Methods for detecting and processing the condition of graphite trays, semiconductor process equipment and media

Abstract

This application discloses a method for detecting and processing the condition of a graphite tray, semiconductor process equipment, and a dielectric. The method includes: acquiring a set of three-dimensional data of multiple detection points on the surface of the graphite tray in a three-dimensional polar coordinate system, the three-dimensional data including the height values ​​of the deposits; filtering the three-dimensional data in the set of three-dimensional data based on the height values ​​to obtain a set of singular points; clustering the three-dimensional data of each singular point in the set of singular points to obtain multiple subsets of singular points; determining the protrusion type of each subset of singular points, including sharp peaks or gentle slopes; and determining whether the graphite tray needs cleaning based on the protrusion type of the multiple sets of singular points. This application, while ensuring a reduction in the number of defects on the wafer surface, avoids premature replacement of the graphite tray, extends the service life of the graphite tray, and reduces production costs. Furthermore, timely cleaning of the graphite tray also improves the stability of wafer doping concentration between process cycles.

Description

Technical Field

[0001] This application belongs to the field of semiconductor technology, and in particular relates to a method for detecting and processing the state of a graphite tray, semiconductor process equipment, and a medium. Background Technology

[0002] In the silicon carbide epitaxial growth process, due to the high process temperature, the silicon carbide wafer must be placed in a high-temperature resistant graphite tray and then placed together into the process reaction chamber (hereinafter referred to as the process chamber) for epitaxial growth. For example... Figure 1 As shown, the process includes a graphite tray 1, a gap 2, a silicon carbide wafer 3, and a silicon carbide wafer positioning edge 4. After the process gas is injected into the high-temperature process chamber, it not only grows an epitaxial layer on the surface of the silicon carbide wafer 3, but also deposits on the surface of the process chamber and the graphite tray 1. This results in the continuous accumulation of different types of silicon carbide particles on the surface of the graphite tray 1 during the process. With the increase of the number of process cycles, the accumulation of particles on the surface of the graphite tray gradually increases. Some of the accumulated particles form unstable protrusions, which pulverize and fall off in subsequent processes, thus increasing the defects on the wafer surface. In addition, a certain amount of dopant gas is also present in the silicon carbide particles accumulated on the surface of the graphite tray during the process, which causes unstable changes in the wafer doping concentration during the process, affecting the uniformity of wafer doping concentration between different process cycles (i.e., furnace cycles). Therefore, both the process chamber and the graphite tray need to be regularly polished and cleaned to remove abnormally accumulated silicon carbide particles on the surface, and in severe cases, they need to be replaced.

[0003] To detect the degree of silicon carbide particle accumulation on the surface of graphite trays, the industry commonly uses software to estimate the cumulative growth film thickness of the graphite tray based on process gas flow rate and process time. The tray's condition is then determined empirically, and a new tray is replaced when it reaches a preset cumulative growth film thickness threshold. However, this method has significant uncertainties during detection. To ensure the tray's condition does not affect the process results, it is usually replaced prematurely, leading to higher production costs and unstable wafer doping concentrations between process cycles. Summary of the Invention

[0004] The purpose of this application is to provide a method for detecting and processing the state of a graphite tray, a semiconductor process equipment, and a storage medium, in order to solve the problems in the related art, such as inaccurate detection of the degree of silicon carbide particle accumulation on the surface of the graphite tray, which leads to high production costs due to premature replacement of the graphite tray, and unstable wafer doping concentration between process cycles.

[0005] To achieve the above objectives, the embodiments of this application adopt the following technical solutions: In a first aspect, embodiments of this application provide a method for detecting and processing the state of a graphite tray, comprising: acquiring a set of three-dimensional data of multiple detection points on the surface of the graphite tray in a three-dimensional polar coordinate system, wherein the three-dimensional data includes the height value of the current accumulation on the graphite tray; filtering the three-dimensional data in the set of three-dimensional data according to the height value to obtain a set of singular points; clustering the three-dimensional data of each singular point in the set of singular points to obtain multiple sets of singular points; determining the protrusion type of each set of singular points, wherein the protrusion type includes a peak-shaped protrusion or a gentle slope-shaped protrusion; and determining whether the graphite tray needs to be cleaned according to the protrusion type corresponding to the multiple sets of singular points.

[0006] Secondly, embodiments of this application provide a semiconductor process apparatus, including: a graphite tray, a three-dimensional scanning device disposed above the graphite tray, and a processing device connected to the three-dimensional scanning device. The three-dimensional scanning device is used to acquire three-dimensional data of multiple detection points on the surface of the graphite tray in a three-dimensional polar coordinate system by scanning. The processing device is used to implement the steps of the graphite tray state detection and processing method as described in the first aspect of this application based on the set of the three-dimensional data.

[0007] Thirdly, embodiments of this application provide a readable storage medium storing a program or instructions, which, when executed by a processor, implement the steps of the graphite tray state detection and processing method as described in the first aspect of this application.

[0008] The above-described technical solutions adopted in the embodiments of this application can achieve the following beneficial effects: In this embodiment of the application, when detecting the condition of a graphite tray, a set of three-dimensional data of multiple detection points on the surface of the graphite tray in a three-dimensional polar coordinate system is acquired. Based on the height value of the current accumulation of material on the graphite tray in the three-dimensional data, the three-dimensional data in the set of three-dimensional data is filtered to obtain a set of singular points. The three-dimensional data of each singular point in the set of singular points are clustered to obtain multiple subsets of singular points. The protrusion type of each subset of singular points is determined, including sharp peaks or gentle slopes. Based on the protrusion type corresponding to the multiple subsets of singular points, it is determined whether the graphite tray needs cleaning. This embodiment of the application, by acquiring a set of three-dimensional data of multiple detection points on the surface of the graphite tray in a three-dimensional polar coordinate system and performing a series of operations such as filtering, clustering, and protrusion type determination on the set of three-dimensional data, can achieve accurate detection of the degree of particle accumulation on the surface of the graphite tray and whether cleaning is necessary. This avoids premature replacement of the graphite tray while ensuring a reduction in the number of defects on the wafer surface, extending the service life of the graphite tray, and reducing production costs. In addition, timely cleaning of the graphite tray also improves the stability of wafer doping concentration between process cycles. Attached Figure Description

