a wafer quality detection method in an ultra-clean environment, and the specific technical scheme is as follows: A wafer quality detection method in an ultra-clean environment, characterized in that it includes: The finished wafers are transferred to the laser inspection station, where the laser scanning inspection is performed on the wafer surface. Wafers with abnormal surfaces are gas-flushed and laser scanning inspection is performed again; When the laser inspection station detects defects on the wafer surface, the wafer surface is heated by thermal radiation, and the specific defects of the wafer are detected by an infrared thermal imager; Perform ultrasonic testing on the inside of the wafer, determine whether the wafer has internal defects based on the ultrasonic reflection signal, generate ultrasonic testing images of the wafer with internal defects, and extract defect data of the wafer with internal defects by constructing a neural network;

[0007] Generate wafer quality inspection report and display wafer inspection results.

[0008] Preferably, the laser detection platform is used to emit a scanning laser beam to the wafer, and receive the laser beam reflected by the wafer through an optical sensor; When the wafer surface is flat, smooth and defect-free, the laser beam is totally reflected on the wafer surface, and the light intensity value of the reflected laser beam received by the optical sensor is fixed and evenly distributed, that is: ,in is the initial light intensity of the laser beam, It is the light intensity value of the reflected laser beam received by the optical sensor; When there is a defect on the wafer surface, the laser beam will be irregularly reflected at the defective part of the wafer surface. Total reflection will still occur at the flat, smooth and defect-free part of the wafer surface. The irregular reflection and total reflection light intensity are superimposed, and the light intensity value of some areas received by the optical sensor changes. The reflected light intensity of irregular reflection at the defective position of the wafer is calculated as follows: ; in, is the coordinate position of the wafer defect, is the standard deviation of the Gaussian distribution, It is the intensity of the reflected light that is irregularly reflected at the defective position of the wafer; ; in, The combined light intensity of total reflection and irregular reflection.

[0009] Preferably, a light intensity threshold is set and light intensity trigger count threshold , dividing the wafers into wafers with normal surfaces and wafers with abnormal surfaces; When the optical sensor receives the superimposed light intensity Greater than the light intensity threshold The number of times exceeds the light intensity triggering threshold , it is determined to be a wafer with abnormal surface; Flushing the wafers with abnormal surfaces with gas, wherein the gas is nitrogen; When the gas flushing is completed, the wafer is laser inspected again. If the optical sensor no longer shows abnormal light intensity, the ultrasonic inspection of the wafer interior is started. If the optical sensor still shows abnormal light intensity, the wafer with abnormal surface is determined to be a surface defective wafer, and the defect coordinate position of the wafer is reversely solved based on the laser beam light intensity distribution collected by the optical sensor: ; in: ; Given the light intensity distribution on the optical sensor, the objective function is minimized by the gradient descent method. The regional range of wafer defect locations is obtained.

[0010] Preferably, when a defect is detected on the wafer surface, a thermal radiation heating instrument is used to heat the area of ​​the wafer defect position, and a high-resolution infrared thermal imager is used to scan the wafer surface to capture infrared changes on the heated wafer surface; An infrared sensor is used to collect infrared characteristic images of the wafer after being heated by a thermal radiation heating instrument, the collected infrared characteristic images are filtered and denoised, and the images are segmented according to the wafer defect positions obtained by pre-detection; Perform data processing on the infrared characteristic image to find the temperature change area: ; in, Indicates the position in the infrared feature image Temperature changes, It's location The temperature after heating, It's location Background temperature; The color of the infrared characteristic image of the wafer is set as a grayscale image, the reference color of the grayscale image of the infrared characteristic image is set, the grayscale color of the largest area in the infrared characteristic image of the wafer is selected as the reference color, non-reference colors are highlighted, the positions of the non-reference colors are divided into box diagrams, the shape types of the non-reference colors in the box diagrams are identified, and the wafer defect types, the specific positions of the wafer defects, and the areas of the wafer defects are marked in the box diagrams; Set the impact weight threshold of wafer surface defects , when the wafer surface defect weight Greater than or equal to the weight threshold When the current wafer is cleared from the inspection process; when the wafer surface defect weight Less than the weight threshold When the wafer enters the next inspection process, the wafer surface defect weight The calculation formula is: ; in, is the number of cracks on the wafer surface, is the influence coefficient of cracks on the wafer surface, is the area of ​​the crack on the wafer surface; is the amount of wafer surface wear, is the influence coefficient of wafer surface wear, is the worn area of ​​the wafer surface.

