Epitaxial film deposition device control system

Abstract

To provide a technique capable of obtaining a stable quality as a quality of an SiC epitaxial film in an epitaxial film deposition device control system and the like.SOLUTION: A control system 1 creates a recipe of processing of an epitaxial film deposition device 20. The control system 1 comprises a processor and a memory. Processing of the epitaxial film deposition device 20 includes processing for depositing an epitaxial film by using epitaxial growth to a substrate (SiC substrate 24). The processor inputs information on the recipe of first processing immediately after maintaining the epitaxial film deposition device 20 and information of an evaluation value of a quality of the epitaxial film 25 of processing results as initial conditions, constructs or renews a model of processing, and creates a recipe such that the evaluation value of the quality is within an allowable range including a target value, as a recipe for the second and subsequent processing immediately after maintenance, based on the model.SELECTED DRAWING: Figure 1

Description

[Technical Field]

[0001] The present invention relates to a technology for managing an epitaxial film forming apparatus. [Background technology]

[0002] The most important challenges in achieving a sustainable society are the depletion of energy resources and excessive emissions of greenhouse gases such as CO2. For this reason, power conversion devices with excellent energy efficiency and low CO2 emissions are becoming increasingly important. Most power conversion devices consist of a power module in which an insulated gate bipolar transistor (IGBT) as a switching element and a PiN diode (PND) as a rectifying element are connected in parallel. Therefore, reducing losses in semiconductor elements directly contributes to energy savings in power conversion devices. As a loss-reducing technology for semiconductor elements, methods for fabricating elements using 4H-type silicon carbide (4H-SiC, hereafter also referred to as SiC) have attracted attention. To improve the reliability and reduce the cost of SiC elements, high-quality, inexpensive SiC epitaxial substrates (hereafter also referred to as substrates) are required.

[0003] Epitaxial film formation equipment is one type of semiconductor manufacturing equipment. One type of epitaxial film formation equipment for SiC is the SiC epitaxial growth equipment. Epitaxial growth is a thin-film crystal growth technology in which crystals are grown on a base crystal substrate and aligned with the crystal plane of the substrate. SiC epitaxial growth is a technology for forming SiC films on off-cut SiC substrates. Because SiC substrates generally have a high donor concentration, the donor concentration and film thickness must be adjusted for each application's voltage resistance. Therefore, epitaxial growth (or epitaxial film formation) is performed to fabricate SiC devices. Requirements for epitaxial growth technology are diverse, including increasing the diameter of epitaxial growth in response to larger substrate diameters, ensuring uniform donor concentration and epitaxial film thickness, high-speed growth, and low crystal defects.

[0004] One of the systems that evaluate, manage, and control semiconductor manufacturing is the management system for epitaxial film formation equipment. This management system manages the film formation conditions (sometimes called "recipe") that are set in the film formation equipment and are used to control the film formation process. This management system grasps and evaluates the status and results of the film formation process based on the film formation conditions set by the film formation equipment, and creates optimal film formation conditions.

[0005] An example of the prior art related to the above is Japanese Patent Application Laid-Open No. 2020-123675 (Patent Document 1). Patent Document 1 describes a technology for a semiconductor manufacturing equipment management system that determines a recipe according to changes in the semiconductor manufacturing equipment over time. [Prior art documents] [Patent documents]

[0006] [Patent Document 1] Japanese Patent Publication No. 2020-123675 Summary of the Invention [Problem to be solved by the invention]

[0007] However, in conventional semiconductor device manufacturing technology, it was found that the quality of the SiC epitaxial film (sometimes abbreviated as "epi film") produced in the film formation process of the film formation equipment fluctuates due to changes in the film formation equipment over time, maintenance, etc. Therefore, there is a problem in that film formation costs increase in order to obtain a stable epi film.

[0008] In particular, the inventors discovered that even if optimal film formation conditions are set for an epitaxial film formation apparatus, the quality of the resulting epitaxial film fluctuates each time the apparatus undergoes maintenance. Although generally difficult to quantify, variations in the apparatus condition occur before and after maintenance. For example, material associated with film formation accumulates on the walls of the vacuum chamber (also called the chamber) of the apparatus. Immediately after maintenance, the previous film formation conditions may no longer be optimal due to such variations in the apparatus condition. Therefore, the quality of the resulting epitaxial film immediately after maintenance may deviate from the range of predetermined quality conditions defined by manufacturing specifications, etc.