[0009] The accompanying drawings, which are included to provide a further understanding of this application and form part of this application, illustrate exemplary embodiments and are used to explain this application, but do not constitute an undue limitation of this application. In the drawings: Figure 1 This is a schematic diagram of a graphite tray status detection device in related technologies; Figure 2 A flowchart illustrating a method for detecting and processing the state of a graphite tray, provided as an embodiment of this application; Figure 3 A front view of a process chamber provided for one embodiment of this application; Figure 4 A top view of a process chamber provided for one embodiment of this application; Figure 5 A flowchart illustrating a method for detecting and processing the state of a graphite tray, provided as another embodiment of this application; Figure 6 A flowchart illustrating a method for detecting and processing the state of a graphite tray, provided as another embodiment of this application; Figure 7 A flowchart illustrating a method for detecting and processing the state of a graphite tray, provided as another embodiment of this application; Figure 8 This is a schematic diagram of the structure of a semiconductor process apparatus provided for another embodiment of this application. Detailed Implementation

[0010] To make the objectives, technical solutions, and advantages of this application clearer, the technical solutions of this application will be clearly and completely described below in conjunction with specific embodiments and corresponding drawings. Obviously, the described embodiments are only a part of the embodiments of this application, and not all of them. Based on the embodiments in this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0011] The terms "first," "second," etc., used in this application are used to distinguish similar objects and not to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that embodiments of this application can be implemented in orders other than those illustrated or described herein. Furthermore, "and / or" in this application indicates at least one of the connected objects, and the character " / " generally indicates that the preceding and following objects are in an "or" relationship. It should be noted that all data involved in this application was obtained with the user's authorization.

[0012] Epitaxial growth refers to the process of growing a specific monolithic thin film on a wafer using epitaxial technology. The substrate wafer and the epitaxial thin film are collectively referred to as an epitaxial wafer. The epitaxial layer growth of silicon carbide wafers mainly adopts the chemical vapor deposition (CVD) method, using SiH4 or TCS as the silicon (Si) source and C2H4 or C3H8 as the carbon (C) source. Under high temperature and reduced pressure conditions, the Si and C atoms generated by the pyrolysis reaction recombine on the wafer surface to form a silicon carbide epitaxial layer.

[0013] In the silicon carbide epitaxial growth process, due to the high process temperature, silicon carbide wafers must be placed in high-temperature resistant graphite trays and placed together in the process module (PM) for epitaxial growth. After the process gas is injected into the high-temperature process module, it not only grows an epitaxial layer on the surface of the silicon carbide wafer, but also deposits on the surface of the process module and the graphite tray. This results in the continuous accumulation of different types of silicon carbide particles on the surface of the graphite tray during the process. With the increase of process cycles, the accumulation of particles on the surface of the graphite tray gradually increases. Some of these particles accumulate to form unstable protrusions, which pulverize and detach during subsequent processes, thus increasing wafer surface defects. In addition, a certain amount of dopant gas is also present in the silicon carbide particles accumulated on the surface of the graphite tray during the process, causing unstable changes in the wafer doping concentration and affecting the uniformity of wafer doping concentration between different process cycles (i.e., furnace cycles). Therefore, both the process module and the graphite tray need to be cleaned regularly to remove abnormally accumulated silicon carbide particles, and in severe cases, they need to be replaced.

[0014] To detect the degree of silicon carbide particle accumulation on the surface of graphite trays, the current industry practice is to estimate the cumulative growth film thickness of the graphite tray using software based on process gas flow rate and process time, and then determine the state of the graphite tray based on experience. A new graphite tray is replaced when it reaches a preset cumulative growth film thickness threshold. However, this method has significant uncertainties during detection. To ensure that the graphite tray state does not affect the process results, it is usually replaced prematurely, resulting in high production costs and unstable wafer doping concentrations between process cycles. Therefore, this application proposes a method for detecting and processing the state of graphite trays, semiconductor process equipment, and a storage medium to solve the problems of inaccurate detection of particle accumulation on the surface of graphite trays, leading to high production costs due to premature replacement of graphite trays, and unstable wafer doping concentrations between process cycles in related technologies.

[0015] The technical solutions provided by the various embodiments of this application are described in detail below with reference to the accompanying drawings.

[0016] Figure 2 This is a schematic flowchart illustrating a method for detecting and processing the state of a graphite tray, provided as an embodiment of this application. Figure 2 As shown, the graphite tray state detection and processing method of this application embodiment may specifically include the following steps: S201, acquire a set of three-dimensional data of multiple detection points on the surface of the graphite tray in a three-dimensional polar coordinate system. The three-dimensional data includes the height value of the current stack of materials on the graphite tray.

[0017] In this embodiment of the application, the execution subject of the graphite tray state detection and processing method is a graphite tray state detection and processing device, which can be disposed in the process chamber of semiconductor process equipment.

[0018] like Figure 3 As shown, the process chamber includes: a graphite tray 1 for supporting the wafer; a rotating base 2 for supporting the graphite tray 1 and driving the graphite tray 1 to rotate; a three-dimensional scanning device 3 for acquiring three-dimensional data of the graphite tray surface, corresponding to the three-dimensional scanning range 6; a chamber body 4; a quartz observation window 5; an outer circle of the graphite tray 7; a gap 8 between the graphite tray 1 and the wafer; a silicon carbide wafer 9; and a rotary motor 10 for driving the rotating base 2 to rotate. Figure 4 for Figure 3 The top view of the process chamber shown.

