A method, system, apparatus, and readable storage medium for high efficiency production of silicon carbide wafers

Abstract

The present application relates to the technical field of semiconductor processing, and particularly relates to a high-efficiency production method, equipment and readable storage medium of silicon carbide wafer. The method adopts a crystal ingot doping concentration detection device to match multiple laser modification equipment, establishes an identity for the crystal ingot, obtains doping concentration distribution data in a detection coordinate system and extracts a reference feature set, and generates a cross-equipment laser processing parameter package; a central control module buffers the detected crystal ingot and schedules it to an idle laser modification equipment. The laser modification equipment does not repeatedly detect the doping concentration, only collects the current reference feature set, calculates a coordinate conversion matrix and a coordinate conversion residual error; when the residual error meets a threshold value, laser modification is performed, otherwise it is returned for rechecking. The present application can improve the utilization rate of the detection device and the laser modification equipment, and ensure the consistency of the cross-equipment processing positions.

Description

Technical Field

[0001] This invention relates to the field of semiconductor processing technology, and more particularly to a high-efficiency method, apparatus and readable storage medium for producing silicon carbide wafers. Background Technology

[0002] Silicon carbide (SiC), as a third-generation wide-bandgap semiconductor material, possesses characteristics such as a large bandgap, high breakdown electric field, high thermal conductivity, and resistance to high temperatures and high voltages. It has been widely used in new energy vehicles, rail transportation, smart grids, photovoltaic inverters, 5G communications, and high-power electronic devices. With the development of power devices towards higher voltage, higher current, and higher frequency, the market demand for large-size, high-quality silicon carbide substrates continues to grow. Silicon carbide ingots are processed into wafers through slicing, peeling, grinding, and polishing processes. The ingot slicing or peeling process directly affects the wafer thickness uniformity, surface damage layer thickness, material utilization, and subsequent epitaxial quality. Due to the high hardness, high brittleness, and strong chemical stability of silicon carbide, traditional mechanical cutting methods such as slurry wire cutting and diamond wire cutting typically suffer from low cutting efficiency, high kerf loss, and difficulty in controlling processing stress and surface damage. Existing public reports also point out that traditional cold-cutting processes for cutting and peeling silicon carbide ingots suffer from low efficiency and high losses, while laser cutting and peeling technology can achieve higher efficiency and higher quality silicon carbide substrate manufacturing.

[0003] In recent years, laser lift-off technology has been increasingly applied to silicon carbide ingot processing. This technology typically focuses a laser at a predetermined depth within the silicon carbide ingot, forming a modified layer, void layer, or crack propagation layer through scanning. The wafer is then separated from the ingot by external force, thermal stress, or subsequent lift-off processes. Compared to traditional wire cutting, laser lift-off offers advantages such as lower material loss, higher processing efficiency, and reduced material consumption. For example, Chinese patent application CN115555736A discloses a method and apparatus for laser lift-off of silicon carbide ingots. It uses an ultrashort pulse laser to form a void modified layer and a crack modified layer at a predetermined depth on the silicon carbide ingot. A short pulse laser then promotes the transverse growth and connection of cracks, thereby reducing longitudinal crack propagation and lowering the cutting loss thickness. This approach primarily focuses on the formation of the laser modified layer and crack propagation structure to improve lift-off quality and reduce material waste.

[0004] However, during the laser modification process of silicon carbide ingots, differences in doping concentration, resistivity, light transmission characteristics, or crystal defect distribution may exist in different regions of the ingot. If the entire ingot is processed with completely uniform laser parameters, it can easily lead to insufficient local modification, local thermal damage, uneven crack propagation, or fluctuations in surface quality after peeling. Therefore, existing technologies have developed schemes that first detect the local material characteristics of the ingot and then adjust the laser processing parameters accordingly. For example, Chinese Patent Publication No. CN111162017A discloses a method and device for detecting small-facet regions, as well as a method for generating wafers and a laser processing device. It distinguishes between small-facet regions and non-small-facet regions by detecting the fluorescence brightness of the silicon carbide ingot, and uses different laser energies and concentrator positions for different regions based on the detected boundary coordinates to improve the uniformity of the peeling layer formation. This scheme illustrates that differentiating laser processing conditions based on local ingot detection results has become an important technical direction for improving the quality of silicon carbide wafer generation.

[0005] For example, Chinese patent application CN121250552A discloses a processing method and its peeling sheet for improving the efficiency of laser ablation of silicon carbide ingots. This method involves initial resistivity testing at different locations on the surface of the polished silicon carbide ingot, calculating the target power value at each location based on the standard relationship between resistivity and laser power, and then setting laser processing conditions for different locations before performing laser refining processing. This reduces rework and improves peeling and slicing efficiency. This approach further demonstrates that detecting local electrical parameters of the ingot and mapping the results to positional laser processing parameters is an effective means of improving laser ablation efficiency and stability.

[0006] However, the aforementioned existing technologies mainly focus on how to detect local characteristics of ingots and how to adjust laser processing parameters based on the detection results. They typically treat detection, parameter generation, and laser refining as sequential steps within the same laser processing device or continuous processing flow. In actual production, the time required for ingot doping concentration detection, resistivity detection, or optical detection is usually significantly shorter than the time required for laser refining scanning. If each laser refining device is equipped with an independent detection unit, and detection and laser refining are performed sequentially before each processing, the detection unit is in a waiting or underutilized state most of the time. At the same time, the laser refining device also incurs additional cycle time loss due to the pre-detection steps, making it difficult to meet the requirements of high efficiency, low cost, and high equipment utilization for mass production of large-size silicon carbide wafers.

[0007] Furthermore, simply using a single detection device in parallel with multiple laser refining devices introduces new process control problems: the doping concentration distribution data generated by the detection device is usually based on the detection coordinate system, while the ingot's attitude, rotation angle, translation amount, and positioning error may change after being handled, buffered, and re-clamped by a robot in different laser refining devices. If the ingot identification, detection coordinates, reference features, processing coordinates, and laser parameter packages cannot be accurately bound and converted, mismatch between the detection area and the actual processing area may occur, leading to the application of laser power, focusing depth, or scanning parameters to the wrong location, thereby affecting the uniformity of the refining layer and the peeling quality. Therefore, the existing technology still lacks a high-efficiency silicon carbide wafer production method and equipment that can effectively combine independent doping concentration detection, multiple laser refining devices, ingot buffering, robot handling, cross-device coordinate transformation, and reuse of processing parameter packages. Summary of the Invention

[0008] This invention aims to provide a high-efficiency production method and equipment for silicon carbide wafers. By separating the ingot doping concentration detection process from the laser refining equipment, and using one ingot doping concentration detection device to match multiple laser refining devices, and combining ingot buffering, robot handling, central control scheduling, ingot identity binding, cross-device coordinate transformation, and processing parameter package reuse mechanisms, asynchronous parallel production of silicon carbide ingot detection and laser refining is achieved. This improves the utilization rate of laser refining equipment, reduces the cost of repeated configuration of detection devices, and ensures the consistency of processing position and the stability of refining quality when different laser refining devices call detection parameters.

[0009] Firstly, in order to achieve the above-mentioned objectives, the present invention adopts the following technical solution:

[0010] A high-efficiency method for producing silicon carbide wafers, the method comprising the following steps:

[0011] S1. Establish a production unit, which includes an ingot doping concentration detection device, an ingot buffer, an ingot handling robot, a central control module, and... Taiwan laser modification equipment, ≥2;

[0012] S2, the first feeding step Each silicon carbide ingot is given an identity identifier. Establish a detection coordinate system in the ingot doping concentration detection device. Doping concentration distribution data were obtained by detection. Extracting the baseline feature set and generate with Bound cross-device laser processing parameter package ;Will , , and Bind the storage and move the silicon carbide ingots that have completed the testing to the ingot cache location;

[0013] S3, when the When the laser modification equipment is idle, select the first ingot from the ingot buffer that has completed doping concentration detection and is not locked. A silicon carbide ingot, and the silicon carbide ingot and its corresponding... Locked to number Taiwan laser modification equipment, Number the laser-modified equipment;

[0014] S4, the The first laser modification equipment received the first After reading a silicon carbide ingot , establish the first Machining coordinate system of laser modification equipment Collect the current set of reference features of the silicon carbide ingot in the processing coordinate system. ,according to and Calculate the coordinate transformation matrix and coordinate transformation residuals ;

[0015] when At that time, Convert to device execution parameter package And perform laser modification processing; when At this time, processing is prohibited, and the product must be returned for verification or repositioning. The maximum allowed coordinate transformation residual threshold; For the first The silicon carbide ingot in the first The equipment execution parameter package in the laser refining equipment.

[0016] Preferably, in step S2, the silicon carbide ingot to be processed is loaded and an ingot identification mark is assigned. The ingot handling robot will... A silicon carbide ingot to be processed is transported to an ingot doping concentration detection device, which establishes a detection coordinate system. The doping concentration of the silicon carbide ingot to be processed is detected on the surface to be processed to obtain doping concentration distribution data in the detection coordinate system. And extract the benchmark feature set of the silicon carbide ingot. The central control module determines the doping concentration distribution data based on this data. and benchmark feature set Generate cross-device laser processing parameter packages .