[0009] An object of the present invention is to provide a technology that can obtain stable quality as the quality of an epitaxial film in a management system for an epitaxial film formation apparatus. [Means for solving the problem]

[0010] A representative embodiment of the present invention has the following configuration: A management system for an epitaxial film formation apparatus according to the embodiment is a management system for generating a recipe for a process for the epitaxial film formation apparatus, and includes a processor, wherein the process for the epitaxial film formation apparatus includes a process for forming an epitaxial film by epitaxial growth on a substrate, and the processor constructs or updates a model of the process by inputting, as initial conditions, a recipe for a first process immediately after maintenance of the epitaxial film formation apparatus and information on an evaluation value of the quality of the epitaxial film resulting from the process, and generates, based on the model, recipes for the second and subsequent processes immediately after the maintenance, in which the evaluation value of the quality falls within an allowable range including a target value. [Effects of the Invention]

[0011] According to a representative embodiment of the present invention, a management system for an epitaxial film forming apparatus can achieve stable quality of the epitaxial film. Problems and advantages other than those described above will be described in the "Description of the Preferred Embodiments of the Invention." [Brief explanation of the drawings]

[0012] [Figure 1] 1 shows the configuration of a management system for an epitaxial film formation apparatus according to a first embodiment. [Figure 2] 1 shows the configuration of a management system according to a first modification of the first embodiment. [Figure 3] 10 shows the configuration of a management system according to a second modification of the first embodiment. [Figure 4] 2 shows a detailed configuration example of a management system according to a first embodiment. [Figure 5] In the first embodiment, an example of the configuration of an epitaxial film formation apparatus will be shown. [Figure 6] 1 shows a functional block configuration of a management system according to a first embodiment. [Figure 7] 3 shows a processing flow of the management system according to the first embodiment. [Figure 8] In the first embodiment, an explanatory diagram of the model configuration etc. is shown. [Figure 9] In the first embodiment, an example of the change over time in the recipe and epitaxial quality will be shown. [Figure 10] 10 shows the configuration of a management system according to a second embodiment. [Figure 11] 10 shows a processing flow of a management system according to a second embodiment. [Figure 12] 10 shows a processing flow of a management system according to a third embodiment. [Figure 13] An example of the relationship between the film forming apparatus, chamber, and model in the management system of the fourth embodiment is shown. [Figure 14] 10 shows examples of GUI displays in modifications of the first to fourth embodiments. [Figure 15] 10 shows a continuation of the GUI display example in the modified examples of the first to fourth embodiments. [Figure 16] 10 shows an example of changes over time in recipe and epitaxial quality in a management system of a comparative example compared to the embodiment. DETAILED DESCRIPTION OF THE INVENTION

[0013] Hereinafter, embodiments of the present invention will be described in detail with reference to the drawings. In the drawings, the same parts are generally designated by the same reference numerals, and repeated explanations will be omitted. In the drawings, the depiction of each component may not represent its actual position, size, shape, range, etc., in order to facilitate understanding of the invention, and the present invention is not necessarily limited to the position, size, shape, range, etc., disclosed in the drawings.

[0014] For the purpose of explanation, when describing processing by a program, the program, functions, processing units, etc. may be described as the main components. However, the main hardware components of these are a processor, or a controller, device, computer, system, etc., configured with the processor. A computer executes processing according to a program loaded into memory using resources such as memory and communication interfaces as appropriate, thereby realizing predetermined functions, processing units, etc. The processor may be configured with semiconductor devices such as a CPU or GPU. The processor may be configured with devices or circuits capable of performing predetermined calculations. Processing is not limited to software program processing, but can also be implemented with dedicated circuits. Dedicated circuits such as FPGAs and ASICs can be applied. The program may be installed as data in advance on the target computer, or may be distributed as data from a program source to the target computer and installed. The program source may be a program distribution server on a communication network or a non-transitory computer-readable storage medium. A program may be configured with multiple program modules. A computer system may be configured with multiple devices. A computer system may be configured with a client-server system, a cloud computing system, an IoT system, etc. For the purpose of explanation, various data and information may be described using expressions such as tables and lists, but are not limited to such structures or formats. Data and information for identifying various elements may be described using expressions such as identification information, identifiers, IDs, names, and numbers, but these expressions are interchangeable.

[0015] [Assignments, etc.] The following provides additional information on the aforementioned issues. A management system and management method for semiconductor manufacturing equipment as a comparative example to the embodiment will be outlined below. The management system of the comparative example records and stores processing history information for film formation equipment. The processing history information is, in other words, time-series data on processing, and includes information such as the recipe used for the processing and the evaluation results of the epitaxial quality as a result of the processing. The management system autonomously estimates changes in the chamber over time from the processing history information and determines whether maintenance is required. The recipe is a set of film formation conditions for achieving film formation results close to the target value using a model. The technology of this comparative example estimates changes in the equipment state over time from processing history information including the recipe and epitaxial quality, and therefore does not involve film formation to confirm the changes over time.