[0019] A graphite tray is placed on a rotating base driven by a rotary motor, with the center of the graphite tray aligned with the center of the rotating base. The rotating base drives the graphite tray to rotate at a set speed v. A 3D scanning device is placed above the graphite tray. This 3D scanning device may include, but is not limited to, an optical sectioning 3D laser measuring instrument, a white light interferometric coaxial 3D measuring instrument, a line scanning 3D imaging device, and a 3D machine vision system.

[0020] A 3D scanning device performs a 3D scan on the surface of a rotating graphite tray, acquiring 3D data of multiple detection points on the surface of the graphite tray in a 3D polar coordinate system. These multiple 3D data points constitute a set of 3D data, for example, a set of 3D data corresponding to N detection points. K { θ,r,h The three-dimensional data includes detection points. K n Angle value at the circumference of the graphite tray i Testing point K n Distance from the center of the graphite tray r The height of the current stack after a single process on the graphite pallet h The height of the accumulation is the thickness of the current accumulation after a single process. It is the sum of the height of the accumulation before the single process and the height increment of the accumulation after the single process. The height increment of the accumulation is the absolute difference between the height of the accumulation before and after the single process.

[0021] It should be noted that the 3D scanning device will perform a 3D scan after the graphite tray enters a constant speed rotation process. The 3D scan ends after the graphite tray has rotated one full circle at a constant speed, and the acquired 3D data is recorded. The detection process only requires scanning and recording the 3D data of one full circle.

[0022] S202, based on the height value, the three-dimensional data in the set of three-dimensional data are filtered to obtain the set of singular points.

[0023] In this embodiment, multiple detection points need to be identified to determine whether each detection point is an outlier or a normal point. Outliers are detection points where particulate matter accumulates abnormally, while normal points are detection points where particulate matter accumulates normally. That is, the set of three-dimensional data from multiple detection points... K { θ,r,h} is divided into a set of singular points K a { θ,r,h} and regular point set K b { θ,r, h Specifically, this can be based on a set of three-dimensional data. K { θ,r,h The height value of each detection point in the set of 3D data. K { θ,r,h By filtering multiple 3D data points in the dataset, a set of singularities is obtained. K a { θ,r,h} and regular point set K b { θ,r,h}

[0024] Since the entire graphite tray contains a large amount of 3D data, the 3D data of singular points needs to be processed to identify and remove the 3D data of regular points from the overall 3D data set. This facilitates the subsequent clustering and partitioning of the 3D data of singular points and saves computing resources.

[0025] S203, cluster the 3D data of each singular point in the singular point set to obtain multiple singular point subsets.

[0026] In this embodiment of the application, based on the set of singularities K a { θ,r,h The three-dimensional data of each singular point in the singular point set. K a { θ,r,h The singular points in the list are clustered based on their location to form a set of singular points. K a { θ,r,h} is divided into multiple sets of singular points K c { θ,r,h}

[0027] S204, determine the bulge type of each singular point set, including peak-shaped bulges or gentle slope-shaped bulges.

[0028] In this embodiment of the application, for each set of singular points K c { θ,r,h}, determine whether the bulge type of the singular point set is a sharp peak or a gentle slope, that is, determine whether the bulges within the singular point set are sharp peaks or gentle slopes, for each singular point set. K c { θ,r,h} corresponds to a convex type.

[0029] S205, determine whether the graphite tray needs to be cleaned based on the protrusion type corresponding to multiple sets of singularities.

[0030] In this embodiment of the application, as a first feasible implementation, it is possible to determine whether the graphite tray needs to be cleaned based on the proportion of the subset of multiple singular point subsets in the multiple subsets of protrusions of the peak type. For example, if the proportion exceeds a preset proportion threshold, it is determined that the graphite tray needs to be cleaned; if the proportion does not exceed the preset proportion threshold, it is determined that the graphite tray does not need to be cleaned.

[0031] As a second feasible implementation, the number of subsets of singular point subsets with peak-shaped protrusions can be used as a basis. T Determine if the graphite tray needs cleaning, for example, if the quantity... T Exceeding the preset second quantity threshold T m2 If the quantity is high, then the graphite tray needs to be cleaned. T The second preset quantity threshold was not exceeded. T m2 If so, then the graphite tray does not need to be cleaned.

[0032] It should be noted here that when it is determined that the graphite tray needs cleaning, the graphite tray rotation stops, and a prompt message is output to remind the user that the graphite tray needs cleaning. This prompt message may also include the locations in the graphite tray that need cleaning, that is, the locations of each singular point in the set of singular points with the protrusion type of spike-shaped protrusions (i.e., the planar polar coordinate data in the 3D data { θ,r (The corresponding position).

[0033] In summary, the graphite tray status detection and processing method of this application embodiment, when detecting the graphite tray status, acquires a set of three-dimensional data of multiple detection points on the graphite tray surface in a three-dimensional polar coordinate system, filters the three-dimensional data in the set of three-dimensional data according to the height value of the current accumulation of the graphite tray in the three-dimensional data to obtain a set of singular points, clusters the three-dimensional data of each singular point in the set of singular points to obtain multiple singular point subsets, determines the protrusion type of each singular point subset, the protrusion type includes sharp peak protrusions or gentle slope protrusions, and determines whether the graphite tray needs to be cleaned according to the protrusion type corresponding to the multiple singular point sets. This application embodiment acquires a set of three-dimensional data of multiple detection points on the surface of a graphite tray in a three-dimensional polar coordinate system, and performs a series of operations such as screening, clustering, and protrusion type determination on the set of three-dimensional data. This enables accurate detection of the degree of particle accumulation on the surface of the graphite tray and whether cleaning is required. Thus, while ensuring a reduction in the number of defects on the wafer surface, premature replacement of the graphite tray is avoided, extending the service life of the graphite tray and reducing production costs. In addition, timely cleaning of the graphite tray also improves the stability of wafer doping concentration between process steps.