[0017] Preferably, in step S2, the reference feature set The current reference feature set includes at least two of the following: the outer circumference of a silicon carbide ingot, positioning notch, crystal orientation marker, end face edge feature points, and pre-defined positioning marks; For the first The silicon carbide ingot was transported to the... After the laser modification equipment, by the first The laser-modified equipment acquires the reference feature set in the processing coordinate system. The corresponding feature set.

[0018] Preferably, in step S2, the ingot doping concentration detection device obtains the doping concentration-related detection value using at least one of fluorescence detection, resistivity detection, photoluminescence detection, and infrared transmittance detection. and the detection value related to the doping concentration Converted to doping concentration distribution data :

[0019] ;

[0020] In the formula, For the first A silicon carbide ingot is detected at the coordinates. The detected value of doping concentration at the location; For the first A silicon carbide ingot is detected at the coordinates. The doping concentration-related detection values ​​obtained at the location; This is the conversion factor between the doping concentration-related detection value and the doping concentration; Calculate the bias amount to convert the doping concentration; and These are two orthogonal coordinates in the detection coordinate system.

[0021] Preferably, in step S2, the cross-device laser processing parameter package include:

[0022] ;

[0023] In the formula, For the first Cross-device laser processing parameter package for a single silicon carbide ingot; For the first The identification mark of a silicon carbide ingot; For the first Doping concentration distribution data of a silicon carbide ingot in the detection coordinate system; For the first Laser power distribution of a silicon carbide ingot in the detection coordinate system; For the first Laser focusing depth distribution of a silicon carbide ingot in the detection coordinate system; For the first Laser scanning velocity distribution of a silicon carbide ingot in the detection coordinate system; For the first Laser scanning spacing distribution of individual silicon carbide ingots in the detection coordinate system; For the first Laser pulse repetition frequency distribution of a silicon carbide ingot in the detection coordinate system; For the first The set of baseline features for a silicon carbide ingot; This identifies the generation time or validity period of a cross-device laser processing parameter package.

[0024] Preferably, in step S4, the device executes the parameter package. Including laser power distribution in the machining coordinate system Laser focusing depth distribution Laser scanning speed distribution Laser scanning spacing distribution and laser pulse repetition frequency distribution And satisfy:

[0025] ;

[0026] ;

[0027] ;

[0028] In the formula, For the first The silicon carbide ingot in the first Laser power in the processing coordinate system of a laser modification equipment; The silicon carbide ingot is positioned in the detection coordinate system relative to the coordinates. The corresponding laser power; The laser focusing depth in the machining coordinate system; To detect the corresponding laser focusing depth in the coordinate system; The laser scanning speed in the machining coordinate system; To detect the corresponding laser scanning speed in the coordinate system; where, Depend on Coordinate transformation matrix Obtained through conversion;

[0029] The laser power distribution and laser focusing depth distribution Based on doping concentration distribution data Make corrections:

[0030] ;

[0031] ;

[0032] In the formula, For the first A silicon carbide ingot is detected at the coordinates. The corresponding laser power at that location; Reference doping concentration The corresponding reference laser power; This is the laser power correction coefficient corresponding to the change in doping concentration; For the first A silicon carbide ingot is detected at the coordinates. The detected value of doping concentration at the location; The preset reference doping concentration; For the first A silicon carbide ingot is detected at the coordinates. The corresponding laser focusing depth; Reference doping concentration The corresponding reference focus depth; This is the focusing depth correction coefficient corresponding to the change in doping concentration.

[0033] Preferably, in step S3, the central control module collects the real-time status of each laser modification device, the cache status of the ingot buffer, and the robot status of the ingot handling robot; when the... When the laser refining equipment is idle, the central control module selects the first ingot from the ingot cache that has completed doping concentration detection and is not locked. A silicon carbide ingot, and the silicon carbide ingot and its corresponding cross-device laser processing parameter package. Locked to number A laser-modified material processing device.

[0034] Preferably, in step S3, the central control module determines the work assignment evaluation value. Select the silicon carbide ingot from the ingot buffer and assign it to an idle laser refining device:

[0035] ;

[0036] In the formula, To make the first The silicon carbide ingots were allocated to the first... Evaluation value of the laser quality improvement equipment dispatch; For the first The waiting time after the laser modification equipment has been idle; For the first The waiting time of a silicon carbide ingot in the ingot cache bit; For the first The silicon carbide ingot is moved from its current position to the next position. Estimated transport time for the laser-modified equipment; For the first Each silicon carbide ingot corresponds to a cross-device laser processing parameter package. The remaining effective period penalty value; , , and These are non-negative weighting coefficients; the central control module executes first. Larger work assignment combinations.

[0037] Preferably, in step S4, the first The first laser modification equipment received the first After processing a silicon carbide ingot, instead of re-performing the doping concentration detection, the ingot's identification identifier is read. , establish the first Machining coordinate system of laser modification equipment Collect the current set of reference features of the silicon carbide ingot in the processing coordinate system. And based on the benchmark feature set Compared with the current benchmark feature set Calculate the coordinate transformation matrix This transforms the coordinates in the detection coordinate system into the machining coordinate system.

[0038] ;

[0039] In the formula, and The first The silicon carbide ingot in the first Two orthogonal coordinates in the processing coordinate system of a laser modification equipment; and The first Two orthogonal coordinates of a silicon carbide ingot in the detection coordinate system; For the first The silicon carbide ingot is transferred from the detection coordinate system to the first... The coordinate transformation matrix of the processing coordinate system of the laser modification equipment;

[0040] No. The laser-modified equipment or central control module calculates the coordinate transformation residual based on the reference feature points. :

[0041] ;

[0042] ;

[0043] In the formula, The number of reference feature points used in the calculation of coordinate transformation residuals; The reference feature point number; and The first The coordinates of each reference feature point in the detection coordinate system; and The first The actual acquired coordinates of each reference feature point in the machining coordinate system; and The first The coordinate transformation matrix of the reference feature points Predicted coordinates of the transformed machining coordinate system;

[0044] Preferably, the method further includes step S5: After the laser modification process is completed, the central control module notifies the ingot handling robot to transfer the processed silicon carbide ingot to the unloading position, ingot stripping equipment, or ingot thinning equipment, and releases the first step. The task of the laser refining equipment is locked; at the same time, the central control module continues to schedule the ingot doping concentration detection device to perform the doping concentration detection of the next silicon carbide ingot, so that the ingot doping concentration detection and the laser refining processing of multiple laser refining equipment are carried out asynchronously and in parallel.

[0045] Preferably, in step S4, when At that time, the first The laser-modified equipment is based on the coordinate transformation matrix. Cross-device laser processing parameter package Convert to device execution parameter package and according to the device execution parameter package For the A silicon carbide ingot undergoes laser modification; when At that time, the central control module prohibits the first A laser modification equipment performs laser modification processing and repositions the silicon carbide ingot to the verification position or reposition it.

[0046] Preferably, in step S4, the coordinate transformation matrix For rigid or approximately rigid transformation matrices:

[0047] ;

[0048] In the formula, For the first The silicon carbide ingot is transferred from the detection coordinate system to the first... The coordinate transformation matrix of the processing coordinate system of the laser modification equipment; This is a coordinate scale correction factor; For the first The silicon carbide ingot was transported to the... The plane rotation angle after the laser modification equipment; To detect the translation amount from the coordinate system to the machining coordinate system in the first coordinate direction; To detect the amount of translation from the coordinate system to the machining coordinate system in the second coordinate direction.

[0049] Secondly, the present invention also provides a high-efficiency silicon carbide wafer production system for implementing the method, comprising:

[0050] The ingot feeding module is used to receive silicon carbide ingots to be processed and provide the feeding position;

[0051] A crystal ingot doping concentration detection device is used to establish a detection coordinate system. The doping concentration of the silicon carbide ingot to be processed is detected on the end face to be processed, and doping concentration distribution data is generated. And extract the benchmark feature set. ;

[0052] Ingot buffer bit, used to buffer silicon carbide ingots that have completed doping concentration detection but have not yet undergone laser modification;

[0053] A laser-modified equipment is used to perform the process according to the equipment's parameter package. Laser modification of silicon carbide ingots;

[0054] Ingot handling robots are used in ingot loading modules, ingot doping concentration detection devices, and ingot buffer positions. The laser-modified equipment and the material feeding station are used to transport silicon carbide ingots.

[0055] The central control module, along with the ingot feeding module, ingot doping concentration detection device, ingot buffer, and ingot handling robot, and... Communication connection of the laser-modified equipment.

[0056] Preferably, the ingot handling robot is one or more of a multi-joint robot, a gantry robot, an RGV transport vehicle, or an AMR mobile robot; the ingot buffer is provided with a physical buffer, an identification structure, and a positioning structure, wherein the identification structure is used to read the ingot identification identifier. The positioning structure is used to limit the planar position and end face orientation of the silicon carbide ingot in the buffer state; the production equipment also includes an ingot stripping device and / or an ingot thinning device, and the central control module is a production line MES system or a device central control module that is communicatively connected to the production line MES system.

[0057] Thirdly, the present invention also provides an electronic device, including a processor, a memory, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the computer program, it implements the method described above.

[0058] Fourthly, the present invention also provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the method described thereon.

