Surface particle defect detection and removal method and computer readable storage medium

Abstract

The invention relates to a surface particle defect detection and removal method and a computer readable storage medium. The method comprises the steps that defect detection equipment acquires gray scale features of the surface of a detected wafer; inputting the gray scale features and the process level of the tested wafer when the gray scale features are obtained into a surface defect identification model, and carrying out surface particle defect identification; the particle removing mechanism moves to the position corresponding to the recognized surface particle defect, and particle removing treatment is carried out; detecting the processed to-be-detected wafer through the defect detection equipment, and judging whether the identified surface particle defect is removed or not; and the surface defect identification model carries out re-learning according to the judgment result. According to the invention, the production efficiency of products can be improved, the product yield is improved, the influence of reworking on product circulation is reduced, and the workload of technicians is reduced.

Description

Technical Field

[0001] The present application relates to semiconductor manufacturing, and more particularly to a method for detecting and removing surface particle defects, and also to a computer-readable storage medium. Background Art

[0002] The semiconductor integrated circuit (IC) industry has experienced exponential growth. As semiconductor processes become more complex, production cycles become longer and longer. Process quality control is an important control method for high-quality semiconductor operations. Inline Defect Monitor Step refers to a defect inspection station after a process loop is completed during the production of semiconductor wafers. By scanning product defects with a defect scanning machine, the process stability of the wafer product can be directly monitored in real time. Figure 1 shown.

[0003] If the defect inspection site detects an abnormality, human judgment is required to decide how to do the rework, which will affect the speed of product circulation. Summary of the invention

[0004] Based on this, it is necessary to provide a method for detecting and removing surface particle defects that can improve production efficiency.

[0005] A method for detecting and removing surface particle defects, comprising: a defect detection device obtains the process level at which a wafer under test is located, and obtains the grayscale features of the surface of the wafer under test at the process level; the process level and the grayscale features are input into a surface defect recognition model to identify surface particle defects; a particle removal mechanism moves to a position corresponding to the identified surface particle defect, and performs a particle removal process; the processed wafer under test is detected by the defect detection device to determine whether the identified surface particle defect has been removed; and the surface defect recognition model is relearned according to the result of the determination.

[0006] In the above-mentioned surface particle defect detection and removal method, the surface defect recognition model recognizes surface particle defects according to grayscale features and process levels, can scientifically and effectively recognize surface particle defects, and control the particle removal mechanism to remove the surface particle defects according to the position of the surface particle defects, and then judge whether the recognition of the surface particle defects is correct according to the result of the surface particle defect removal, and feed back to the surface defect recognition model for re-learning, so as to continuously optimize the recognition accuracy of the surface defect recognition model. Therefore, this method can improve the production efficiency of the product, improve the product yield and reduce the impact of rework on product circulation.

[0007] In one of the embodiments, the method also includes a step of obtaining the size of the grayscale abnormal area; the step of identifying surface particle defects includes: judging whether the grayscale abnormal area is a surface particle defect based on the process level, size and grayscale characteristics of the grayscale abnormal area.

[0008] In one of the embodiments, the surface defect recognition model is configured to classify grayscale abnormal regions of different sizes into sets of different size intervals for training.

[0009] In one embodiment, the classification of grayscale abnormal regions of different sizes into sets of different size intervals for training includes: classifying the grayscale abnormal regions into sets with size characteristic values ​​less than 0.2 microns, or ≥0.2 microns and less than 1 micron, or ≥1 micron.

[0010] In one embodiment, the surface defect recognition model is configured to classify grayscale abnormal regions of different process levels into different sets for training.

[0011] In one embodiment, the particle removal mechanism includes an air sweep module.

[0012] In one of the embodiments, the step of performing particle removal processing includes: adjusting the wind pressure of the wind sweep module according to the size of the grayscale abnormal area.

[0013] In one of the embodiments, the wind sweep module includes a positioning unit, and the positioning unit is used to synchronize with the coordinate information of the defect detection device.