[0034] Figure 5 This is a flowchart illustrating a method for detecting and processing the state of a graphite tray, provided as another embodiment of this application. Figure 5 As shown, in Figure 2 Based on the illustrated embodiment, the graphite tray state detection and processing method of this application embodiment may specifically include the following steps: S501, acquire a set of three-dimensional data of multiple detection points on the surface of the graphite tray in a three-dimensional polar coordinate system. The three-dimensional data includes the height value of the current stack of materials on the graphite tray.

[0035] In this embodiment, step S501 is the same as step S201 in the above embodiment, and will not be repeated here.

[0036] Step S202 in the above embodiment, "filtering the three-dimensional data in the set of three-dimensional data according to the height value to obtain the set of singular points," may specifically include the following step S502: S502, divides 3D data whose height values ​​are outside the preset height range into a set of singularities.

[0037] In this embodiment of the application, a height range can be preset, for example, set to (h1, h2), then if the detection point K i height value h i Located outside the preset height range, i.e. h i h i If the value is greater than h2, then the detection point will be...​K i Partition to the set of singularities K a Correspondingly, if the detection point K i height value h i Located within the preset height range, i.e., h1≤ h i If h2 ≤ h2, then the detection point will be... K i Divide into regular point set K b .

[0038] The height range can be determined by the following steps: calculate the mean and standard deviation of all height values ​​in the set of 3D data; determine the height range based on the mean and standard deviation of the height values.

[0039] Specifically, the mean of all height values ​​in a set of 3D data can be calculated using the following formula. m h :

[0040] in, N The total number of detection points in the set of 3D data. h i For the first i The height value in the three-dimensional data of each detection point.

[0041] The standard deviation of all height values ​​in a set of 3D data can be calculated using the following formula. s h :

[0042] The height range can be preset to ( m h -2 s h , m h +2 s h If the detection point K i height value h i Located outside the preset height range, i.e. h i < m h -2 s h or h i > mh +2 s h Then the detection point K i Partition to the set of singularities K a Correspondingly, if the detection point K i height value h i Located within the preset height range, i.e. m h -2 s h ≤ h i ≤ m h +2 s h Then the detection point K i Divide into regular point set K b .

[0043]

[0044] S503 clusters the 3D data of each singular point in the singular point set to obtain multiple singular point subsets.

[0045] S504, determine the bulge type of each set of singular points, including peak-shaped bulges or gentle slope-shaped bulges.

[0046] S505, determine whether the graphite tray needs cleaning based on the protrusion type corresponding to multiple sets of singularities.

[0047] In this embodiment, steps S503-S505 are the same as steps S203-S205 in the above embodiment, and will not be repeated here.

[0048] Furthermore, such as Figure 6 As shown, step S503 in the above embodiment, "clustering the 3D data of each singular point in the singular point set to obtain multiple singular point subsets," may specifically include the following steps: S601 converts the planar polar coordinate data in each three-dimensional data of the singular point set into planar rectangular coordinate data in the planar rectangular coordinate system to obtain the planar singular point set.

[0049] In this embodiment of the application, the singularity set is... K a { θ,r,h The planar polar coordinate data in each of the three-dimensional data in} θ,rThe following formula can be used to convert the data into Cartesian coordinates in a Cartesian coordinate system: x,y}, thus obtaining the set of singular points in the plane. M { x,y}.in, M { x,y In} M { i} and singularity set K a { θ,r,h In} K a { i One-to-one correspondence.

[0050]

[0051] S602, cluster the Cartesian coordinate data of each singular point in the set of plane singular points to obtain multiple plane singular point subsets.

[0052] In this embodiment, multiple planar singularity subsets can be obtained through the following steps: any unclustered singularity in the planar singularity subset is taken as the central singularity of the current planar singularity subset; among the unclustered singularities in the planar singularity subset, it is searched to see if there are singularities that satisfy the clustering conditions, including being located within a circle centered at the central singularity with a preset radius threshold; if not, the clustering of the current planar singularity subset ends; if it exists, the singularity that satisfies the clustering conditions is added to the current planar singularity subset, and the singularity that satisfies the clustering conditions is taken as the new central singularity.

[0053] Specifically, the set of singular points in a plane can be... M { x,y Any singular point that is not clustered in} M 0{ x 0 ,y 0} is the set of singular points in the current plane. W Find the central singularity of 0 and traverse the set. M { x,y}, search for the existence of singularities among the unclustered singularities. M i { x i ,y i}, located in M 0{ x 0 ,y Using 0 as the center and a preset radius threshold... f The area inside the circle with radius is considered. If a singularity exists... M i { xi ,y i}, then the singular point M i { x i ,y i Add to the current set of singular points in the plane. W In 0, and in the set M { x,y Continue searching for singularities within the range. M i+1 { x i+1 ,y i+1}, located in M i { x i ,y i Using} as the center and the radius threshold f Within a circle with radius , continue searching until no singular point matching the conditions is found.

[0054] Repeat the above steps until a set is reached. M { x,y All singular points in the array are moved to a subset of singular points in the plane. W 0、 W 1. W 2, ... W n until.

[0055] It should be noted here that the radius threshold f It can be set as a singularity to be clustered. M i { x i ,y i The corresponding height value h i The absolute value, because if the singular point M i { x i ,y i If there is an abnormal protrusion at}, then there should be a similar protruding point within its spherical range.

[0056] S603, based on multiple planar singular point subsets, divide the singular point set into multiple singular point subsets.

[0057] In this application embodiment, as a first feasible implementation, it can be based on each set of plane singular points. Wand singularity set K a The correspondence between elements in the set of singularities. K a The corresponding division is into multiple sets of singular points. K c .

[0058] As a second feasible implementation, it can be based on each set of plane singular points. W and singularity set K a The correspondence between elements in the set of singularities. K a Divided into multiple candidate singular point sets K w For multiple candidate singularity sets K w By merging these elements, we obtain multiple sets of singular ideas. K c .