[0011] Preferably, performing ultrasonic testing on the wafer after the surface testing to detect whether the wafer has internal defects includes: The wafer is placed on a uniformly textured ultrasonic testing table, and an ultrasonic probe is used to transmit an ultrasonic signal to the wafer and receive an ultrasonic reflection signal; According to the wafer thickness, set the ultrasonic signal frequency: ,in is the ultrasonic signal frequency; is the elastic modulus of the wafer; is the density of the wafer material; is the wafer thickness.

[0012] Preferably, the ultrasonic detection is used to detect internal delamination, internal cracks and internal voids of the wafer; Strategy for wafer internal layer detection: There are differences in acoustic impedance between different material layers, and the total reflection coefficient is used Describes the degree of reflection of ultrasound between different material layers: ; is the acoustic impedance of the Nth layer of material, , N is the total number of layers, is the density of the Nth layer of material, is the speed of sound in the Nth layer of material; The wafer is ultrasonically inspected in the ultrasonic inspection station, and the total reflection coefficient Below a fixed value, delamination exists in the wafer.

[0013] Preferably, the strategy for detecting cracks inside the wafer is: considering the influence of the shape and direction of the crack on the scattering amplitude, a crack reflection coefficient model including the crack direction angle is constructed: ; in, is the ultrasonic reflection coefficient of the crack inside the wafer, is the wafer diameter, The scattering amplitude varies with the incident angle and crack direction angle Distribution of Indicates the degree of ultrasonic absorption by the crack; The crack absorption coefficient represents the ultrasonic signal frequency, incident angle, and crack orientation angle; is the wafer thickness; Setting the reflectance threshold for wafer crack detection , when the ultrasonic reflection coefficient of the wafer crack Greater than threshold When the wafer has internal cracks; Strategy for void detection inside wafer: Consider the shape of the void, the impact of multipath propagation and interference effects on scattering, and build a void reflection coefficient model: ; in, is the ultrasonic reflection coefficient of the cavity inside the wafer, is the wafer diameter, represents the distribution of scattering amplitude with incident angle and hole position; The absorption coefficient of the cavity to ultrasound; The cavity absorption coefficient represents the relationship between the ultrasonic signal frequency, incident angle and cavity position; is the wafer thickness; Setting reflectance threshold for wafer void detection , when the ultrasonic reflection coefficient of the wafer cavity Greater than threshold There are internal voids in the wafer.

[0014] Preferably, when the wafer completes ultrasonic scanning, the ultrasonic data of the wafer with internal defects is visualized to generate an ultrasonic image, and a neural network model is constructed to identify the ultrasonic image; Construct a convolutional neural network to extract wafer defect features from ultrasonic images; The wafer defect characteristics include the area of ​​internal delamination of the wafer and quantity , the area of ​​cracks inside the wafer and quantity And the area of ​​the void inside the wafer and quantity ; Setting the impact weight of wafer internal defects , when the wafer internal defect weight Greater than or equal to weight When the defect weight inside the wafer is Less than weight When the wafer enters the next inspection process, the internal defect weight of the wafer The calculation method is as follows: ; in, is the influence weight of internal wafer delamination; is the influence weight of cracks inside the wafer; is the influence weight of the void inside the wafer.

[0015] Preferably, according to the surface inspection and internal inspection results of the wafer, wafers without defects in both the surface inspection and the internal inspection are classified as high-quality wafers; Wafers that meet the defect impact weights during wafer surface and internal inspections are classified as defective wafers, and a report file is generated. The report file is used to record the defect type, defect location, defect area, wafer number and wafer storage location of the defective wafer.

[0016] A computer server comprises: a processor and a memory; wherein the memory stores a computer program that can be called by the processor; the processor executes the wafer quality detection method in an ultra-clean environment by calling the computer program stored in the memory.

[0017] A computer-readable storage medium stores an erasable computer program; when the computer program is run on a computer device, the computer device executes the wafer quality detection method in an ultra-clean environment.