[0014] In one of the embodiments, the defect detection device obtains the process level at which the wafer under test is located, and the step of obtaining the grayscale characteristics of the surface of the wafer under test at the process level includes: generating a grayscale image of the surface of the wafer under test.

[0015] In one embodiment, the step of re-learning based on the result of the judgment includes: if the identified surface particle defect has not been removed, then feeding back information representing that the recognition result is wrong to the surface defect recognition model; if the identified surface particle defect has been removed, then feeding back information representing that the recognition result is correct to the surface defect recognition model.

[0016] It is also necessary to provide a computer-readable storage medium having a computer program stored thereon, which implements the steps in the method of any of the above embodiments when the computer program is executed by a processor.

[0017] It is also necessary to provide a computer device, including a memory and a processor, wherein a computer program is stored in the memory, and when the processor executes the computer program, the steps in the method described in any of the above embodiments are implemented.

[0018] It is also necessary to provide a computer program product, comprising a computer program, which, when executed by a processor, implements the steps of the method described in any of the aforementioned embodiments.

[0019] It is also necessary to provide a surface particle defect detection and removal system, including a defect detection device, a particle removal mechanism and an artificial intelligence module, wherein the artificial intelligence module is configured with a surface defect recognition model; the defect detection device is used to obtain the grayscale features of the surface of the wafer under test, and the artificial intelligence module is used to input the grayscale features and the process level at which the wafer under test is located when the grayscale features are obtained into the surface defect recognition model to perform surface particle defect recognition; the particle removal mechanism is used to move to the position corresponding to the identified surface particle defect and perform particle removal processing; the artificial intelligence module is also used to detect the processed wafer under test through the defect detection device, determine whether the identified surface particle defect has been removed, and enable the surface defect recognition model to be re-learned according to the result of the judgment.

[0020] In one embodiment, the defect detection device obtains the grayscale features of the surface of the wafer under test, including generating a grayscale image of the surface of the wafer under test, and obtaining the size and grayscale features of the grayscale abnormal area according to the grayscale image; the artificial intelligence module performs surface particle defect recognition, including judging whether the grayscale abnormal area is a surface particle defect according to the process level, size and grayscale features of the grayscale abnormal area.

[0021] In one embodiment, the particle removal mechanism includes an air sweep module.

[0022] In one of the embodiments, the wind sweep module includes a positioning unit, which is used to synchronize with coordinate information of the defect detection device.

[0023] In one of the embodiments, the wind sweep module adjusts the wind pressure of the wind sweep module according to the size of the grayscale abnormal area.

[0024] In one of the embodiments, if the identified surface particle defect has not been removed, the artificial intelligence module will feedback information indicating that the recognition result is wrong to the surface defect recognition model; if the identified surface particle defect has been removed, the artificial intelligence module will feedback information indicating that the recognition result is correct to the surface defect recognition model.

[0025] In one of the embodiments, the surface defect recognition model is configured to classify grayscale abnormal regions of different sizes into sets of different size intervals for training.

[0026] In one embodiment, the classification of grayscale abnormal regions of different sizes into sets of different size intervals for training includes: classifying the grayscale abnormal regions into sets with size characteristic values ​​less than 0.2 microns, or ≥0.2 microns and less than 1 micron, or ≥1 micron.

[0027] In one embodiment, the surface defect recognition model is configured to classify grayscale abnormal regions of different process levels into different sets for training. BRIEF DESCRIPTION OF THE DRAWINGS

[0028] In order to better describe and illustrate the embodiments and / or examples of the inventions disclosed herein, reference may be made to one or more drawings. The additional details or examples used to describe the drawings should not be considered as limiting the scope of any of the disclosed inventions, the embodiments and / or examples currently described, and the best modes of these inventions currently understood.