[0016] FIG. 16 shows graphs illustrating an example of temporal changes in (a) film formation conditions and (b) epitaxial quality in a comparative example. The horizontal axis represents the film formation period (corresponding time or number of times). For example, from time t1 to time t2, the film formation conditions are set so that the epitaxial quality is maintained at a substantially constant value v1 within the target range V0. The predetermined target range V0 is centered around value V1 and ranges from lower limit V2 to upper limit V3. For example, at time tm1, maintenance of the film formation apparatus was performed. This caused inherent fluctuations in the apparatus condition. As a result, from time t3 to time t4 after the maintenance, the epitaxial quality was set to value v2, which was outside the target range V0, even under the same film formation conditions as before the maintenance. Similarly, at time tm2, maintenance of the film formation apparatus was performed. As a result, from time t5 to time t6 after the maintenance, the epitaxial quality was set to value v3, which was outside the target range V0, even under the same film formation conditions as before the maintenance.

[0017] As described above, in the comparative example, the inherent fluctuation of the equipment state before and after maintenance was not taken into consideration, and the recipe was determined even immediately after maintenance, assuming that the equipment state was the same before and after maintenance. Therefore, as described above, there were cases where the epitaxial quality deviated from the target range V0 for multiple film formations performed immediately after maintenance (for example, each film formation shown by multiple plots from time t3).

[0018] Fluctuations in the state of a film formation apparatus are generally difficult to quantify because they are caused by a variety of factors. In the embodiment, the mechanism does not require quantification of the state of the film formation apparatus, and the generation and adjustment of an optimal recipe before and after maintenance is controlled using an evaluation value of the epitaxial quality of the film formation result as output for the recipe as input of the model.

[0019] <First Embodiment> 1 to 9, a technology of a management system for an epitaxial film formation apparatus according to a first embodiment of the present invention will be described. The management system according to the first embodiment is a system having a function of generating and proposing a recipe, which is an optimal film formation condition adapted to the state of the epitaxial film formation apparatus, and is mainly realized by a computer system. The management method according to the first embodiment is a method having steps executed by the management system according to the first embodiment. The management system according to the first embodiment uses, as an initial condition, the value of the quality of the epitaxial film (sometimes referred to as epi quality) resulting from the epitaxial film formation immediately after maintenance of the epitaxial film formation apparatus, and constructs and updates a model for generating film formation conditions related to epitaxial film formation by machine learning. Note that "immediately after maintenance" refers, in other words, to the time when the first film formation process is performed after maintenance.

[0020] [Management system] FIG. 1 shows an outline of the configuration of a management system for an epitaxial film formation apparatus according to a first embodiment. The management system 1 of the first embodiment is a management system for an epitaxial film formation apparatus 20, and is implemented by a computer system 10. The computer system 10 includes an integrated management unit 11, an apparatus control unit 12, a recipe search unit 13, an analysis and evaluation unit 14, an input device 15A, and an output device 15B, which are interconnected. The input device 15A and the output device 15B may be externally connected devices. A user U1 operates and uses the management system 1 implemented by the computer system 10 via the input device 15A and the output device 15B. The user U1 is, for example, a person who manages the manufacturing process (particularly the film formation process). The computer system 10 is connected to a control unit 21 of the epitaxial film formation apparatus 20 via wired or wireless communication 200.

[0021] The epitaxial film formation apparatus 20 includes a control unit 21 and a chamber 22. The control unit 21 controls the driving of each part of the film formation apparatus. The chamber 22 is composed of, for example, a vacuum chamber, and includes a stage 23 inside the vacuum chamber. The movement of the stage 23, for example, in the horizontal direction (for example, translation and rotation), is controlled. A target substrate, such as a SiC substrate 24, is placed on the stage 23. The film formation apparatus 20 forms an epitaxial film 25 by performing an epitaxial film formation process on the surface of the SiC substrate 24. During this epitaxial film formation process, a predetermined substance is deposited in the chamber 22. The deposition location 26 is, for example, the sidewall surface of the chamber 22. The film formation apparatus 20 includes a sensor 27, which will be described later. The sensor 27 measures predetermined physical quantities, such as the thickness and amount, of the deposit at the deposition location 26. The configuration of the chamber 22 and the deposition method are not limited to this example and may be various.