[0018] Beneficial effects of the present invention: The present invention not only detects surface defects of wafers, but also further detects internal defects of wafers; Through laser pre-detection, rapid pre-detection of wafer surface defects can be achieved, reducing the occupation of production resources by over-detection of wafer surfaces. Detailed detection of abnormal wafers can be performed to detect defects on the wafer surface. The present invention also performs ultrasonic testing on the inside of the wafer to further control the quality of the wafer. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] Figure 1 A flow chart of a wafer quality detection method in an ultra-clean environment provided by the present invention; Figure 2 This is a structural diagram of the wafer quality detection method in an ultra-clean environment provided by the present invention. DETAILED DESCRIPTION

[0020] In order to better understand the present invention, a more detailed description will be made of various aspects of the present invention with reference to the accompanying drawings. It should be understood that these detailed descriptions are only descriptions of exemplary embodiments of the present invention, and are not intended to limit the scope of the present invention in any way. Throughout the specification, the same reference numerals refer to the same elements. The expression "and / or" includes any and all combinations of one or more of the associated listed items.

[0021] In the accompanying drawings, the size, dimensions and shape of the elements have been slightly adjusted for ease of illustration. The accompanying drawings are for illustration only and are not strictly drawn to scale. As used herein, the terms "substantially", "approximately" and similar terms are used as terms of approximation, not as terms of degree, and are intended to illustrate the inherent deviations in measurements or calculations that will be recognized by those of ordinary skill in the art. In addition, in the present invention, the order in which the steps are processed does not necessarily represent the order in which these processes occur in actual operation, unless otherwise specified or can be derived from the context.

[0022] It should also be understood that expressions such as "comprises", "including", "having", "includes" and / or "comprising" are open rather than closed expressions in this specification, which indicate the presence of the stated features, elements and / or components, but do not exclude the presence of one or more other features, elements, components and / or combinations thereof. In addition, when expressions such as "at least one of..." appear after a list of listed features, they modify the entire list of features rather than just the individual elements in the list. In addition, when describing embodiments of the present invention, "may" is used to mean "one or more embodiments of the present invention". And, the term "exemplary" is intended to refer to an example or illustration.

[0023] Unless otherwise defined, all terms (including engineering terms and scientific and technological terms) used in this document have the same meaning as commonly understood by ordinary technicians in the field to which the present invention belongs. It should also be understood that unless otherwise clearly stated in the present invention, words defined in commonly used dictionaries should be interpreted as having the same meaning as their meaning in the context of the relevant technology, and should not be interpreted in an idealized or overly formal sense.

[0024] It should be noted that, in the absence of conflict, the embodiments and features of the embodiments of the present invention can be combined with each other. The present invention will be described in detail below with reference to the accompanying drawings and in combination with the embodiments.

[0025] Example 1 Reference Figure 1 , which is the first embodiment of the present invention, provides a wafer quality inspection method in an ultra-clean environment.

[0026] S1: The finished wafers are transferred to the laser inspection station, where the surface of the wafers is inspected by laser scanning. Wafers with abnormal surfaces are flushed with gas and then inspected by laser scanning again.

[0027] The laser detection platform is used for emitting a scanning laser beam to the wafer and receiving the laser beam reflected by the wafer through an optical sensor.

[0028] The laser beam emitted to the wafer is reflected on the wafer surface and received by the optical sensor. When a defect appears on the wafer surface, the reflected laser beam changes the propagation path. The optical sensor determines whether there is a defect on the wafer surface based on the light intensity distribution received by the optical sensor.

[0029] When the wafer surface is flat, smooth and defect-free, the laser beam is totally reflected on the wafer surface, and the light intensity value of the reflected laser beam received by the optical sensor is fixed and evenly distributed, that is: ,in is the initial light intensity of the laser beam, It is the light intensity value of the reflected laser beam received by the optical sensor.

[0030] When there is a defect on the wafer surface, the incident laser beam will be irregularly reflected at the defect on the wafer surface. Total reflection will still occur at the flat, smooth and defect-free surface of the wafer. The irregular reflection and total reflection light intensity are superimposed, and the light intensity value of some areas received by the optical sensor changes. The reflected light intensity of irregular reflection at the defect position of the wafer is calculated as follows: ; in, is the coordinate position of the wafer defect; is the standard deviation of the Gaussian distribution, It is the intensity of the reflected light caused by irregular reflection at the defective position of the wafer.

[0031] ; in, It is the superimposed light intensity of total reflection and irregular reflection.

[0032] Set light intensity threshold and light intensity trigger count threshold , the wafers are divided into wafers with normal surfaces and wafers with abnormal surfaces.

[0033] When the optical sensor receives the superimposed light intensity Greater than the light intensity threshold The number of times exceeds the light intensity triggering threshold , it is judged as a wafer with abnormal surface.