[0059] This invention employs a continuous production mechanism of independent detection, parameter binding, cache scheduling, coordinate transformation, and multi-machine refining. This eliminates the need for the ingot doping concentration detection device to be repeatedly configured with each laser refining machine. Instead, it serves as a shared detection unit, providing doping concentration distribution data and cross-machine laser processing parameter packages to multiple laser refining machines. This reduces equipment procurement and maintenance costs and avoids cycle time losses caused by repeatedly performing the detection process before each processing cycle. Because the central control module can dispatch tasks based on the idle status of each laser refining machine, the ingot cache status, and the robot status, the short-time doping concentration detection process and the long-time laser refining process can be asynchronously parallelized, significantly improving the overall continuous production efficiency. Furthermore, this invention establishes a detection coordinate system and extracts a set of reference features for the ingot during the detection stage. After the laser refining equipment receives the ingot, it re-acquires the current set of reference features and calculates the coordinate transformation matrix and coordinate transformation residual. This ensures that the doping concentration distribution data and laser processing parameters generated under the detection coordinate system can be accurately mapped to the processing coordinate system of any laser refining equipment. When the coordinate transformation residual exceeds a threshold, processing is prohibited or the kernel is returned, thereby avoiding mismatch between the detection area and the actual processing area due to ingot handling, buffering, rotation, or clamping deviations. Therefore, this invention not only improves production cycle time and equipment utilization but also ensures the accuracy of positional processing parameters such as laser power, focusing depth, and scanning speed when called across equipment. This is beneficial for forming a uniform and stable refining layer, reducing localized insufficient refining, thermal damage, or peeling quality fluctuations, and improving the consistency and yield of silicon carbide wafers in mass production. Attached Figure Description

[0060] Figure 1 This is a schematic diagram of the overall layout of a high-efficiency silicon carbide wafer production equipment according to the present invention.

[0061] Figure 2 This is a schematic diagram of the process for the high-efficiency production method of silicon carbide wafers according to the present invention.

[0062] Figure 3 This is a schematic diagram illustrating the relationship between ingot doping concentration detection and cross-device laser processing parameter package generation in this invention.

[0063] Figure 4This diagram illustrates the coordinate transformation and residual judgment between the detection coordinate system and the machining coordinate system in this invention.

[0064] Figure 5 This is a schematic diagram illustrating the scheduling relationship between the central control module, the ingot buffer, and multiple laser refining devices in this invention.

[0065] Figure 6 This is a comparison diagram of the production cycle time between the embodiments and comparative examples of the present invention.

[0066] Figure 7 This is a residual distribution diagram of the cross-device coordinate transformation of the present invention.

[0067] Figure 8 This is a comparison chart of the quality of the modified layer and the peeling quality of the present invention.

[0068] Explanation of reference numerals in the attached figures: Detailed Implementation

[0069] The embodiments of the present invention will be further described below with reference to the accompanying drawings. It should be understood that the following embodiments are intended to help those skilled in the art understand and implement the present invention, and do not constitute a limitation on the scope of protection of the present invention. Without departing from the core concept of the present invention, those skilled in the art can make adaptive adjustments to the number of equipment, spatial layout, and parameter range according to production cycle, ingot size, laser refining equipment model, handling robot type, and detection method.

[0070] The core concept of this invention lies in separating the doping concentration detection process of silicon carbide ingots from each laser refining device, forming a shared ingot doping concentration detection device. After detection, instead of generating ordinary laser processing parameters, a cross-device laser processing parameter package is generated, which is jointly bound to the ingot's identity, detection coordinate system, doping concentration distribution data, and reference feature set. When the ingot is transported to any laser refining device, that device no longer repeats the doping concentration detection. Instead, it converts the laser processing parameters generated in the detection coordinate system to the current processing coordinate system through identity reading, posture verification, coordinate transformation, and residual judgment, and performs processing accordingly. Thus, this invention simultaneously solves two technical problems: first, the problem of low equipment utilization due to the mismatch between the detection process and the laser refining process cycle; and second, the problem of easy mismatch between the detection coordinates and the actual processing coordinates when processing across devices after independent detection.

[0071] I. Terminology Explanation

[0072] 1. Silicon carbide ingots refer to SiC crystal rods or blocks formed by sublimation, physical vapor transport, or other crystal growth methods. They can be further processed into silicon carbide wafers through laser modification, peeling, grinding, thinning, and polishing.

[0073] 2. Doping concentration detection refers to the use of fluorescence, resistivity, photoluminescence, infrared transmittance or other detection methods related to doping concentration to perform zoned detection on the end face or near-surface area of ​​silicon carbide ingots in order to obtain detection data reflecting the electrical or optical differences of the material at different locations.

[0074] 3. Detection coordinate system , refers to the planar coordinate system established by the ingot doping concentration detection device during the detection stage, which is used to describe the positions of each detection point, reference feature point and doping concentration distribution data on the ingot end face.

[0075] 4. Machining coordinate system , refers to the The planar coordinate system established by the laser modification equipment after receiving the ingot is used to describe the actual clamping position, rotation attitude and laser scanning path of the ingot on the current stage.

[0076] 5. Baseline Feature Set This refers to the testing phase from the first... The set of features extracted from a silicon carbide ingot that can be used for positioning and coordinate matching includes at least two of the following: outer circle contour, positioning notch, crystal orientation identifier, edge feature point, and pre-defined positioning mark.

[0077] 6. Current benchmark feature set , refers to the The silicon carbide ingot was transported to the... After the laser refining equipment is installed, the attitude verification unit of the laser refining equipment collects the set of reference features. The corresponding current feature set.

[0078] 7. Cross-device laser processing parameter package This refers to the generation on the detection device side, and its association with the ingot's identity identifier. The bound data packet includes information such as doping concentration distribution data in the detection coordinate system, laser power distribution, focusing depth distribution, scanning speed distribution, scanning spacing distribution, pulse repetition frequency distribution, reference feature set, and validity period identifier.

[0079] 8. Equipment execution parameter package , refers to the One silicon carbide ingot was assigned to the first... After the laser modification equipment is installed, the laser processing parameter package is used across the equipment. Coordinate transformation matrix The converted result can be directly supplied to the first A set of processing parameters executed by a laser-modified equipment.

[0080] 9. Coordinate transformation residuals , refers to the average deviation between the predicted position of the reference feature point after being mapped to the machining coordinate system by the coordinate transformation matrix during the detection stage and the actual acquisition position in the machining coordinate system. It is used to determine whether the matching between the detection coordinate system and the machining coordinate system is reliable.

[0081] 10. Asynchronous parallel production refers to the process where the ingot doping concentration detection device, ingot buffer, ingot handling robot, and multiple laser refining devices do not wait to be executed in a fixed order within a single device, but are instead operated in parallel under the unified scheduling of the central control module, enabling the detection process, handling process, and multiple laser refining processes to run in parallel.

[0082] II. System Structure

[0083] like Figure 1 As shown, the high-efficiency silicon carbide wafer production equipment of the present invention includes an ingot feeding module, an ingot doping concentration detection device, an ingot buffer, an ingot handling robot, a central control module, and... Taiwan laser modification equipment, among which greater than or equal to Integers. Figure 1 In the illustrated embodiment, Four laser refining devices are positioned on the left and right sides of the central inspection and handling area. The central area is equipped with an ingot doping concentration detection device, an ingot buffer, a loading position, and a unloading position. An ingot handling robot is located in the central area and can handle silicon carbide ingots between the loading position, the ingot doping concentration detection device, the ingot buffer, the various laser refining devices, and the unloading position.

[0084] The ingot loading module receives silicon carbide ingots to be processed. It can be a manual loading platform, an automatic hopper, a lifting hopper, or an automated conveyor mechanism connected to upstream cleaning and polishing processes. The ingot loading module is equipped with an ingot identification generation or reading structure, such as a QR code reader, RFID reader, laser marking unit, or batch number reading unit associated with the MES system. After the silicon carbide ingot enters the production unit, it is assigned an ingot identification tag. The label remains unchanged during subsequent inspection, caching, handling, dispatching, laser modification, and unloading processes.

[0085] A silicon carbide ingot doping concentration detection device is used to detect the doping concentration on the surface of a silicon carbide ingot to be processed. This device can employ one or more combinations of fluorescence detection, resistivity detection, photoluminescence detection, or infrared transmittance detection. For fluorescence detection, the device may include an excitation source, a fluorescence receiver, an imaging unit, and a data processing unit; for resistivity detection, the device may include a multi-point probe, an electrode contact module, and a resistivity conversion module; for photoluminescence detection, the device may include an excitation source, a spectral acquisition unit, and a spectral analysis module. This invention does not limit the specific detection mechanism, as long as it can obtain detection values ​​related to the doping concentration at different locations on the ingot and form location-based distribution data.

[0086] The ingot cache includes a physical cache and a parameter cache. The physical cache is used to temporarily store silicon carbide ingots that have undergone doping concentration detection but have not yet entered the laser refining equipment; the parameter cache is used to store ingot identification information. The system includes bound doping concentration distribution data, a set of baseline features, and cross-device laser processing parameter packages. Multiple ingot positioning stations can be configured in the physical buffer, each equipped with a limiting ring, vacuum adsorption structure, mechanical positioning block, or end-face orientation restriction structure to prevent uncontrolled displacement of the ingot during the buffering process. The parameter buffer area can be located in the central control module, the production line MES system, or a separate edge computing controller.