[0059] Specifically, since multiple consecutive protrusions are not easily broken, multiple candidate singularity sets can be considered during data clustering. K w Merging is performed to reduce the number of singularity sets, thereby reducing the number of cleanups. For example, if the number of singularity sets with spike-shaped protrusions is greater than or equal to 10, the graphite tray needs to be cleaned. If the number of singularity sets is 10 before merging, and 9 after merging, then the 9 sets do not need to be cleaned, reducing the number of cleanups.

[0060] Among them, multiple candidate singular idea sets K w The merging process can be achieved through the following steps: calculate the three-dimensional coordinates of the center point of the candidate singular point subset; calculate the distance between the center points of any two candidate singular point subsets based on the three-dimensional coordinates of the center point; if the distance is less than a preset distance threshold, then merge the two candidate singular point subsets.

[0061] Specifically, calculate the set of candidate singular points. K w Any point in the middle and other points K i distance d and the point and set K w Sum of squared distances to all points within the area D Sum of squared distances D The smallest point K 0 as a set K w The center point, corresponding coordinatesα 0{ i 0 ,r 0 ,h 0}. Distance d and the sum of squared distances D The calculation formula is as follows: ：

[0062] Calculate any two candidate singular point sets K w and K w+1 center point coordinates α w and α w+1 distance γ w If the distance γ w If the distance is less than the preset distance threshold γ, then the set will be... K w and K w+1 Merge until any two candidate singular sets are reached. K w and K w+1 Multiple sets of singular points are obtained until the distance between the center points is greater than the distance threshold γ. K c The distance threshold γ can be set as follows: Two candidate singularity sets are obtained respectively. K w and K w+1 Height of the highest point inside h w and h p Set the distance threshold γ= h w + h p .

[0063] In some embodiments, before step S504 "determine the protrusion type of each singular point subset", the following steps may be included: determining whether the singular point subset is a set of detected noise points; if not, then performing the step of determining the protrusion type of each singular point subset; if so, then not processing the singular point subset.

[0064] Specifically, if the set of singular ideas K c The number of internal singularities is less than a preset first threshold. T m1 Then the set of singular points is determined. K cTo detect a set of noise points, no processing is performed. If the set of singular points... K c The number of internal singularities is equal to or greater than a preset first threshold. T m1 Then the set of singular points is determined. K c If the set of noise points is not detected, proceed to step S504, "Determine the bulge type of each singular point subset".

[0065] It should be noted here that, because noise points typically appear as a single peak value in data sampling, affecting nearby individual points, they often manifest as clusters of a few isolated points during clustering. This differs significantly from the actual trend of particulate matter accumulation data. Therefore, in general, the first quantity threshold is... T m1 Set it to 5 or below.

[0066] Furthermore, such as Figure 7 As shown, step S504, "determining the bulge type of each singular point set," may specifically include the following steps: S701, determine the concave / convex point type of the singular points in each singular point subset, including concave points and convex points.

[0067] In this embodiment of the application, the singularity is determined to be concave or convex based on the three-dimensional data of each singularity in the singularity subset.

[0068] As a feasible implementation method, the type of concavity / convexity of a singularity can be determined through the following steps: Traversing the set of strange ideas K c { θ,r,h}, a collection of unusual ideas K c { θ,r,h Inner height value h The largest singularity K m { i m ,r m ,h m} is identified as the highest singularity; the highest singularity is calculated. K m { i m ,r m ,h m Height value h m and a collection of strange ideasK c { θ,r,h The height value of any singular point within} h height difference h m -h If the height difference h m -h Less than the highest singularity K m { i m , r m ,h m The planar projection distance between} and any singular point L If the height difference is 0, then the singularity is determined to be a concave point; if the height difference is 0, then the singularity is determined to be a concave point. h m -h Equal to or greater than the highest singularity K m { i m ,r m ,h m The planar projection distance between} and any singular point L If so, the singularity is determined to be a convex point. The planar projection spacing is... L The calculation formula is as follows:

[0069] S702, if the number of singular points of type concave is less than the number of singular points of type convex within the singular point subset, and the maximum height value within the singular point subset is greater than a preset height threshold, then the convexity type of the singular point subset is determined to be a peak-shaped convexity.

[0070] In this embodiment of the application, if the set of singular points K c { θ,r,h Within a given set, the number of concave points is less than the number of convex points, and the set of singular points is... K c { θ,r,h Maximum height value within} h m Greater than the preset height threshold h max Then the set of singular points is determined. K c { θ,r,h Within the area, there are sharp, pointed protrusions.

[0071] S703, if the number of singular points of type concave within the singular point subset is equal to or greater than the number of singular points of type convex, or if the maximum height value within the singular point subset is equal to or less than the height threshold, then the convexity type of the singular point subset is determined to be a gentle slope convexity.

[0072] In this embodiment of the application, if the set of singular points K c { θ,r,h Within a given set, the number of concave points is equal to or greater than the number of convex points, or, the set of singular points. K c { θ,r,h Maximum height value within} h m Equal to or less than the preset height threshold h max Then the set of singular points is determined. K c { θ,r,h The area within the range is characterized by gently sloping protrusions.