[0034] The wafer with surface abnormality is flushed with gas, wherein the gas is nitrogen.

[0035] When performing gas flushing on wafers with abnormal surfaces, the wafers are placed on a flushing table and the force is evenly applied to the wafers. The gas flushing nozzle flushes the wafers to prevent the wafers from being damaged during gas flushing.

[0036] When the gas flushing is completed, the wafer is laser inspected again. If the optical sensor no longer shows abnormal intensity, the ultrasonic inspection of the wafer interior is started. If the optical sensor still shows abnormal light intensity for many times, the wafer with abnormal surface is determined to be a surface defective wafer, and the defect coordinate position of the wafer is reversely solved based on the laser beam light intensity distribution collected by the optical sensor: ; in: ; is the light intensity distribution on the optical sensor, and the objective function is minimized by gradient descent. The regional range of wafer defect locations is obtained.

[0037] The steps of the gradient descent method are as follows: When the surface defect image of the wafer is obtained, the objective function is defined .

[0038] Calculate the objective function Relative to the parameters The gradient indicates the direction of change of the target Hanna tree in the parameter space: ; All partial derivatives form a vector, representing the objective function At the point The gradient at ; Update , reduce the objective function value: ; in, is the learning rate, which indicates the step size of parameter update in each iteration.

[0039] Repeat the above steps until the gradient approaches zero.

[0040] S2: When the laser inspection station detects defects on the wafer surface, the wafer surface is heated by thermal radiation, and the specific defects of the wafer are detected by an infrared thermal imager.

[0041] When defects are detected on the wafer surface, a thermal radiation heating instrument is used to heat the area at the defective location of the wafer. The heating temperature is half of the maximum temperature the wafer can withstand. A high-resolution infrared thermal imager is used to scan the wafer surface to capture infrared changes in the heated wafer surface. When defects appear on the wafer surface, the location of the wafer defect has a thermal difference from the normal area of ​​the wafer when heated by the thermal radiation heating instrument, and the infrared characteristics of the wafer defect area are significantly different from those of the normal wafer area.

[0042] An infrared sensor is used to collect infrared characteristic images of the wafer after being heated by a thermal radiation heating instrument, the collected infrared characteristic images are filtered and denoised, and the defect area image is segmented according to the wafer defect position obtained by pre-detection.

[0043] Perform data processing on the infrared characteristic image to find the temperature change area: ; in, Indicates the position in the infrared feature image Temperature changes, It's location The temperature after heating, It's location background temperature.

[0044] The color of the wafer infrared characteristic image is set as a grayscale image, the reference color of the infrared characteristic image grayscale image is set, the grayscale color of the largest area in the wafer infrared characteristic image is selected as the reference color, the non-reference colors are highlighted, the positions of the non-reference colors are divided into boxes, the shape types of the non-reference colors in the boxes are identified, and the wafer defect types, the specific positions of the wafer defects, and the areas of the wafer defects are marked in the boxes.

[0045] Set the impact weight threshold of wafer surface defects , when the wafer surface defect weight Greater than or equal to the weight threshold When the current wafer is cleared from the inspection process; when the wafer surface defect weight Less than the weight threshold When the wafer enters the next inspection process, the wafer surface defect weight The calculation formula is: ; in, is the number of cracks on the wafer surface, is the influence coefficient of cracks on the wafer surface, is the area of ​​the crack on the wafer surface; is the amount of wafer surface wear, is the influence coefficient of wafer surface wear, is the worn area of ​​the wafer surface.

[0046] S3: Perform ultrasonic inspection on the inside of the wafer, determine whether the wafer has internal defects based on the ultrasonic reflection signal, generate an ultrasonic inspection image of the wafer with internal defects, and extract defect data of the wafer with internal defects by constructing a neural network.

[0047] Perform ultrasonic testing on the wafer after surface inspection to detect whether there are internal defects in the wafer.

[0048] The wafer is placed on a uniformly textured ultrasonic testing table, and an ultrasonic probe is used to transmit ultrasonic signals to the wafer and receive ultrasonic reflection signals.

[0049] According to the wafer thickness, set the ultrasonic signal frequency: ,in is the ultrasonic signal frequency; is the elastic modulus of the wafer; is the wafer material density; is the wafer thickness.

[0050] The ultrasonic testing is used to detect internal delamination, internal cracks and internal voids of the wafer.