[0087] Ingot handling robots are used to move silicon carbide ingots between workstations. They can be articulated robots, gantry robots, RGV (Automated Guided Vehicle) transport vehicles, AMR (Automatic Mobile Robot) mobile robots, or combinations thereof. The end effector of the ingot handling robot can employ vacuum suction grippers, flexible jaws, mechanical clamping mechanisms, or composite pick-and-place mechanisms to ensure that the ingot end faces are not scratched during handling. Considering the typically high value and high processing precision requirements of silicon carbide ingots, the handling robot can be equipped with force control detection, posture detection, and clamping abnormality alarm functions.

[0088] The central control module is the core control unit of this invention, enabling high-efficiency production and cross-equipment parameter reuse. The central control module communicates with the ingot feeding module, ingot doping concentration detection device, ingot buffer, ingot handling robot, and various laser modification devices. The central control module is used for task assignment, status acquisition, parameter package management, ingot identity binding, buffer scheduling, work assignment calculation, coordinate transformation data management, residual threshold determination, and anomaly handling. The central control module can be part of an equipment PLC, industrial computer, edge controller, or production line MES system, or it can be a dedicated equipment control module that communicates with the MES system.

[0089] Each laser-modified silicon carbide ingot includes a stage, optical path, laser, identification reading unit, attitude verification unit, coordinate transformation unit, and laser execution unit. The stage supports and holds the silicon carbide ingot; the optical path focuses the laser emitted by the laser to a predetermined depth inside the ingot; and the identification reading unit reads the ingot's identification markings. The attitude verification unit is used to collect the current reference feature set. The coordinate transformation unit is used to calculate the coordinate transformation matrix. and coordinate transformation residuals The laser actuator is used to execute the equipment parameter package. Complete laser-modified processing.

[0090] III. Specific Technical Route for Implementing the Method of the Invention

[0091] The method of the present invention can be followed Figure 2 The illustrated process mainly includes production unit establishment, ingot identification and detection, parameter package generation, ingot caching, equipment dispatching, coordinate transformation, residual judgment, laser modification, and subsequent transportation. Each step is described below, with doping concentration detection, cross-equipment parameter package generation, coordinate transformation, residual judgment, and multi-machine asynchronous scheduling being the key implementation aspects of this invention.

[0092] S1. Establish production units

[0093] In this embodiment, firstly, establish as follows Figure 1 The production unit shown is equipped with an ingot doping concentration detection device, an ingot buffer, an ingot handling robot, a central control module, and... A laser-modified treatment device. Preferably, The value is determined based on the doping concentration detection cycle, the robot handling cycle, and the laser modification processing cycle.

[0094] The central control module can determine the actual number of laser refining devices activated based on the following cycle time relationship:

[0095] ;

[0096] In the formula, The cycle ratio is used to characterize the cycle matching relationship between a single ingot doping concentration detection device and multiple laser modification devices. The average time for a single silicon carbide ingot to undergo laser modification in a single laser modification device; The average time for a single silicon carbide ingot to complete doping concentration detection and generate detection data in the ingot doping concentration detection device; The average time for the ingot handling robot to complete one ingot transfer.

[0097] In actual configurations, the number of laser refining devices actually in use... It can be satisfied:

[0098] ;

[0099] In the formula, This refers to the actual number of laser-based refining equipment in use. The total number of laser refining devices configured in the production unit; For beat ratio; Indicates not greater than The largest integer.

[0100] This step rationally configures multiple laser modification devices according to the cycle time of detection, handling, and modification, avoiding significant bottlenecks between the detection device and the laser modification equipment. Compared with the traditional sequential processing method of one laser modification device per detection unit, this invention provides a hardware foundation for subsequent continuous and efficient production by sharing detection devices and multi-machine asynchronous processing.

[0101] S2. Ingot identification, doping concentration detection and benchmark feature extraction

[0102] After the silicon carbide ingot to be processed enters the ingot feeding module, the central control module establishes an ingot identification for it. The aforementioned The identifier can be formed by combining batch number, ingot number, loading time, processing batch number, and production line number, or it can be directly obtained by reading QR code, RFID tag, or laser marking. The purpose of establishing ingot identification is to ensure that subsequent testing data, cache location, parameter package, handling tasks, and processing records can all be traced back to the same ingot, avoiding the misuse of parameter packages between different ingots.

[0103] Ingot handling robot will be the first A silicon carbide ingot to be processed is transported to the ingot doping concentration detection device. The ingot doping concentration detection device establishes a detection coordinate system before detection. In a preferred embodiment, the origin of the detection coordinate system is... The center of the crystal ingot end face can be taken. The axis can be taken as the direction passing through the positioning notch, crystal orientation mark, or preset positioning mark. shaft and The axes are orthogonal. If the ingot is a circular or cylindrical end face with a positioning notch, the center of the end face can be determined by fitting the outer circle, and the angular reference of the detection coordinate system can be determined by the direction of the positioning notch. If a pre-defined positioning mark is set on the end face of the ingot, the translation and rotation reference of the coordinate system can be determined by two or more positioning marks.

[0104] The ingot doping concentration detection device performs zoned detection on the ingot's surface to be processed. Specifically, the surface to be processed can be divided into several detection grids, each grid corresponding to a detection coordinate. The detection device obtains doping concentration-related detection values ​​for each detection point or detection area. Doping concentration distribution data are obtained based on pre-calibrated conversion relationships. In one linear conversion method, the following formula can be used:

[0105] ;

[0106] In the formula, For the first A silicon carbide ingot at the detection coordinates The detected value of doping concentration at the location; For the first A silicon carbide ingot at the detection coordinates The doping concentration-related detection values ​​obtained at the location; This is the conversion factor between the doping concentration-related detection value and the doping concentration; Calculate the bias amount to convert the doping concentration; and These are two orthogonal coordinates in the detection coordinate system.

[0107] In another implementation, if the relationship between the detected value and the doping concentration is non-linear, a piecewise calibration table, polynomial fitting, or machine learning regression model can be used for conversion. This invention does not limit the specific conversion model, as long as it can obtain positional doping concentration distribution data that can be used to generate laser processing parameters based on the detected value.

[0108] While detecting the doping concentration distribution, the ingot doping concentration detection device also extracts the first... The set of reference features for each crystal ingot The reference feature set It can include at least two of the following: outer circular profile, positioning notch, crystal orientation marker, end face edge feature points, and pre-defined positioning marks. By simultaneously acquiring doping concentration distribution data and a set of reference features, this invention can bind material distribution information and spatial positioning information together in the same detection coordinate system. This is a significant improvement of this invention, distinguishing it from simply detecting resistivity or fluorescence intensity and processing directly in the same device.

[0109] In this step, if doping concentration distribution data is obtained only from an independent detection device without extracting a set of reference features for cross-device coordinate matching, the parameter distribution in the detection coordinate system will be difficult to directly correspond to the actual processing position after the ingot is transported to different laser refining devices due to rotation, translation, and attitude errors caused by reclamping. In existing same-machine detection and processing methods, the detection coordinates and processing coordinates are usually within the same device coordinate framework, resulting in small coordinate deviations. However, this invention employs an asynchronous parallel structure with a shared detection device and multiple laser refining devices, necessitating the establishment of a cross-device coordinate consistency mechanism. Therefore, this step will... and Simultaneously collect and bind data for subsequent calculation of the coordinate transformation matrix. and residual Provide the necessary data foundation.

[0110] S3. Cross-device laser processing parameter package generation and binding storage

[0111] Obtaining doping concentration distribution data and benchmark feature set Subsequently, the central control module generates a cross-equipment laser processing parameter package according to preset process rules. Preferably, the cross-device laser processing parameter package Includes the following data:

[0112] ;

[0113] In the formula, For the first Cross-device laser processing parameter package for a single silicon carbide ingot; For the first The identification mark of a silicon carbide ingot; For the first Doping concentration distribution data of a silicon carbide ingot in the detection coordinate system; For the first Laser power distribution of a silicon carbide ingot in the detection coordinate system; For the first Laser focusing depth distribution of a silicon carbide ingot in the detection coordinate system; For the first Laser scanning velocity distribution of a silicon carbide ingot in the detection coordinate system; For the first Laser scanning spacing distribution of individual silicon carbide ingots in the detection coordinate system; For the first Laser pulse repetition frequency distribution of a silicon carbide ingot in the detection coordinate system; For the first The set of baseline features for a silicon carbide ingot; This identifies the generation time or validity period of a cross-device laser processing parameter package.

[0114] In one embodiment, the laser power distribution and laser focusing depth distribution Based on doping concentration distribution data Make corrections:

[0115] ;

[0116] In the formula, For the first A silicon carbide ingot is detected at the coordinates. The corresponding laser power at that location; Reference doping concentration The corresponding reference laser power; This is the laser power correction coefficient corresponding to the change in doping concentration; For the first A silicon carbide ingot is detected at the coordinates. The detected value of doping concentration at the location; This is the preset baseline doping concentration.

[0117] ;

[0118] In the formula, For the first A silicon carbide ingot is detected at the coordinates. The corresponding laser focusing depth; Reference doping concentration The corresponding reference focus depth; This is the focusing depth correction coefficient corresponding to the change in doping concentration; For the first A silicon carbide ingot is detected at the coordinates. The detected value of doping concentration at the location; This is the preset baseline doping concentration.