[0073] Furthermore, since the silicon carbide particles deposited on the inner wall of the chamber during the process also contain a certain amount of doped gas, which can cause unstable changes in the wafer doping concentration during the process, affecting the stability (i.e., consistency) of the wafer doping concentration between different process steps, the graphite tray state detection and processing method of this application embodiment may further include the following steps to determine whether the process chamber needs to be processed: Figure 3 It also includes a machine vision light source 12 for providing back illumination for the graphite tray 1; a machine vision camera 13 for acquiring images of the bottom of the graphite tray 1 and identifying the serial number (SN) code to distinguish different graphite trays 1, corresponding to the machine vision detection range 11. For the same graphite tray, it includes a collection of its three-dimensional data. K { θ,r,h The 3D data in the dataset is filtered to obtain a set of regular points. K b { θ,r,h}, set of regular points K b { θ,r,h} is the set of singular points K a { θ,r,h In a collection of three-dimensional data K { θ,r,h The complement of}; based on the regular set of points K b { θ,r,h The average height of each regular point in} h b Determine the thickness of the current accumulation on the graphite tray. h0, where the thickness value is... h 0 and the mean of height values h b A positive correlation is established; based on the thickness values ​​of the graphite tray deposits before and after a single process, the thickness increment Δ of the graphite tray deposits in a single process is determined. h (Since the graphite trays are replaced between two cleanings of the process chamber (e.g., the graphite tray is cleaned once every 10 process cleanings, and the process chamber is cleaned once every 30 process cleanings), and the graphite trays are partially ground and cleaned during actual use, it is necessary to determine the thickness increment of the deposit for each graphite tray in a single process cleaning); The thickness increments of the deposits for each graphite tray used after the last cleaning of the process chamber are summed to obtain the current thickness of the deposit in the process chamber. H If the current thickness of the deposit in the process chamber H Exceeding the preset thickness threshold H m If this is detected, the process chamber needs cleaning, and a prompt message will be output to remind the user to clean the process chamber. The formula for calculating thickness H is as follows:

[0074] For example, after the process chamber was cleaned last time, graphite tray 1 and graphite tray 2 were used. Graphite tray 1 was used for two processes. The thickness of the deposit before the first process was 0 micrometers, and the thickness of the deposit after the first process was 10 micrometers. Therefore, the thickness increment of the deposit on graphite tray 1 after the first process was 10 micrometers. Suppose that after the first process, the sharp protrusions on graphite tray 1 were locally ground and cleaned, so that the thickness of the deposit on graphite tray 1 became 8 micrometers. Then, graphite tray 1 was used for the second process. The thickness of the deposit before the second process was 8 micrometers. Suppose that the thickness of the deposit after the second process was calculated to be 18 micrometers. Therefore, the thickness increment of the deposit on graphite tray 1 after the second process was 10 micrometers. At this time, the thickness of the deposit in the process chamber is the sum of the thickness increments of the deposits on graphite tray 1 in the two processes, that is, 10 + 10 = 20 micrometers. Assuming that graphite tray 2 has been used for three processes, and the thickness increments of the deposits in the three processes are 11, 12, and 11 respectively, then the thickness of the deposits in the process chamber is the sum of the thickness increments of the deposits from the two processes of graphite tray 1 and the three processes of graphite tray 2, which is 10 + 10 + 11 + 12 + 11 = 54 micrometers.

[0075] Furthermore, the method for detecting and processing the state of the graphite tray in this application embodiment may further include the following step of adjusting the flow rate of the doping gas: Based on the current doped gas flow rate in the process L m and the thickness of the process chamber deposits after the current process Hn-1 Calculate the adjustment value Δ for the doped gas flow rate in the next process. L m ; Adjustment value Δ based on the doped gas flow rate L m Determine the flow rate of the doped gas in the next process. L m .

[0076] Specifically, record the wafer doping concentration after each process. U n The flow rate of doped gas during the process L m Obtain the change in wafer doping concentration Δ U n With the thickness of particulate matter accumulation in the chamber H n-1 The relationship curve between the two can be fitted to obtain the following reference formula:

[0077] in, G m The preset doping gas weights.

[0078] To ensure the stability of wafer doping concentration in each process, the desired change in wafer doping concentration Δ is... U n If =0, then the adjustment value Δ for the doped gas flow rate can be obtained as follows. L m The calculation formula is as follows:

[0079] Based on the current doped gas flow rate in the process L m Adjustment value △ for doped gas flow rate L m Determine the flow rate of the doped gas in the next process. L m = L m +△ L m Additionally, after each process is completed, new data can be recorded and the fitted curve can be further optimized to obtain a new reference formula.

[0080] It should be noted that the above curve fitting method can be implemented using a machine learning algorithm based on neural networks. By continuously accumulating data, the accuracy of the predicted adjustment value of the doping gas flow rate can be gradually improved, ensuring the stability of wafer doping concentration between different process steps.

[0081] In summary, the graphite tray status detection and processing method of this application embodiment acquires a set of three-dimensional data of multiple detection points on the graphite tray surface in a three-dimensional polar coordinate system, and performs a series of operations such as singular point screening, clustering, and protrusion type determination on the set of three-dimensional data. This enables precise detection of the degree of particle accumulation on the graphite tray surface and whether cleaning is required. This avoids premature replacement of the graphite tray while ensuring a reduction in the number of wafer surface defects, extending the lifespan of the graphite tray, and reducing production costs. Furthermore, timely cleaning of the graphite tray also improves the stability of wafer doping concentration between process cycles. By performing routine point screening on the set of three-dimensional data and calculating the current thickness of the deposits in the process chamber, precise detection of the degree of particle accumulation on the process chamber surface and whether cleaning is required can be achieved, ensuring the stability of wafer doping concentration between process cycles. Additionally, by adjusting the doping gas flow rate during each process, the stability of wafer doping concentration between process cycles can be further ensured.

[0082] This application also provides a semiconductor process apparatus. For example... Figure 8 As shown, the semiconductor process equipment 800 includes: a graphite tray 1, a three-dimensional scanning device 3 disposed above the graphite tray 1, and a processing device 801 connected to the three-dimensional scanning device 3. The three-dimensional scanning device 3 is used to acquire three-dimensional data of multiple detection points on the surface of the graphite tray 1 in a three-dimensional polar coordinate system by scanning. The processing device 801 is used to implement the steps of any of the above-described graphite tray state detection processing method embodiments based on the set of three-dimensional data.