[0051] Strategy for wafer internal layer detection: There are differences in acoustic impedance between different material layers, and the total reflection coefficient is used Describes the degree of reflection of ultrasound between different material layers: ; is the acoustic impedance of the Nth layer of material, , N is the total number of layers, is the density of the Nth layer of material, is the speed of sound in the Nth layer of material.

[0052] The wafer is ultrasonically tested in the ultrasonic testing station. The total reflection coefficient is a fixed value. If the total reflection coefficient is Below a fixed value, delamination exists in the wafer.

[0053] Strategy for detecting cracks inside wafers: Considering the influence of crack shape and direction on scattering amplitude, a crack reflection coefficient model including crack direction angle is constructed: ; in, is the ultrasonic reflection coefficient of the crack inside the wafer, is the wafer diameter, The scattering amplitude varies with the incident angle and crack direction angle Distribution of Indicates the degree of ultrasonic absorption by the crack; The crack absorption coefficient represents the ultrasonic signal frequency, incident angle, and crack orientation angle; is the wafer thickness.

[0054] Setting the reflectance threshold for wafer crack detection , when the ultrasonic reflection coefficient of the wafer crack Greater than threshold There are internal cracks in the wafer.

[0055] Strategy for void detection inside wafer: Consider the shape of the void, the impact of multipath propagation and interference effects on scattering, and construct a void reflection coefficient model.

[0056] ; in, is the ultrasonic reflection coefficient of the cavity inside the wafer, is the wafer diameter, represents the distribution of scattering amplitude with incident angle and hole position; The absorption coefficient of the cavity to ultrasound; The cavity absorption coefficient represents the relationship between the ultrasonic signal frequency, incident angle and cavity position; is the wafer thickness.

[0057] Setting reflectance threshold for wafer void detection , when the ultrasonic reflection coefficient of the wafer cavity Greater than threshold There are internal voids in the wafer.

[0058] When the wafer completes ultrasonic scanning, the ultrasonic data of the wafer with internal defects is visualized to generate an ultrasonic image, and a neural network model is built to identify the ultrasonic image.

[0059] Construct a convolutional neural network to extract wafer defect features from ultrasonic images.

[0060] The wafer defect characteristics include the area of ​​internal delamination of the wafer and quantity , the area of ​​cracks inside the wafer and quantity And the area of ​​the void inside the wafer and quantity .

[0061] Setting the impact weight of wafer internal defects , when the wafer internal defect weight Greater than or equal to the weight threshold When the defect weight inside the wafer is Less than the weight threshold When the wafer enters the next inspection process, the internal defect weight of the wafer The calculation method is as follows: ; in, is the influence weight of internal wafer delamination; is the influence weight of cracks inside the wafer; is the influence weight of the void inside the wafer.

[0062] S4: Generate a wafer quality inspection report and display the wafer inspection results.

[0063] Based on the surface inspection and internal inspection results of the wafer, wafers without defects in both surface inspection and internal inspection are classified as high-quality wafers.

[0064] Wafers that have defects detected by wafer surface inspection and internal inspection but meet the defect impact weights are classified as defective wafers and a report file is generated.

[0065] The report file is used to record the defect type, defect location, defect area, wafer number and wafer storage location of the defective wafer.

[0066] Example 2

[0067] Reference Figure 2 , a second embodiment of the present invention provides a wafer quality inspection system in an ultra-clean environment.

[0068] The system includes a laser detection module, an infrared detection module, an ultrasonic detection module and a quality inspection report module.

[0069] The laser detection module is used to transfer the finished wafers to the laser detection station, perform laser scanning detection on the wafer surface on the laser detection station, perform gas flushing on wafers with abnormal surfaces, and perform laser scanning detection again.

[0070] The infrared detection module is used to heat the wafer surface by thermal radiation when the laser detection platform detects defects on the wafer surface, and detect the specific defects of the wafer through an infrared thermal imager.

[0071] The ultrasonic detection module is used to perform ultrasonic detection on the inside of the wafer, determine whether the wafer has internal defects based on the ultrasonic reflection signal, generate an ultrasonic detection image of the wafer with internal defects, and extract defect data of the wafer with internal defects by constructing a neural network.

[0072] The quality inspection report module is used to generate a wafer quality inspection report and display the wafer inspection results.

[0073] Example 3

[0074] The present invention also provides an electronic device. The electronic device may include one or more processors and one or more memories. The memories store computer-readable codes, and when the computer-readable codes are run by the one or more processors, the wafer quality detection method in an ultra-clean environment as described above may be executed.