[0119] In practice, , , , and This can be obtained through process databases, calibration curves, empirical tables, piecewise functions, or closed-loop experimental data. For example, for regions with high doping concentration and strong absorption, the laser power can be appropriately reduced or the focusing depth adjusted; for regions with low doping concentration and high modification threshold, the laser power can be appropriately increased, the scanning speed reduced, or the scanning spacing decreased to improve the continuity of the modified layer.

[0120] After generating the cross-device laser processing parameter package, the central control module will... , , and Binding and storage. Once the ingot has been inspected, it is moved to the ingot buffer position by the ingot handling robot. The physical buffer position records the actual storage location of the ingot, and the parameter buffer records the data packet corresponding to that ingot. The central control module marks the ingot's status as inspected, unlocked, or awaiting dispatch.

[0121] In this step, the parameter package is no longer a temporary processing parameter for use within the same device, but a data structure that can be used across devices. Because It simultaneously includes doping concentration distribution data, laser processing parameter distribution, and a set of reference features. Therefore, when an ingot is dispatched to any laser refining device, the correspondence between the parameters and the actual processing position can be restored through coordinate transformation. This method avoids the need to repeatedly set up a complete doping concentration detection device on each laser refining device, and also avoids the risk of not being able to guarantee position consistency by simply sharing detection data.

[0122] S4, Multi-machine Asynchronous Dispatch and Ingot Cache Scheduling

[0123] When there is a detected and unlocked ingot in the ingot buffer slot, the central control module collects the equipment status of each laser modification device, the buffer status of the ingot buffer slot, and the robot status of the ingot handling robot in real time. The equipment status includes at least idle, processing, processing completed, and fault; the buffer status includes buffer slot occupancy, ingot identity, parameter package status, lock status, and waiting time; the robot status includes idle, handling, abnormal, and maintenance status.

[0124] When the When the laser refining equipment is idle, the central control module selects a non-locked ingot from the ingot buffer that has completed doping concentration detection. A crystal ingot is generated, and the ingot is associated with the corresponding cross-device laser processing parameter package. Locked to number A laser-modified equipment. To reduce idle waiting time, shorten transportation distances, and avoid parameter package expiration, the central control module can select dispatch combinations based on dispatch evaluation values:

[0125] ;

[0126] In the formula, To make the first The silicon carbide ingots were allocated to the first... Evaluation value of the laser quality improvement equipment dispatch; For the first The waiting time after the laser modification equipment has been idle; For the first The waiting time of a silicon carbide ingot in the ingot cache bit; For the first The silicon carbide ingot is moved from its current position to the next position. Estimated transport time for the laser-modified equipment; For the first Each silicon carbide ingot corresponds to a cross-device laser processing parameter package. The remaining effective period penalty value; , , and These are the non-negative weighting coefficients.

[0127] Central control module preferred Larger work assignment combinations. If a laser modification machine has been idle for a long time, then If the value is large, the system tends to prioritize replenishing that device; if a crystal ingot waits in the cache for a long time, then... If the ingot is relatively large, the system tends to prioritize processing that ingot; if an ingot is far from a laser modification device, then... A larger value results in a lower work assignment evaluation score; if a parameter package is about to expire, then... The penalty can be set to a smaller value or prioritized based on system strategy to avoid parameter invalidation. In practice, the above-mentioned work assignment evaluation value can be adjusted according to production needs and is not limited to a linear expression.

[0128] Regarding cache control, when the number of unlocked ingots and completed detection is too small, the central control module prioritizes issuing detection tasks to the ingot doping concentration detection device; when the cache is nearly full, the central control module suspends new detection tasks to avoid cache congestion. The following constraints can be adopted:

[0129] ;

[0130] In the formula, This refers to the number of silicon carbide ingots in the ingot buffer that are unlocked and have completed doping concentration detection. This is a preset lower limit for the cache, used to reduce idle waiting time for laser refining equipment; The preset cache limit is used to limit the number of ingot cache bits occupied and avoid cache congestion.

[0131] S5. Establishment of machining coordinate system and calculation of coordinate transformation matrix

[0132] No. The first laser modification equipment received the first After a silicon carbide ingot is processed, the doping concentration detection is not repeated. Instead, the ingot's identification identifier is first read by the identification reading unit. Central control module or the first Taiwan laser modification equipment according to Retrieve the cross-device laser processing parameter package corresponding to the crystal ingot Subsequently, the first Establishing a processing coordinate system for a laser-modified equipment The attitude verification unit collects the current reference feature set of the ingot on the current stage. .

[0133] The attitude verification unit can employ camera vision, laser ranging, contour scanning, edge detection, or a combination of sensing methods. If the reference feature set... If the ingot contains an outer circular profile and a positioning notch, the attitude verification unit can acquire an image of the ingot's end face, identify the location of the outer circular profile and the positioning notch, and determine the translation and rotation angle of the ingot on the stage. If the reference feature set... If a pre-defined positioning marker is included, the corresponding point can be directly obtained through image recognition. This verification process does not involve re-detecting the doping concentration, but rather rapidly acquiring the ingot posture and reference features, requiring a time far less than that for complete doping concentration detection.

[0134] Based on the baseline feature set of the detection phase and the current benchmark feature set The coordinate transformation unit calculates the coordinate transformation matrix. This converts the end face coordinates in the inspection coordinate system to the end face coordinates in the machining coordinate system.

[0135] ;

[0136] In the formula, and The first The silicon carbide ingot in the first Two orthogonal coordinates in the processing coordinate system of a laser modification equipment; and The first Two orthogonal coordinates of a silicon carbide ingot in the detection coordinate system; For the first The silicon carbide ingot is transferred from the detection coordinate system to the first... The coordinate transformation matrix of the processing coordinate system of the laser modification equipment; Number the laser-modified equipment; This is the ingot serial number.

[0137] In a preferred embodiment, the coordinate transformation matrix For rigid or approximately rigid transformation matrices:

[0138] ;

[0139] In the formula, For the first The silicon carbide ingot is transferred from the detection coordinate system to the first... The coordinate transformation matrix of the processing coordinate system of the laser modification equipment; This is a coordinate scale correction factor; For the first The silicon carbide ingot was transported to the... The plane rotation angle after the laser modification equipment; To detect the translation amount from the coordinate system to the machining coordinate system in the first coordinate direction; To detect the amount of translation from the coordinate system to the machining coordinate system in the second coordinate direction.

[0140] The technical significance of this step lies in the fact that there will inevitably be different stage coordinates, clamping errors, and changes in transport posture between an independent detection device and multiple laser refining devices. Without a coordinate transformation matrix... Detection coordinate system , and It is impossible to reliably map the actual laser scanning position to the processing coordinate system. Through this step, the present invention elevates shared detection from a simple production efficiency optimization to a technical solution for precise cross-device processing, enabling laser modification equipment to accurately use the detection results without re-performing doping concentration detection.

[0141] S6. Coordinate Transformation Residual Judgment and Equipment Execution Parameter Package Generation

[0142] Simply calculating the coordinate transformation matrix is ​​insufficient to guarantee processing reliability, as local contamination, identification errors, notch occlusion, clamping slippage, or datum feature mismatch may occur during ingot handling and clamping. To avoid misalignment of laser processing parameters due to erroneous coordinate transformations, this invention further calculates the coordinate transformation residual. .

[0143] First, coordinate transformation prediction is performed on the benchmark feature points participating in the review:

[0144] ;

[0145] In the formula, and The first The coordinate transformation matrix of the reference feature points Predicted coordinates of the transformed machining coordinate system; This is the coordinate transformation matrix; and The first The coordinates of each reference feature point in the detection coordinate system; The reference feature point number.

[0146] Then, the coordinate transformation residual is calculated based on the difference between the predicted coordinates and the actual acquired coordinates:

[0147] ;

[0148] In the formula, For the first The silicon carbide ingot in the first Coordinate transformation residuals on a laser refining device; The number of reference feature points used in the calculation of coordinate transformation residuals; The reference feature point number; and The first The actual acquired coordinates of each reference feature point in the machining coordinate system; and The first The coordinate transformation matrix of the reference feature points Predicted coordinates of the transformed machining coordinate system.

[0149] when When this occurs, it indicates that the transformation relationship between the detection coordinate system and the machining coordinate system meets the accuracy requirements. The laser modification equipment can integrate laser processing parameter packages across different devices. Convert to device execution parameter package .when If this occurs, it indicates an anomaly in ingot posture recognition, positioning, or coordinate transformation, and the central control module will prohibit the [further actions]. The laser modification equipment performs laser modification processing and schedules the ingot to the verification position or repositioning position.

[0150] If processing is permitted, the equipment executes the parameter package. Including laser power distribution in the machining coordinate system Laser focusing depth distribution Laser scanning speed distribution Laser scanning spacing distribution and laser pulse repetition frequency distribution When the coordinate transformation relationship holds, the following conditions must be met:

[0151] ;

[0152] In the formula, For the first The silicon carbide ingot in the first Laser power in the processing coordinate system of a laser modification equipment; The silicon carbide ingot is positioned in the detection coordinate system relative to the coordinates. The corresponding laser power; Depend on Coordinate transformation matrix Obtained through conversion.