[0083] The semiconductor process equipment of this application embodiment acquires a set of three-dimensional data of multiple detection points on the surface of a graphite tray in a three-dimensional polar coordinate system. By performing a series of operations on the set of three-dimensional data, such as singularity filtering, clustering, and protrusion type determination, it can accurately detect the degree of particle accumulation on the graphite tray surface and whether cleaning is necessary. This avoids premature replacement of the graphite tray while ensuring a reduction in the number of wafer surface defects, extending the lifespan of the graphite tray, and reducing production costs. Furthermore, timely cleaning of the graphite tray also improves the stability of wafer doping concentration between process cycles. By performing routine point filtering on the set of three-dimensional data and calculating the current thickness of the deposit in the process chamber, accurate detection of the degree of particle accumulation on the process chamber surface and whether cleaning is necessary can be achieved, ensuring the stability of wafer doping concentration between process cycles. In addition, by adjusting the doping gas flow rate during each process, the stability of wafer doping concentration between process cycles can be further ensured.

[0084] This application also proposes a readable storage medium storing one or more computer programs, the one or more computer programs including instructions. When the program or instructions are executed by a processor in a semiconductor process apparatus including multiple applications, the processor in the semiconductor process apparatus is able to execute the various processes of the above-described graphite tray state detection processing method embodiments, and specifically to execute the steps of any of the above-described graphite tray state detection processing method embodiments.

[0085] The readable storage medium of this application embodiment acquires a set of three-dimensional data of multiple detection points on the surface of a graphite tray in a three-dimensional polar coordinate system. By performing a series of operations on the set of three-dimensional data, such as singularity filtering, clustering, and protrusion type determination, it can accurately detect the degree of particle accumulation on the graphite tray surface and whether cleaning is necessary. This avoids premature replacement of the graphite tray while ensuring a reduction in the number of wafer surface defects, extending the lifespan of the graphite tray, and reducing production costs. Furthermore, timely cleaning of the graphite tray also improves the stability of wafer doping concentration between process cycles. By performing routine point filtering on the set of three-dimensional data and calculating the current thickness of the deposits in the process chamber, accurate detection of the degree of particle accumulation on the process chamber surface and whether cleaning is necessary can be achieved, ensuring the stability of wafer doping concentration between process cycles. In addition, by adjusting the doping gas flow rate during each process, the stability of wafer doping concentration between process cycles can be further ensured.

[0086] The systems, devices, modules, or units described in the above embodiments can be implemented by computer chips or entities, or by products with certain functions. A typical implementation device is a computer. Specifically, a computer can be, for example, a personal computer, laptop computer, cellular phone, camera phone, smartphone, personal digital assistant, media player, navigation device, email device, game console, tablet computer, wearable device, or any combination of these devices.

[0087] For ease of description, the above devices are described separately by function as various units. Of course, in implementing this application, the functions of each unit can be implemented in one or more software and / or hardware.

[0088] Those skilled in the art will understand that embodiments of this application can be provided as methods, systems, or computer program products. Therefore, this application can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, this application can take the form of a computer program product embodied on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0089] This application is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of this application. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart... Figure one One or more processes and / or boxes Figure one A device that provides the functions specified in one or more boxes.

[0090] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure one One or more processes and / or boxes Figure one The function specified in one or more boxes.

[0091] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure one One or more processes and / or boxes Figure one The steps of the function specified in one or more boxes.

[0092] In a typical configuration, a computing device includes one or more processors (CPU), input / output interfaces, network interfaces, and memory.

[0093] Memory may include non-persistent storage in computer-readable media, such as random access memory (RAM) and / or non-volatile memory, such as read-only memory (ROM) or flash RAM. Memory is an example of computer-readable media.

[0094] Computer-readable media includes both permanent and non-permanent, removable and non-removable media that can store information using any method or technology. Information can be computer-readable instructions, data structures, modules of programs, or other data. Examples of computer storage media include, but are not limited to, phase-change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technologies, CD-ROM, digital versatile optical disc (DVD) or other optical storage, magnetic tape, magnetic magnetic disk storage or other magnetic storage devices, or any other non-transferable medium that can be used to store information accessible by a computing device. As defined herein, computer-readable media does not include transient computer-readable media, such as modulated data signals and carrier waves.

[0095] It should also be noted that the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such process, method, article, or apparatus. Unless otherwise specified, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes that element.

[0096] This application can be described in the general context of computer-executable instructions, such as program modules, that are executed by a computer. Generally, program modules include routines, programs, objects, components, data structures, etc., that perform a specific task or implement a specific abstract data type. This application can also be practiced in distributed computing environments where tasks are performed by remote processing devices connected via a communication network. In distributed computing environments, program modules can reside in local and remote computer storage media, including storage devices.

[0097] The various embodiments in this specification are described in a progressive manner. Similar or identical parts between embodiments can be referred to mutually. Each embodiment focuses on describing the differences from other embodiments. In particular, the system embodiments are basically similar to the method embodiments, so the description is relatively simple; relevant parts can be referred to the descriptions in the method embodiments.

[0098] The above are merely embodiments of this application and are not intended to limit the scope of this application. Various modifications and variations can be made to this application by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this application should be included within the scope of the claims of this application.

Claims

1. A method for detecting and processing the state of a graphite tray, characterized in that, include: A set of three-dimensional data of multiple detection points on the surface of a graphite tray in a three-dimensional polar coordinate system is obtained, wherein the three-dimensional data includes the height value of the current accumulation of the graphite tray; Based on the height value, the three-dimensional data in the set of three-dimensional data is filtered to obtain a set of singular points; Cluster the three-dimensional data of each singular point in the singular point set to obtain multiple singular point subsets; Determine the protrusion type of each set of singular points, the protrusion type including sharp peaks or gentle slopes; Based on the protrusion type corresponding to the multiple sets of singularities, determine whether the graphite tray needs cleaning.