[0075] The method or system according to the embodiment of the present invention may also be implemented with the aid of the architecture of the electronic device of the present invention.

[0076] An electronic device may include a bus, one or more CPUs, read-only memory (ROM), random access memory (RAM), a communication port for connecting to a network, input / output components, a hard disk, and the like.

[0077] A storage device in an electronic device, such as a ROM or a hard disk, can store a wafer quality detection method in an ultra-clean environment provided by the present invention.

[0078] A wafer quality detection method in an ultra-clean environment comprises the following steps: transmitting a wafer that has been produced to a laser detection station, performing laser scanning detection on the surface of the wafer on the laser detection station, performing gas flushing on wafers with abnormal surfaces, and performing laser scanning detection again; when the laser detection station detects that there are defects on the surface of the wafer, the surface of the wafer is heated by thermal radiation, and specific defects of the wafer are detected by an infrared thermal imager; ultrasonic detection is performed on the inside of the wafer, and it is determined whether the wafer has internal defects based on ultrasonic reflection signals, an ultrasonic detection image of the wafer with internal defects is generated, and defect data of the wafer with internal defects is extracted by constructing a neural network; and a wafer quality detection report is generated to display the wafer detection results.

[0079] Furthermore, the electronic device may also include a user interface. Of course, the architecture of the present invention is only exemplary, and when implementing different devices, one or more components in the electronic device disclosed in the present invention may be omitted according to actual needs.

[0080] Example 4

[0081] The invention also discloses a computer-readable storage medium.

[0082] The computer-readable storage medium has computer-readable instructions stored thereon.

[0083] When the computer-readable instructions are executed by a processor, a wafer quality inspection method in an ultra-clean environment according to an embodiment of the present invention described with reference to the above drawings may be executed.

[0084] The storage medium includes, but is not limited to, for example, volatile memory and / or non-volatile memory. Volatile memory may include, for example, random access memory (RAM) and cache memory (cache), etc. Non-volatile memory may include, for example, read-only memory (ROM), hard disk, flash memory, etc. In addition, according to an embodiment of the present invention, the process described above with reference to the flowchart may be implemented as a computer software program.

[0085] For example, the present invention provides a non-temporary machine-readable storage medium, which stores machine-readable instructions, and the machine-readable instructions can be run by a processor to execute instructions corresponding to the method steps provided by the present invention, for example: transferring the completed wafer to a laser inspection station, performing laser scanning inspection on the wafer surface on the laser inspection station, gas flushing the wafer with abnormal surface, and laser scanning inspection again; when the laser inspection station detects that there are defects on the wafer surface, the wafer surface is heated by thermal radiation, and the specific defects of the wafer are detected by an infrared thermal imager; ultrasonic inspection is performed on the inside of the wafer, and it is determined whether the wafer has internal defects based on the ultrasonic reflection signal, and an ultrasonic inspection image of the wafer with internal defects is generated, and the defect data of the wafer with internal defects is extracted by constructing a neural network; a wafer quality inspection report is generated to display the wafer inspection results.

[0086] When the computer program is executed by the central processing unit (CPU), the above functions defined in the method of the present invention are performed. The method, apparatus, and device of the present invention may be implemented in many ways. For example, the method, apparatus, and device of the present invention may be implemented by software, hardware, firmware, or any combination of software, hardware, and firmware.

[0087] The above sequence for the steps of the method is for illustration only, and the steps of the method of the present invention are not limited to the sequence specifically described above unless otherwise specifically stated.

[0088] In addition, in some embodiments, the present invention can also be implemented as a program recorded in a recording medium, which includes machine-readable instructions for implementing the method according to the present invention. Therefore, the present invention also covers a recording medium storing a program for executing the method according to the present invention.

[0089] In addition, the parts of the above technical solutions provided in the embodiments of the present invention that are consistent with the implementation principles of the corresponding technical solutions in the prior art are not described in detail to avoid excessive redundancy.

[0090] The specific implementation modes as described above further describe the purpose, technical solutions and beneficial effects of the present invention in detail. It should be understood that the above description is only a specific implementation mode of the present invention and is not intended to limit the present invention. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.