[0153] ;

[0154] In the formula, For the first The silicon carbide ingot in the first Laser focusing depth in the processing coordinate system of a laser modification equipment; The silicon carbide ingot is positioned in the detection coordinate system relative to the coordinates. The corresponding laser focusing depth; Depend on Coordinate transformation matrix Obtained through conversion.

[0155] ;

[0156] In the formula, For the first The silicon carbide ingot in the first Laser scanning speed in the processing coordinate system of a laser modification equipment; The silicon carbide ingot is positioned in the detection coordinate system relative to the coordinates. The corresponding laser scanning speed; Depend on Coordinate transformation matrix Obtained through conversion.

[0157] This step, together with S5, constitutes the core technology that distinguishes this invention from conventional multi-machine parallel production lines. Conventional production lines can improve logistics efficiency through robots and buffers, but they cannot solve the spatial mismatch problem when inspection data is reused across devices. This invention addresses this issue through residual thresholding. A safety check is established before processing, ensuring that the laser parameter package is only executed if the coordinate transformation is reliable. This avoids errors in local power application, misalignment of focusing depth, discontinuity in the modified layer, or fluctuations in peeling quality. This mechanism guarantees both production efficiency and processing accuracy, representing a significant improvement of this invention compared to existing technologies.

[0158] S7, Laser Modification Processing, Material Cutting and Subsequent Process Connection

[0159] Once the coordinate transformation residuals meet the requirements, the first... The laser modification equipment is executed according to the equipment parameter package. For the first A silicon carbide ingot undergoes laser modification. Laser modification may include forming a modified layer, a void layer, or a crack propagation layer within the ingot along a predetermined scanning path. The laser actuation unit... , , , and By controlling the laser power, focusing position, scanning speed, scanning spacing, and pulse repetition frequency, appropriate modification energy can be obtained for regions with different doping concentrations.

[0160] After laser modification is completed, the first The laser-modified silicon carbide ingot sends a processing completion signal to the central control module. The central control module then instructs the ingot handling robot to transfer the processed silicon carbide ingot to the unloading station, ingot stripping equipment, or ingot thinning equipment. If the production line integrates ingot stripping equipment, the laser-modified ingot can directly enter the stripping process; if thinning or surface treatment is required first, the central control module can schedule it to the ingot thinning equipment based on the status of the downstream equipment. After the transfer is completed, the central control module releases the first... The task of the laser refining equipment is locked, and the equipment status is updated to idle or pending cleaning so that it can continue to receive the next ingot task.

[0161] This step has a relatively low inventive contribution; its main function is to complete the production loop after laser refining and to connect with subsequent processes. However, in the overall scheme of this invention, this step ensures that multiple laser refining devices can continuously release and receive tasks, thereby maintaining the stable operation of asynchronous parallel production.

[0162] IV. Specific Application Examples and Experimental Data

[0163] (I) Experimental Objective

[0164] To verify the technical effect of the present invention—namely, matching one ingot doping concentration detection device with multiple laser modification devices and achieving high-efficiency production through ingot identification, cross-device laser processing parameter packages, coordinate transformation matrix, and coordinate transformation residual judgment—this embodiment constructs the following... Figure 1 The high-efficiency production test platform for silicon carbide wafers shown is in accordance with... Figure 2 The method flow shown was verified through continuous processing. The experiment focused on the following: First, the effect of asynchronous parallel operation of an independent doping concentration detection device and multiple laser modification devices on improving the overall production capacity and equipment utilization; Second, the detection coordinate system. Coordinate system of each laser-modified processing equipment Through coordinate transformation matrix After mapping, can the consistency between the detection data and the actual processing position be guaranteed? Third, through coordinate transformation residuals... The threshold judgment mechanism should be used to effectively avoid parameter misalignment caused by ingot handling, clamping deviations, or posture errors. Fourth, doping concentration distribution data should be used. Generate positional laser processing parameter package Afterwards, it remains to be seen whether the continuity of the modified layer, the uniformity of the peel thickness, and the wafer yield can be improved.

[0165] (II) Test equipment and samples

[0166] This embodiment uses 4H-SiC silicon carbide ingots as test samples. The ingot diameter is 150mm. The ingot's end face to be processed is rough polished before entering the testing process. The experimental production unit includes one ingot doping concentration detection device, one six-station ingot buffer, one multi-joint ingot handling robot, and four laser refining devices. The four laser refining devices are denoted as follows: , , and Among them, the ingot doping concentration detection device adopts a combination of photoluminescence detection and end face image recognition, which can simultaneously obtain the doping concentration-related detection values ​​and the ingot reference feature set; the laser modification equipment is equipped with an identity reading unit, an attitude verification unit, a coordinate transformation unit and a laser execution unit.

[0167] Two control groups were set up in the experiment:

[0168] Comparative Example 1: The traditional single-machine sequential processing method is adopted, that is, each laser modification equipment performs ingot identification reading, doping concentration detection, laser parameter generation and laser modification processing in sequence within its own equipment. There is no shared doping concentration detection device or cross-equipment parameter package reuse mechanism.

[0169] Example 1: Using the solution of the present invention, a single independent ingot doping concentration detection device completes the doping concentration detection and reference feature extraction, generating a cross-device laser processing parameter package. After being cached, the ingot is dispatched by the central control module to an idle laser modification device, and then processed through a coordinate transformation matrix. and coordinate transformation residuals Perform a pre-processing verification.

[0170] (III) Generation of test data and cross-device parameter packages

[0171] In Example 1, after each crystal ingot enters the crystal ingot doping concentration detection device, a detection coordinate system is first established. The origin of the coordinate system is the center of the fitted circle of the outer circle of the ingot, and the direction of the center of the positioning notch of the ingot is taken as the coordinate origin. Axial direction. The detection device performs zone detection on the end face of the ingot to be processed according to a 5mm×5mm grid to obtain detection values ​​related to doping concentration. And converted into doping concentration distribution data through calibration relationships. :

[0172] ;

[0173] In the formula, For the first Each crystal ingot is detected at coordinates The detected value of doping concentration at the location; This is the detection value related to the doping concentration at this location; This is the conversion factor between the detected value and the doping concentration; This is for calculating the offset.

[0174] After the test is completed, the central control module according to Generate laser power distribution and focus depth distribution :

[0175] ;

[0176] In the formula, For the first Each crystal ingot is detected at coordinates The corresponding laser power at that location; Reference doping concentration The corresponding reference laser power; This is the laser power correction coefficient corresponding to the change in doping concentration; To detect coordinates The detected value of doping concentration at the location; This is the preset baseline doping concentration.

[0177] ;

[0178] In the formula, For the first Each crystal ingot is detected at coordinates The corresponding laser focusing depth; Reference doping concentration The corresponding reference focus depth; This is the focusing depth correction coefficient corresponding to the change in doping concentration.

[0179] Simultaneously, the detection device extracts the outer contour of the ingot, locates the center point of the notch, and three edge feature points to form a set of reference features. and with the ingot identification mark Doping concentration distribution data Together with laser processing parameters, they are packaged into a cross-device laser processing parameter package. .

[0180] Table 1 shows the main test results obtained for the 10 crystal ingots during the testing phase.

[0181]

[0182] As shown in Table 1, there are varying degrees of differences in doping concentration distribution on the end faces of each ingot, with the maximum difference being... to Without zoning parameter correction, insufficient or excessive local refining energy can easily occur. This invention generates a positional parameter package in the detection coordinate system, enabling different laser refining devices to access the precise processing parameters corresponding to the same ingot.

[0183] IV. Cross-device coordinate transformation verification

[0184] To verify Figure 4 The effectiveness of the coordinate transformation and residual judgment mechanism shown will be demonstrated by randomly assigning the completed ingots to four laser modification devices. , , and Each laser-modified ingot receiving unit only performs identity reading and attitude verification; it does not re-perform doping concentration detection. The attitude verification unit collects the current set of reference features. and the baseline feature set in the detection phase. Matching, calculating the coordinate transformation matrix :

[0185] ;

[0186] In the formula, and These are the coordinates in the machining coordinate system; and These are the coordinates in the detection coordinate system; For the first The crystal ingot is transferred from the detection coordinate system to the first... The coordinate transformation matrix of the processing coordinate system of the laser-modified equipment.

[0187] After coordinate transformation, the coordinate transformation residuals are further calculated. :

[0188] ;

[0189] In the formula, For the first The ingot in the first Coordinate transformation residuals on a laser refining device; The number of reference feature points used in residual calculation; The reference feature point number; and The first The actual acquired coordinates of each reference feature point in the machining coordinate system; and The first The coordinate transformation matrix of the reference feature points The predicted coordinates obtained from the transformation.

[0190] In this embodiment, the maximum allowable coordinate transformation residual threshold is set as follows:

[0191] ;

[0192] In the formula, The maximum coordinate transformation residual threshold that allows laser refining processes to be performed.

[0193] Table 2 presents the verification results of coordinate transformation for 12 randomly selected cross-device dispatching operations.