[0029] Figure 1 It is a schematic diagram of defect monitoring at an online monitoring station;

[0030] Figure 2 is a flow chart of a method for detecting and removing surface particle defects in one embodiment of the present application;

[0031] Figure 3 is a schematic diagram of an exemplary grayscale image;

[0032] Figure 4 is a schematic diagram of an exemplary wind sweep module for removing surface particle defects;

[0033] Figure 5 This is a comparison photo of particle defects on the wafer surface before and after wind sweeping;

[0034] Figure 6 is a schematic diagram of a surface defect recognition model in an embodiment of the present application for classifying defects according to their sizes;

[0035] Figure 7 is a schematic diagram of a box surrounding a defect. DETAILED DESCRIPTION

[0036] In order to facilitate the understanding of the present invention, the present invention will be described more fully below with reference to the relevant drawings. The preferred embodiments of the present invention are shown in the drawings. However, the present invention can be implemented in many different forms and is not limited to the embodiments described herein. On the contrary, the purpose of providing these embodiments is to make the disclosure of the present invention more thorough and comprehensive.

[0037] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as those commonly understood by those skilled in the art of the present invention. The terms used herein in the specification of the present invention are only for the purpose of describing specific embodiments and are not intended to limit the present invention. The term "and / or" used herein includes any and all combinations of one or more related listed items.

[0038] The purpose of the terms used herein is only to describe specific embodiments and is not intended to be limiting of the present invention. When used herein, the singular forms "one", "an" and "said / the" are also intended to include plural forms, unless the context clearly indicates otherwise. It should also be understood that the terms "consisting of" and / or "comprising", when used in this specification, determine the presence of the features, integers, steps, operations, elements and / or parts, but do not exclude the presence or addition of one or more other features, integers, steps, operations, elements, parts and / or groups. When used herein, the term "and / or" includes any and all combinations of the relevant listed items.

[0039] Figure 2 : is a flow chart of a method for detecting and removing surface particle defects in an embodiment of the present application, comprising the following steps:

[0040] S110, the defect detection equipment obtains the process level of the wafer under test and the grayscale characteristics of the surface of the wafer under test.

[0041] In one embodiment of the present application, the defect detection device may be Figure 1 The defect scanning machine used in the online defect monitoring site shown in the figure. The defect scanning machine can generate a grayscale image of the surface of the wafer under test. The location of the surface particle defect of the wafer will have grayscale abnormality, that is, the grayscale deviates from the normal value range, and is thus identified as a grayscale abnormal area. See Figure 3 It is also necessary to obtain the process level of the wafer under test.

[0042] S120, inputting the grayscale features and the process level into a surface defect recognition model to perform surface particle defect recognition.

[0043] First, a surface defect recognition model needs to be established. Known artificial intelligence (AI) or machine learning models and their establishment methods can be used. The surface defect recognition model of the present application distinguishes defect types by grayscale differences. Due to the different process conditions of each process level, the grayscale of the grayscale images obtained by the defect scanning machine at different process levels will also be different. Therefore, the grayscale abnormal areas of different process levels can be classified into different sets for training. In this way, the grayscale intervals corresponding to surface particle defects at different process levels are also different. Thus, an accurate surface defect recognition model is established to accurately identify surface particle defects in different process levels and effectively remove them in subsequent steps. For example, for the grayscale abnormal area of ​​the polysilicon (Poly) level, when its grayscale is the first grayscale interval, it will be identified as a surface particle defect by the surface defect recognition model; for the grayscale abnormal area of ​​the metal level, when its grayscale is the second grayscale interval, it will be identified as a surface particle defect by the surface defect recognition model; for the grayscale abnormal area of ​​the oxide layer level, when its grayscale is the third grayscale interval, it will be identified as a surface particle defect by the surface defect recognition model; and so on...

[0044] In step S120, the grayscale features including the grayscale and the process level at which the wafer under test is located when the grayscale features are obtained are input into the established surface defect recognition model to perform surface particle defect recognition.

[0045] S130, the particle removal mechanism moves to a position corresponding to the identified surface particle defect and performs a particle removal process.