Claims

1. A wafer quality detection method in an ultra-clean environment, characterized in that: include: The finished wafers are transferred to the laser inspection station, where the laser scanning inspection is performed on the wafer surface. Wafers with abnormal surfaces are gas-flushed and laser scanning inspection is performed again; When the laser inspection station detects defects on the wafer surface, the wafer surface is heated by thermal radiation, and the specific defects of the wafer are detected by an infrared thermal imager; Perform ultrasonic testing on the inside of the wafer, determine whether the wafer has internal defects based on the ultrasonic reflection signal, generate ultrasonic testing images of the wafer with internal defects, and extract defect data of the wafer with internal defects by constructing a neural network; Generate wafer quality inspection report and display wafer inspection results.

2. The wafer quality inspection method in an ultra-clean environment according to claim 1, characterized in that: The laser detection platform is used to emit a scanning laser beam to the wafer, and receive the laser beam reflected by the wafer through an optical sensor; When the wafer surface is flat, smooth and defect-free, the laser beam is totally reflected on the wafer surface, and the light intensity value of the reflected laser beam received by the optical sensor is fixed and evenly distributed, that is: ,in is the initial light intensity of the laser beam, It is the light intensity value of the reflected laser beam received by the optical sensor; When there is a defect on the wafer surface, the laser beam will be irregularly reflected at the defective part of the wafer surface. Total reflection will still occur at the flat, smooth and defect-free part of the wafer surface. The irregular reflection and total reflection light intensity are superimposed, and the light intensity value of some areas received by the optical sensor changes. The reflected light intensity of irregular reflection at the defective position of the wafer is calculated as follows: ； in, is the coordinate position of the wafer defect, is the standard deviation of the Gaussian distribution, It is the intensity of the reflected light that is irregularly reflected at the defective position of the wafer; ； in, It is the superimposed light intensity of total reflection and irregular reflection.

3. The wafer quality inspection method in an ultra-clean environment according to claim 2, characterized in that: Set light intensity threshold and light intensity trigger count threshold , dividing the wafers into wafers with normal surfaces and wafers with abnormal surfaces; When the optical sensor receives the superimposed light intensity Greater than the light intensity threshold The number of times exceeds the light intensity triggering threshold , it is determined to be a wafer with abnormal surface; Flushing the wafers with abnormal surfaces with gas, wherein the gas is nitrogen; When the gas flushing is completed, the wafer is laser inspected again. If the optical sensor no longer shows abnormal light intensity, the ultrasonic inspection of the wafer interior is started. If the optical sensor still shows abnormal light intensity, the wafer with abnormal surface is determined to be a surface defective wafer, and the defect coordinate position of the wafer is reversely solved based on the laser beam light intensity distribution collected by the optical sensor: ； in: ； Given the light intensity distribution on the optical sensor, the objective function is minimized by the gradient descent method. The regional range of wafer defect locations is obtained.

4. The wafer quality inspection method in an ultra-clean environment according to claim 3, characterized in that: When defects are detected on the wafer surface, a thermal radiation heating instrument is used to heat the area of ​​the wafer defect location, and a high-resolution infrared thermal imager is used to scan the wafer surface to capture infrared changes on the heated wafer surface; An infrared sensor is used to collect infrared characteristic images of the wafer after being heated by a thermal radiation heating instrument, the collected infrared characteristic images are filtered and denoised, and the images are segmented according to the wafer defect positions obtained by pre-detection; Perform data processing on the infrared characteristic image to find the temperature change area: ； in, Indicates the position in the infrared feature image Temperature changes, It's location The temperature after heating, It's location Background temperature; The color of the infrared characteristic image of the wafer is set as a grayscale image, the reference color of the grayscale image of the infrared characteristic image is set, the grayscale color of the largest area in the infrared characteristic image of the wafer is selected as the reference color, non-reference colors are highlighted, the positions of the non-reference colors are divided into box diagrams, the shape types of the non-reference colors in the box diagrams are identified, and the wafer defect types, the specific positions of the wafer defects, and the areas of the wafer defects are marked in the box diagrams; Set the impact weight threshold of wafer surface defects , when the wafer surface defect weight Greater than or equal to the weight threshold When the current wafer is cleared from the inspection process; when the wafer surface defect weight Less than the weight threshold When the wafer enters the next inspection process, the wafer surface defect weight The calculation formula is: ； in, is the number of cracks on the wafer surface, is the influence coefficient of cracks on the wafer surface, is the area of ​​the crack on the wafer surface; is the amount of wafer surface wear, is the influence coefficient of wafer surface wear, is the worn area of ​​the wafer surface.