[0194]

[0195] As shown in Table 2, most ingots exhibit varying degrees of translational and rotational changes after being handled and reclamped across equipment. Without coordinate transformation, the laser power distribution and focus depth distribution in the detection coordinate system cannot directly correspond to the actual position in the processing coordinate system. The coordinate transformation matrix of this invention... The coordinate transformation residuals of 10 normally clamped samples were all controlled within [a certain range]. Within, satisfy The processing requirements. After deliberately introducing clamping deviations in the 11th and 12th operations, the coordinate transformation residuals were respectively and The system automatically determines to return the kernel, indicating that the residual threshold mechanism can effectively prevent parameter misalignment during processing.

[0196] (v) Verification of production cycle time and equipment utilization rate

[0197] To verify Figure 5 The efficiency improvement effect of multi-device asynchronous scheduling is shown. Comparative Example 1 and Example 1 are used to continuously process 40 crystal ingots. The detection time, handling time, laser modification time, average total cycle time per wafer, equipment utilization rate and output per unit time are statistically analyzed.

[0198] Table 3 shows the comparison results of the two schemes in terms of rhythm.

[0199]

[0200] Table 3 illustrates that in Comparative Example 1, each laser refining device requires waiting for doping concentration detection to complete before processing, resulting in significant pre-processing delays. Furthermore, the dispersed configuration of detection units leads to low utilization rates for individual units. Example 1 employs a shared detection device and four laser refining devices operating asynchronously in parallel. After detection, the ingot is buffered, allowing any laser refining device to quickly call the corresponding parameter package and complete attitude verification when idle, avoiding redundant detection. In an 8-hour continuous test, Comparative Example 1 produced 49 ingots, while Example 1 produced 63 ingots, representing an increase of approximately 28.6%. The average utilization rate of the laser refining devices increased from 77.8% to 93.5%, and the utilization rate of the shared doping concentration detection device increased from 22.1% to 81.6%.

[0201] According to the present invention, the beat ratio is:

[0202] ;

[0203] In the formula, For beat ratio; The average time for laser-induced quality improvement; The average time for doping concentration detection; This represents the average handling time.

[0204] Substitute the average data from Example 1:

[0205] ;

[0206] In the formula, This indicates that one detection device can theoretically be matched with approximately four laser modification devices. This result is consistent with... Figure 1 The layout of one ingot doping concentration detection device matched with four laser refining devices, indicating that the equipment configuration of the present invention has a good match with the actual cycle time.

[0207] (vi) Verification of the uniformity and peeling quality of the modified layer

[0208] To verify the improvement effect of this invention on the quality of laser refining, 20 ingots from Comparative Example 1 and Example 1 were selected for subsequent peeling, and the thickness deviation of the peeled sheet, the continuity of the refining layer, the number of surface thermal damage points, and the peeling success rate were detected. Among them, the continuity of the refining layer was comprehensively evaluated by cross-sectional observation and optical detection after peeling, and the number of thermal damage points was counted as the number of visible abnormal points on the surface of a unit wafer.

[0209] Table 4 shows the comparison results of the processing quality of the two schemes.

[0210]

[0211] As shown in Table 4, the standard deviation of the peel thickness in Example 1 decreased from that in Comparative Example 1. Reduce to The continuity score of the modified layer improved from 86.2 to 94.8, the average number of thermal damage points per wafer decreased from 5.6 to 2.1, and the peeling success rate increased from 92.5% to 98.0%. This indicates that the doping concentration distribution data... Generating a positional laser processing parameter package and executing it accurately in the processing coordinate system after coordinate transformation can reduce local insufficient modification and thermal damage, and improve the uniformity of the modified layer and the peeling quality.

[0212] Further observation revealed that in Comparative Example 1, due to a slight offset between the detection data and the processing position, some ingots in certain areas failed to receive adequate power correction, resulting in local discontinuities in the modified layer. In contrast, Example 1, through the reference feature set... Current benchmark feature set and coordinate transformation matrix Matching the detection coordinates with the processing coordinates ensures that the laser power and focusing depth in different regions can be accurately applied to the corresponding positions, thus significantly improving the continuity of the modified layer.

[0213] (vii) The effect of residual threshold mechanism on intercepting abnormal processing

[0214] To further verify the coordinate transformation residual threshold To enhance its protective function, the experiment included 12 artificially created abnormal clamping scenarios, such as slight ingot misalignment, positioning notch obstruction, end-face edge contamination, and clamping deflection. The system follows... Figure 4 The process shown is used for attitude verification and residual judgment.

[0215] Table 5 presents the results of the abnormal clamping interception test.

[0216]

[0217] As shown in Table 5, the coordinate transformation residuals exceeded the set threshold in all 12 abnormal clamping tests. The system automatically prohibited laser modification processing and returned the ingots to the kernel in all cases, with no erroneous releases. This result demonstrates that the residual threshold judgment mechanism of this invention can provide effective safety verification before cross-device parameter package calls, avoiding the application of laser processing parameters to the wrong location due to abnormal ingot posture.

[0218] Figure 6 This is a comparison chart of the production cycle times of the embodiments and comparative examples of the present invention. The chart can be presented as a bar chart, with the horizontal axis representing Comparative Example 1 and Embodiment 1, and the vertical axis representing the number of ingots produced in 8 hours, the utilization rate of the laser modification equipment, and the utilization rate of the detection device, to visually demonstrate the effect of the present invention in improving continuous production efficiency. Figure 7This is a residual distribution diagram of the cross-equipment coordinate transformation according to the present invention. The diagram can be in the form of a scatter plot or a box plot, with the horizontal axis representing the laser refining equipment number. , , , The vertical axis represents the coordinate transformation residual. And mark the threshold. This is used to demonstrate that the residual of normally clamped samples is below a threshold, and that abnormally clamped samples can be effectively intercepted. Figure 8 This is a comparison chart of the quality of the modified layer and the peeling quality of the present invention. The chart can be presented in bar chart form to show the standard deviation of the peeling sheet thickness, the continuity score of the modified layer, the number of thermal damage points, and the peeling success rate, illustrating that the present invention not only improves production capacity but also maintains or improves processing quality.

[0219] (viii) Experimental Conclusions

[0220] As can be seen from the above specific application examples, this invention can achieve the following effects under laboratory pilot production conditions: First, by matching one ingot doping concentration detection device with four laser refining devices, the detection process and the laser refining process are asynchronous and parallel, significantly improving the overall production output and equipment utilization rate; second, through ingot identification... , benchmark feature set Coordinate transformation matrix and coordinate transformation residuals To ensure accurate mapping between the detection coordinate system and the processing coordinate system after cross-equipment handling, and to avoid misaligned parameter package calls; thirdly, through doping concentration distribution data The system generates positional laser power and focusing depth to achieve matched laser refining energy for regions with different doping concentrations, thereby improving the continuity of the refining layer and the quality of peeling off; fourthly, it uses residual thresholding... Abnormal clamping is automatically intercepted, reducing the risk of erroneous processing. This demonstrates that, compared to traditional single-machine sequential processing methods, this invention not only improves the production efficiency of silicon carbide wafers but also enhances the consistency of cross-device processing and the peeling yield.

[0221] The foregoing description of embodiments of the present invention, through which those skilled in the art are able to implement or use the present invention, will be readily apparent to those skilled in the art. Various modifications to these embodiments will be readily apparent to those skilled in the art. The general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the invention. Therefore, the present invention is not to be limited to the embodiments shown herein, but is to be accorded the widest scope consistent with the principles and novelty disclosed herein.

[0222] 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.

[0223] 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 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0224] 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 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0225] 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 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0226] In a typical configuration, a computing device includes one or more processors (CPU), input / output interfaces, network interfaces, and memory.

[0227] Memory may include non-persistent memory 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.

[0228] 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 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.

Claims

1. A high-efficiency method for producing silicon carbide wafers, characterized in that, The method includes the following steps: S1. Establish a production unit, which includes an ingot doping concentration detection device, an ingot buffer, an ingot handling robot, a central control module, and... Taiwan laser modification equipment, ≥2; S2, the first feeding step Each silicon carbide ingot is given an ingot identification mark. Establish a detection coordinate system in the ingot doping concentration detection device. Doping concentration distribution data were obtained by detection. Extracting the baseline feature set and generate with Bound cross-device laser processing parameter package ;Will , , and Bind the storage and move the silicon carbide ingots that have completed the testing to the ingot cache location; S3, when the When the laser modification equipment is idle, select the first ingot from the ingot buffer that has completed doping concentration detection and is not locked. A silicon carbide ingot, and the silicon carbide ingot and its corresponding... Locked to number Taiwan laser modification equipment, Number the laser-modified equipment; S4, the The first laser modification equipment received the first After reading a silicon carbide ingot , establish the first Processing coordinate system of laser modification equipment Collect the current set of reference features of the silicon carbide ingot in the processing coordinate system. ,according to and Calculate the coordinate transformation matrix and coordinate transformation residuals ; when At that time, Convert to device execution parameter package And perform laser modification processing; when At this time, processing is prohibited, and the product must be returned for verification or repositioning. The maximum allowed coordinate transformation residual threshold; For the first The silicon carbide ingot in the first The equipment execution parameter package in the laser refining equipment.