2. The method according to claim 1, characterized in that, The process of filtering the three-dimensional data in the set of three-dimensional data to obtain a set of singular points includes: The three-dimensional data whose height values ​​are outside the preset height range are assigned to the singularity set.

3. The method according to claim 2, characterized in that, Also includes: Calculate the mean and standard deviation of all height values ​​in the set of three-dimensional data; The height range is determined based on the mean and standard deviation of the height values.

4. The method according to claim 1, characterized in that, The three-dimensional data of each singular point in the singular point set are clustered to obtain multiple singular point subsets, including: The planar polar coordinate data in each of the three-dimensional data in the singular point set is converted into planar rectangular coordinate data in a planar rectangular coordinate system to obtain the planar singular point set. Clustering the Cartesian coordinate data of each singular point in the set of plane singular points yields multiple subsets of plane singular points; Based on the plurality of planar singular point subsets, the singular point set is divided into the plurality of singular point subsets.

5. The method according to claim 4, characterized in that, The planar rectangular coordinate data of each singular point in the set of planar singular points are clustered to obtain multiple subsets of planar singular points, including: Take any unclustered singular point in the set of plane singular points as the center singular point of the current subset of plane singular points; In the set of unclustered singularities in the plane, it is searched to see if there are any singularities that satisfy the clustering conditions. The clustering conditions include being located inside a circle with the central singularity as the center and a preset radius threshold as the radius. If it does not exist, then the clustering of the current set of singular points in the plane ends; If they exist, the singular points that satisfy the clustering conditions are added to the current set of singular points in the plane, and the singular points that satisfy the clustering conditions are used as the new central singular points.

6. The method according to claim 4, characterized in that, The step of dividing the singular point set into multiple singular point sets based on the multiple planar singular point subsets includes: Based on the plurality of planar singular point subsets, the singular point set is divided into a plurality of corresponding candidate singular point subsets; The multiple candidate singularity sets are merged to obtain the multiple singularity sets.

7. The method according to claim 6, characterized in that, The merging of the plurality of candidate singularity subsets to obtain the plurality of singularity subsets includes: Calculate the three-dimensional coordinates of the center point of the candidate singular point subset; Based on the three-dimensional coordinates of the center point, calculate the distance between the center points of any two candidate singular point subsets; If the distance is less than a preset distance threshold, the two candidate singularity sets are merged.

8. The method according to claim 1, characterized in that, Before determining the bulge type of each of the singular point sets, the method further includes: Determine whether the subset of singular points is a set of detected noise points; If not, then perform the step of determining the bulge type of each of the singular point sets.

9. The method according to claim 8, characterized in that, The step of determining whether the subset of singular points is a set of detected noise points includes: If the number of singular points in the singular point subset is less than a preset first number threshold, then the singular point subset is determined to be a set of detected noise points; If the number of singular points in the singular point subset is equal to or greater than the first quantity threshold, then the singular point subset is determined not to be a set of detected noise points.

10. The method according to claim 1, characterized in that, Determining the bulge type of each of the singular point sets includes: Determine the concave / convex point type of the singular points within each set of singular points, wherein the concave / convex point type includes concave points and convex points; If the number of singular points of type concave is less than the number of singular points of type convex within the singular point subset, and the maximum height value within the singular point subset is greater than a preset height threshold, then the convexity type of the singular point subset is determined to be a peak-shaped convexity. If the number of singular points of type concave within the singular point subset is equal to or greater than the number of singular points of type convex, or if the maximum height value within the singular point subset is equal to or less than the height threshold, then the convexity type of the singular point subset is determined to be a gentle slope convexity.

11. The method according to claim 10, characterized in that, Determining the concavity / convexity type of singular points within each of the singular point subsets includes: The singular point with the largest height value within the set of singular points is determined as the highest singular point; Calculate the height difference between the height value of the highest singular point and the height value of any singular point within the subset of singular points; If the height difference is less than the planar projection distance between the highest singular point and any singular point, then the concave / convex point type is determined to be a concave point; If the height difference is equal to or greater than the planar projection distance between the highest singular point and any singular point, then the concave-convex point type is determined to be a convex point.

12. The method according to claim 1, characterized in that, Determining whether the graphite tray needs cleaning based on the protrusion type corresponding to the plurality of singularity sets includes: If the number of spike-shaped protrusions in the set of singularities exceeds a preset second threshold, then the graphite tray needs to be cleaned.

13. The method according to claim 1, characterized in that, Also includes: The three-dimensional data in the set of three-dimensional data is filtered to obtain a set of regular points, which is the complement of the set of singular points in the set of three-dimensional data. The thickness value of the current accumulation of the graphite tray is determined based on the average height value of each regular point in the set of regular points. Based on the thickness values ​​of the graphite tray deposits before and after a single process, determine the thickness increment value of the graphite tray deposits for a single process. The thickness increment of the process material in each process is calculated by summing the thickness increments of the graphite trays used in each process after the previous cleaning of the process chamber; If the thickness of the current buildup in the process chamber exceeds a preset thickness threshold, then the process chamber is determined to need cleaning.

14. The method according to claim 13, characterized in that, Also includes: Based on the doping gas flow rate during the current process and the thickness of the process chamber deposit after the current process, calculate the adjustment value of the doping gas flow rate for the next process. The dopant gas flow rate is determined for the next process based on the adjustment value of the dopant gas flow rate.

15. A semiconductor process apparatus, characterized in that, The method includes: a graphite tray, a three-dimensional scanning device disposed above the graphite tray, and a processing device connected to the three-dimensional scanning device. The three-dimensional scanning device is used to acquire three-dimensional data of multiple detection points on the surface of the graphite tray in a three-dimensional polar coordinate system by scanning. The processing device is used to implement the steps of the method as described in any one of claims 1-14 based on the set of the three-dimensional data.

16. A readable storage medium, characterized in that, A program or instructions are stored on the readable storage medium, which, when executed by a processor, implement the steps of the method as described in any one of claims 1-14.

Images