[0046] In one embodiment of the present application, the particle removal mechanism includes an air sweep module. Figure 4 It is a schematic diagram of an exemplary wind sweep module for removing surface particle defects. There are surface particle defects on the surface of the wafer 10 under test. The surface particle defects are removed by moving the nozzle 20 of the wind sweep module to a position aligned with the surface particle defects and sweeping the surface particle defects.

[0047] In one embodiment of the present application, the wind sweep module includes a positioning unit. The positioning unit is used to synchronize with the coordinate information of the defect detection device, so that the nozzle 20 can be moved to a position facing the surface particle defect for targeted blowing. In one embodiment of the present application, the blowing angle of the wind sweep module is 45 degrees, that is, the wind direction is at an angle of 45 degrees to the wafer plane.

[0048] It is understandable that multiple surface particle defects may be identified on the surface of the wafer being tested. In step S130, the particle removal mechanism moves to the position corresponding to each identified surface particle defect in sequence, and performs particle removal processing at each position.

[0049] S140, detecting and determining whether the identified surface particle defects are removed by a defect detection device.

[0050] After step S130 is completed, the defect scanning machine of step S110 is used again to scan the wafer under test for defects, and determine whether the identified surface particle defects have been removed. Specifically, the determination can be made based on the grayscale image, that is, to confirm whether the grayscale of the grayscale abnormal area has returned to normal. Figure 5 These are comparison photos of particle defects on the wafer surface before and after wind scanning.

[0051] S150, the surface defect recognition model is re-learned according to the judgment result.

[0052] The removal result of the surface particle defect is fed back to the surface defect recognition model for re-learning to continuously optimize the recognition accuracy of the surface defect recognition model. Specifically, if the judgment result of step S140 is that the identified surface particle defect has not been removed, information indicating that the recognition result is wrong is fed back to the surface defect recognition model; if the judgment result of step S140 is that the identified surface particle defect has been removed, information indicating that the recognition result is correct is fed back to the surface defect recognition model.

[0053] In the above-mentioned surface particle defect detection and removal method, the surface defect recognition model recognizes surface particle defects according to grayscale features and process levels, can scientifically and effectively recognize surface particle defects in different process levels, and control the particle removal mechanism to remove the surface particle defects according to the position of the surface particle defects, and then judge whether the recognition of the surface particle defects is correct according to the result of the removal of the surface particle defects, and feed back to the surface defect recognition model for re-learning, so as to continuously optimize the recognition accuracy of the surface defect recognition model. Therefore, this method can improve the production efficiency of the product, improve the product yield, reduce the impact of rework on product circulation, and reduce the workload of process personnel.

[0054] In one embodiment of the present application, establishing a surface defect recognition model includes scanning a large number of products (e.g., 100 batches of wafer products) through a defect scanning machine at an online defect monitoring site, and establishing a surface defect recognition model based on grayscale images of the surfaces of these wafer products. In one embodiment of the present application, when establishing a surface defect recognition model, the grayscale abnormal areas on the surface of the wafer product are first classified according to their sizes. Figure 6, first, classify each grayscale abnormal area into defect sets of different size intervals according to the size of the grayscale abnormal area, and then train different sets separately. In one embodiment of the present application, grayscale abnormal areas with size characteristic values ​​less than 0.2 microns are classified into a set of small defects, grayscale abnormal areas with size characteristic values ​​≥ 0.2 microns and less than 1 micron are classified into a set of medium defects, and grayscale abnormal areas with size characteristic values ​​≥ 1 micron are classified into a set of large defects. In other embodiments, the size range of each set can also be divided according to the actual situation of the process, and more or fewer sets can be established according to the actual process conditions. Since sets of different sizes are trained separately, surface particle defects of different sizes will have different grayscale differences. See. Figure 7 For the aforementioned size characteristic value, the defect (grayscale abnormal area) can be surrounded by a square box, and the size characteristic value is the larger one of the length and width of the square box.