2. The method according to claim 1, characterized in that, In step S2, the silicon carbide ingot to be processed is loaded and an ingot identification mark is assigned. The ingot handling robot will... A silicon carbide ingot to be processed is transported to an ingot doping concentration detection device, which establishes a detection coordinate system. The doping concentration of the silicon carbide ingot to be processed is detected on the surface to be processed to obtain doping concentration distribution data in the detection coordinate system. And extract the benchmark feature set of the silicon carbide ingot. The central control module determines the doping concentration distribution data based on this data. and benchmark feature set Generate cross-device laser processing parameter packages ; And / or, in step S2, the reference feature set The current reference feature set includes at least two of the following: the outer circumference of a silicon carbide ingot, positioning notch, crystal orientation marker, end face edge feature points, and pre-defined positioning marks; For the first The silicon carbide ingot was transported to the... After the laser modification equipment, by the first The laser-modified equipment acquires the reference feature set in the processing coordinate system. The corresponding feature set; And / or, in step S2, the ingot doping concentration detection device obtains the doping concentration-related detection value using at least one of fluorescence detection, resistivity detection, photoluminescence detection, and infrared transmittance detection. and the detection value related to the doping concentration Converted to doping concentration distribution data : ； In the formula, For the first A silicon carbide ingot at the detection coordinates The detected value of doping concentration at the location; For the first A silicon carbide ingot at the detection coordinates The doping concentration-related detection values ​​obtained at the location; This is the conversion factor between the doping concentration-related detection value and the doping concentration; Calculate the bias amount to convert the doping concentration; and These are two orthogonal coordinates in the detection coordinate system.

3. The method according to claim 2, characterized in that, In step S2, the cross-device laser processing parameter package include: In the formula, For the first A cross-device laser processing parameter package for a silicon carbide ingot; For the first The identification mark of a silicon carbide ingot; For the first Doping concentration distribution data of a silicon carbide ingot in the detection coordinate system; For the first Laser power distribution of a silicon carbide ingot in the detection coordinate system; For the first Laser focusing depth distribution of a silicon carbide ingot in the detection coordinate system; For the first Laser scanning velocity distribution of a silicon carbide ingot in the detection coordinate system; For the first Laser scanning spacing distribution of individual silicon carbide ingots in the detection coordinate system; For the first Laser pulse repetition frequency distribution of a silicon carbide ingot in the detection coordinate system; For the first The set of baseline features for a silicon carbide ingot; This serves as an identifier for the generation time or validity period of cross-device laser processing parameter packages; And / or, in step S4, the device executes the parameter package. Including laser power distribution in the machining coordinate system Laser focusing depth distribution Laser scanning speed distribution Laser scanning spacing distribution and laser pulse repetition frequency distribution And satisfy: ； ； ； In the formula, For the first The silicon carbide ingot in the first Laser power in the processing coordinate system of a laser modification equipment; The silicon carbide ingot is positioned in the detection coordinate system relative to the coordinates. The corresponding laser power; The laser focusing depth in the machining coordinate system; To detect the corresponding laser focusing depth in the coordinate system; The laser scanning speed in the machining coordinate system; To detect the corresponding laser scanning speed in the coordinate system; where, Depend on Coordinate transformation matrix Obtained through conversion; The laser power distribution and laser focusing depth distribution Based on doping concentration distribution data Make corrections: ; In the formula, For the first A silicon carbide ingot at the detection coordinates The corresponding laser power at that location; Reference doping concentration The corresponding reference laser power; This is the laser power correction coefficient corresponding to the change in doping concentration; For the first A silicon carbide ingot at the detection coordinates The detected value of doping concentration at the location; The preset reference doping concentration; For the first A silicon carbide ingot at the detection coordinates The corresponding laser focusing depth; Reference doping concentration The corresponding reference focus depth; This is the focusing depth correction coefficient corresponding to the change in doping concentration.

4. The method according to claim 1, characterized in that, In step S3, the central control module collects real-time data on the equipment status of each laser modification device, the buffer status of the ingot buffer, and the robot status of the ingot handling robot; when the... When the laser refining equipment is idle, the central control module selects the first ingot from the ingot cache that has completed doping concentration detection and is not locked. A silicon carbide ingot, and the silicon carbide ingot and its corresponding cross-device laser processing parameter package. Locked to number A laser-modified equipment; And / or, in step S3, the central control module determines the work assignment evaluation value. Select the silicon carbide ingot from the ingot buffer and assign it to an idle laser refining device: ； In the formula, To make the first The silicon carbide ingots were allocated to the first... Evaluation value of the laser-modified equipment dispatch; For the first The waiting time after the laser modification equipment has been idle; For the first The waiting time of a silicon carbide ingot in the ingot cache bit; For the first The silicon carbide ingot is moved from its current position to the next position. Estimated transport time for the laser-modified equipment; For the first Each silicon carbide ingot corresponds to a cross-device laser processing parameter package. The remaining effective period penalty value; , , and These are non-negative weighting coefficients; the central control module executes first. Larger work assignment combinations.

5. The method according to claim 1, characterized in that, In step S4, the first The first laser modification equipment received the first After processing a silicon carbide ingot, instead of re-performing the doping concentration detection, the ingot's identification identifier is read. , establish the first Processing coordinate system of laser modification equipment Collect the current set of reference features of the silicon carbide ingot in the processing coordinate system. And based on the benchmark feature set Compared with the current benchmark feature set Calculate the coordinate transformation matrix This transforms the coordinates in the detection coordinate system into the machining coordinate system. In the formula, and The first The silicon carbide ingot in the first Two orthogonal coordinates in the processing coordinate system of a laser modification equipment; and The first Two orthogonal coordinates of a silicon carbide ingot in the detection coordinate system; For the first The silicon carbide ingot is transferred from the detection coordinate system to the first... The coordinate transformation matrix of the processing coordinate system of the laser modification equipment; No. The laser-modified equipment or central control module calculates the coordinate transformation residual based on the reference feature points. : ; In the formula, The number of reference feature points used in the calculation of coordinate transformation residuals; The reference feature point number; and The first The coordinates of each reference feature point in the detection coordinate system; and The first The actual acquired coordinates of each reference feature point in the machining coordinate system; and The first The coordinate transformation matrix of the reference feature points Predicted coordinates of the transformed machining coordinate system; And / or, the method further includes step S5: After the laser modification process is completed, the central control module notifies the ingot handling robot to transfer the processed silicon carbide ingot to the unloading position, ingot stripping equipment, or ingot thinning equipment, and releases the first... The task of the laser refining equipment is locked; at the same time, the central control module continues to schedule the ingot doping concentration detection device to perform the doping concentration detection of the next silicon carbide ingot, so that the ingot doping concentration detection and the laser refining processing of multiple laser refining equipment are carried out asynchronously and in parallel.

6. The method according to claim 5, characterized in that, In step S4, when At that time, the first The laser-modified equipment is based on the coordinate transformation matrix. Cross-device laser processing parameter package Convert to device execution parameter package and according to the device execution parameter package For the A silicon carbide ingot undergoes laser modification; when At that time, the central control module prohibits the first A laser modification equipment performs laser modification processing and repositions the silicon carbide ingot to the verification position or reposition it. And / or, in step S4, the coordinate transformation matrix For rigid or approximately rigid transformation matrices: In the formula, For the first The silicon carbide ingot is transferred from the detection coordinate system to the first... The coordinate transformation matrix of the processing coordinate system of the laser modification equipment; This is a coordinate scale correction factor; For the first The silicon carbide ingot was transported to the... The plane rotation angle after the laser modification equipment; To detect the translation amount from the coordinate system to the machining coordinate system in the first coordinate direction; To detect the amount of translation from the coordinate system to the machining coordinate system in the second coordinate direction.

7. A high-efficiency silicon carbide wafer production system, characterized in that, The system is used to implement the method according to any one of claims 1-6, comprising: The ingot feeding module is used to receive silicon carbide ingots to be processed and provide the feeding position; A crystal ingot doping concentration detection device is used to establish a detection coordinate system. The doping concentration of the silicon carbide ingot to be processed is detected on the end face to be processed, and doping concentration distribution data is generated. And extract the benchmark feature set. ; Ingot buffer bit, used to buffer silicon carbide ingots that have completed doping concentration detection but have not yet undergone laser modification; A laser-modified equipment is used to perform the operation according to the equipment's parameter package. Laser modification of silicon carbide ingots; Ingot handling robots are used in ingot feeding modules, ingot doping concentration detection devices, and ingot buffer positions. The laser-modified equipment and the material feeding station are used to transport silicon carbide ingots. The central control module, along with the ingot feeding module, ingot doping concentration detection device, ingot buffer, and ingot handling robot, and... Communication connection of the laser-modified equipment.

8. The high-efficiency silicon carbide wafer production system according to claim 7, characterized in that, The ingot handling robot is one or more of a multi-joint robot, a gantry robot, an RGV transport vehicle, or an AMR mobile robot; the ingot buffer is equipped with a physical buffer, an identification structure, and a positioning structure, the identification structure being used to read the ingot's identification identifier. The positioning structure is used to limit the planar position and end face orientation of the silicon carbide ingot in the buffer state; the production equipment also includes an ingot stripping device and / or an ingot thinning device, and the central control module is a production line MES system or a device central control module that is communicatively connected to the production line MES system.

9. An electronic device, characterized in that, The method includes a processor, a memory, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the computer program, it implements the method according to any one of claims 1 to 6.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by a processor, it implements the method described in any one of claims 1 to 6.

Images