5. The wafer quality inspection method in an ultra-clean environment according to claim 4, characterized in that: Perform ultrasonic testing on the wafer after surface inspection to detect whether there are internal defects in the wafer, including: The wafer is placed on a uniformly textured ultrasonic testing table, and an ultrasonic probe is used to transmit an ultrasonic signal to the wafer and receive an ultrasonic reflection signal; According to the wafer thickness, set the ultrasonic signal frequency: ,in is the ultrasonic signal frequency; is the elastic modulus of the wafer; is the density of the wafer material; is the wafer thickness.

6. The wafer quality inspection method in an ultra-clean environment according to claim 5, characterized in that: The ultrasonic testing is used to detect internal delamination, internal cracks and internal voids of the wafer; Strategy for wafer internal layer detection: There are differences in acoustic impedance between different material layers, and the total reflection coefficient is used Describes the degree of reflection of ultrasound between different material layers: ； is the acoustic impedance of the Nth layer of material, , N is the total number of layers, is the density of the Nth layer of material, is the speed of sound in the Nth layer of material; The wafer is ultrasonically tested in the ultrasonic testing station. The total reflection coefficient is a fixed value. If the total reflection coefficient is Below a fixed value, delamination exists in the wafer.

7. The wafer quality inspection method in an ultra-clean environment according to claim 6, characterized in that: Strategy for detecting cracks inside wafers: Considering the influence of crack shape and direction on scattering amplitude, a crack reflection coefficient model including crack direction angle is constructed: ； in, is the ultrasonic reflection coefficient of the crack inside the wafer, is the wafer diameter, The scattering amplitude varies with the incident angle and crack direction angle Distribution of Indicates the degree of ultrasonic absorption by the crack; The crack absorption coefficient represents the ultrasonic signal frequency, incident angle, and crack orientation angle; is the wafer thickness; Setting the reflectance threshold for wafer crack detection , when the ultrasonic reflection coefficient of the wafer crack Greater than threshold When , there are internal cracks in the wafer; Strategy for void detection inside wafer: Consider the shape of the void, the impact of multipath propagation and interference effects on scattering, and build a void reflection coefficient model: ； in, is the ultrasonic reflection coefficient of the cavity inside the wafer, is the wafer diameter, represents the distribution of scattering amplitude with incident angle and hole position; The absorption coefficient of the cavity to ultrasound; The cavity absorption coefficient represents the relationship between the ultrasonic signal frequency, incident angle and cavity position; is the wafer thickness; Setting reflectance threshold for wafer void detection , when the ultrasonic reflection coefficient of the wafer cavity Greater than threshold There are internal voids in the wafer.

8. The wafer quality inspection method in an ultra-clean environment according to claim 7, characterized in that: When the wafer completes ultrasonic scanning, the ultrasonic data of the wafer with internal defects is visualized to generate ultrasonic images, and a neural network model is built to identify the ultrasonic images; Construct a convolutional neural network to extract wafer defect features from ultrasonic images; The wafer defect characteristics include the area of ​​internal delamination of the wafer and quantity , the area of ​​cracks inside the wafer and quantity And the area of ​​the void inside the wafer and quantity ; Setting the impact weight of wafer internal defects , when the wafer internal defect weight Greater than or equal to weight When the defect weight inside the wafer is Less than weight When the wafer enters the next inspection process, the internal defect weight of the wafer The calculation method is as follows: ； in, is the influence weight of internal wafer delamination; is the influence weight of cracks inside the wafer; is the influence weight of the void inside the wafer.

9. The wafer quality inspection method in an ultra-clean environment according to claim 8, characterized in that: According to the surface inspection and internal inspection results of the wafer, the wafers without defects in both the surface inspection and the internal inspection are classified as high-quality wafers; Wafers that meet the defect impact weights during wafer surface and internal inspections are classified as defective wafers, and a report file is generated. The report file is used to record the defect type, defect location, defect area, wafer number and wafer storage location of the defective wafer.

10. A computer server, characterized in that: include: A processor and a memory; wherein the memory stores a computer program that can be called by the processor; the processor executes the wafer quality detection method in an ultra-clean environment described in any one of claims 1 to 9 by calling the computer program stored in the memory.

11. A computer-readable storage medium, characterized in that: An erasable computer program is stored thereon; when the computer program is run on a computer device, the computer device executes a wafer quality detection method in an ultra-clean environment as described in any one of claims 1-9.

Images