[0055] Accordingly, step S110 also needs to obtain the size of the grayscale abnormal region; in step S120, it is determined whether the grayscale abnormal region is a surface particle defect according to the size of the grayscale abnormal region and the grayscale characteristics of the grayscale abnormal region. That is, it is determined whether the grayscale abnormal region is a surface particle defect according to the process level, size and grayscale of the grayscale abnormal region.

[0056] In one embodiment of the present application, step S130 includes adjusting the wind pressure of the wind sweep module according to the size of the grayscale abnormal area, that is, for larger grayscale abnormal areas, stronger wind pressure is used for sweeping. In one embodiment of the present application, the typical wind pressure value is 0.7 megapascals.

[0057] It should be understood that, although the various steps in the flowchart of the present application are displayed in sequence according to the indication of the arrows, these steps are not necessarily executed in sequence in the order indicated by the arrows. Unless there is a clear description in this article, the execution of these steps is not strictly limited in order, and these steps can be executed in other orders. Moreover, at least a part of the steps in the flowchart of the present application may include multiple steps or multiple stages, and these steps or stages are not necessarily executed at the same time, but can be executed at different times, and the execution order of these steps or stages is not necessarily to be carried out in sequence, but can be executed in turn or alternately with other steps or at least a part of the steps or stages in other steps.

[0058] Those of ordinary skill in the art can understand that all or part of the processes in the above-mentioned embodiment methods can be completed by instructing the relevant hardware through a computer program, and the computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to memory, storage, database or other media used in the embodiments provided in this application can include at least one of non-volatile and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory or optical memory, etc. Volatile memory can include random access memory (RAM) or external cache memory. As an illustration and not limitation, RAM can be in various forms, such as static random access memory (SRAM) or dynamic random access memory (DRAM).

[0059] The present application also provides a computer-readable storage medium on which a computer program is stored. When the computer program is executed by a processor, the steps in the method described in any of the above embodiments are implemented.

[0060] The present application also provides a computer device, including a memory and a processor, wherein a computer program is stored in the memory, and when the processor executes the computer program, the steps in the method described in any of the above embodiments are implemented.

[0061] The present application also provides a computer program product, including a computer program, which implements the steps of the method described in any of the above embodiments when executed by a processor.

[0062] The present application correspondingly provides a surface particle defect detection and removal system, which includes a defect detection device, a particle removal mechanism and an artificial intelligence module, wherein the artificial intelligence module is configured with a surface defect recognition model; the defect detection device is used to obtain the process level at which the wafer under test is located, and to obtain the grayscale features of the surface of the wafer under test at the process level, and the artificial intelligence module is used to input the process level and the grayscale features into the surface defect recognition model to perform surface particle defect recognition; the particle removal mechanism is used to move to the position corresponding to the identified surface particle defect and perform particle removal processing; the artificial intelligence module is also used to detect the processed wafer under test through the defect detection device, determine whether the identified surface particle defect has been removed, and enable the surface defect recognition model to be re-learned according to the result of the judgment.

[0063] In one embodiment of the present application, the defect detection device obtains the grayscale features of the surface of the wafer under test, including generating a grayscale image of the surface of the wafer under test, and obtaining the size and grayscale features of the grayscale abnormal area according to the grayscale image. The artificial intelligence module performs surface particle defect recognition, including judging whether the grayscale abnormal area is a surface particle defect according to the process level, size and grayscale features of the grayscale abnormal area.

[0064] In one embodiment of the present application, the particle removal mechanism includes an air sweep module.

[0065] In one embodiment of the present application, the wind sweep module includes a positioning unit, and the positioning unit is used to synchronize with the coordinate information of the defect detection equipment.

[0066] In one embodiment of the present application, if the identified surface particle defect has not been removed, the artificial intelligence module will feedback information indicating that the recognition result is wrong to the surface defect recognition model; if the identified surface particle defect has been removed, the artificial intelligence module will feedback information indicating that the recognition result is correct to the surface defect recognition model.

[0067] In the description of this specification, the description with reference to the terms "some embodiments", "other embodiments", "ideal embodiments", etc. means that the specific features, structures, materials or characteristics described in conjunction with the embodiment or example are included in at least one embodiment or example of the present invention. In this specification, the schematic descriptions of the above terms do not necessarily refer to the same embodiment or example.

[0068] The technical features of the above-described embodiments may be arbitrarily combined. To make the description concise, not all possible combinations of the technical features in the above-described embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0069] The above-mentioned embodiments only express several implementation methods of the present invention, and the descriptions thereof are relatively specific and detailed, but they cannot be understood as limiting the scope of the invention patent. It should be pointed out that, for ordinary technicians in this field, several variations and improvements can be made without departing from the concept of the present invention, and these all belong to the protection scope of the present invention. Therefore, the protection scope of the patent of the present invention shall be subject to the attached claims.

Claims

1. A method for detecting and removing surface particle defects, comprising: The defect detection equipment obtains the process level at which the wafer under test is located, and obtains the grayscale characteristics of the surface of the wafer under test at the process level; Inputting the process level and the grayscale feature into a surface defect recognition model to perform surface particle defect recognition; The particle removal mechanism moves to a position corresponding to the identified surface particle defect and performs particle removal processing; The processed wafer to be tested is tested by the defect detection device to determine whether the identified surface particle defects are removed; The surface defect recognition model is relearned according to the result of the judgment.

2. The method for detecting and removing surface particle defects according to claim 1, characterized in that: It also includes the step of obtaining the size of the grayscale abnormal area; The step of identifying surface particle defects comprises: According to the process level, size and grayscale characteristics of the grayscale abnormal area, it is determined whether the grayscale abnormal area is a surface particle defect.

3. The method for detecting and removing surface particle defects according to claim 2, characterized in that: The surface defect recognition model is configured to classify grayscale abnormal regions of different sizes into sets of different size intervals for training.

4. The method for detecting and removing surface particle defects according to claim 1, characterized in that: The surface defect recognition model is configured to classify grayscale abnormal areas of different process levels into different sets for training.

5. The method for detecting and removing surface particle defects according to claim 1, characterized in that: The particle removal mechanism includes an air sweep module.

6. The method for detecting and removing surface particle defects according to claim 5, characterized in that: The wind sweep module includes a positioning unit, and the positioning unit is used to synchronize with the coordinate information of the defect detection equipment.

7. The method for detecting and removing surface particle defects according to claim 1, characterized in that: The step of the defect detection device acquiring the process level at which the wafer under test is located, and acquiring the grayscale characteristics of the surface of the wafer under test at the process level includes: generating a grayscale image of the surface of the wafer under test.

8. The method for detecting and removing surface particle defects according to claim 1, characterized in that: The step of relearning according to the result of the judgment comprises: If the identified surface particle defect is not removed, feeding back information indicating that the identification result is wrong to the surface defect identification model; If the identified surface particle defects have been removed, information indicating that the identification result is correct is fed back to the surface defect identification model.

9. A system for detecting and removing surface particle defects, characterized in that: It comprises a defect detection device, a particle removal mechanism and an artificial intelligence module, wherein the artificial intelligence module is configured with a surface defect recognition model; the defect detection device is used to obtain the process level at which the wafer under test is located, and obtain the grayscale features of the surface of the wafer under test at the process level, and the artificial intelligence module is used to input the process level and the grayscale features into the surface defect recognition model to perform surface particle defect recognition; the particle removal mechanism is used to move to the position corresponding to the recognized surface particle defect and perform particle removal processing; The artificial intelligence module is also used to detect the processed wafer under test through the defect detection equipment, determine whether the identified surface particle defects are removed, and re-learn the surface defect recognition model according to the judgment result.

10. A computer-readable storage medium having a computer program stored thereon, wherein the computer program, when executed by a processor, implements the steps in the method according to any one of claims 1 to 8.

Images