[0022] The integrated management unit 11 is a unit that inputs, outputs, stores, manages, etc., necessary data and information. The equipment control unit 12 is a unit that controls the operation of the film formation apparatus 20 while communicating with the film formation apparatus 20. The recipe search unit 13 is a unit that performs processing to search for an optimal recipe to be set in the film formation apparatus based on machine learning, etc. The analysis and evaluation unit 14 is a unit that analyzes and evaluates the results of film formation using the recipe and obtains evaluation values, etc., of the quality of the epitaxial film. The analysis and evaluation unit 14 or user U1 obtains values, such as the film thickness, impurity concentration, and crystal defect density of the epitaxial film, as a result of the analysis and evaluation.

[0023] [Management System - Variant (1)] FIG. 2 shows an example implementation of the management system 1 according to a modification of the first embodiment. In the implementation example of FIG. 2, the same components as those in FIG. 1 are connected to each other via a LAN 201. In the configuration example of FIG. 2, each unit, such as the integrated management unit 11, is implemented as a computer system. That is, the system includes a computer system CS1 for the integrated management unit 11, a computer system CS2 for the device control unit 12, a computer system CS3 for the recipe search unit 13, and a computer system CS4 for the analysis and evaluation unit 14, all of which are connected to each other via the LAN 201. Each computer system may be configured by connecting multiple computers, or may be a machine in which, for example, multiple GPUs are connected in parallel. In the configuration example of FIG. 2, parallel distributed processing using multiple computer systems enables rapid processing.

[0024] [Management System - Variant (2)] FIG. 3 shows an implementation example of the management system 1 according to another variation of the first embodiment. In the implementation example of FIG. 3, as part of the same components as those in FIG. 1, a recipe search unit 13 is provided in the form of a server or the like on a cloud computing system 202 via a wide-area communication network. A computer system CS5 including the integrated management unit 11 and the like is connected to the recipe search unit 13, which is a server or the like on the cloud computing system 202 via the wide-area communication network. The processing of the recipe search unit 13 involves machine learning and the like, and may therefore be relatively heavy, requiring high performance and a large amount of computational resources. Therefore, in this implementation example, implementing the recipe search unit 13 on the cloud computing system 202 enables rapid processing. Furthermore, in this implementation example, one recipe search unit 13 can be shared among multiple computer systems associated with multiple film forming apparatuses.

[0025] [Management System - Details] FIG. 4 shows a detailed configuration example of the management system 1 of the first embodiment of FIG. 1. In the configuration of FIG. 4, each unit, such as the integrated management unit 11, is composed of a computer or circuit having a processor 116, a ROM 117, a RAM 118, an interface 115, and buses connecting them to each other. The functions of each unit may be realized by processing using a computer program or may be implemented by a dedicated circuit such as an FPGA. The configuration of FIG. 4 also includes a controller 211 connected to the device control unit 12 and a controller 212 connected to the analysis and evaluation unit 14, and these controllers correspond to control units or operation units through which a user U1 inputs settings and instructions.

[0026] [Epitaxial film formation equipment] Figure 5 shows an example of the implementation of the epitaxial deposition apparatus 20 of Figure 1. Figure 5(A) shows an example of the implementation in the form of a cluster apparatus. This cluster apparatus includes a substrate analysis chamber 22A, a cleaning chamber 22B, an analysis and evaluation chamber 22C, a processing chamber 22D, a regeneration chamber 22E, and a load lock chamber / transfer chamber 22F.

[0027] 5B shows an example of an implementation configuration in the form of a SiC epitaxial device. This SiC epitaxial device is a SiC epitaxial cluster device equipped with analysis, reclamation, and growth chambers, and includes a substrate analysis chamber 22a, an epitaxial analysis chamber 22b, a defect analysis chamber 22c, an epitaxial growth chamber 22d, a CMP chamber 22e, and a load lock chamber / transfer chamber 22f.

[0028] The management system 1 of the first embodiment manages semiconductor manufacturing equipment including an epitaxial film formation apparatus 20. The management system 1 also requires, as its components, an analysis and evaluation device for substrates and epitaxial films, or a means for acquiring analysis and evaluation result data from the analysis and evaluation device. The analysis and evaluation device or means corresponds to the analysis and evaluation unit 14 in FIG. 1, the analysis and evaluation chamber 22C in FIG. 5A, and the epitaxial analysis chamber 22b in FIG. 5B. The necessary components may be integrated into a single device as a cluster device, as in the example of FIG. 5A, or may be a SiC epitaxial device, as in FIG. 5B. Alternatively, the management system 1 may be provided with substrate analysis chambers 22A and 22a, and the management system 1 may automatically acquire and use substrate information from the substrate analysis chambers, as in the third embodiment described below.

[0029] The substrate analysis chambers 22A and 22a are chambers for analyzing and evaluating the substrate (SiC substrate 24 in FIG. 1) on which a film is to be formed, and as a result, substrate information described below is obtained. The processing chamber 22D is a chamber for processing the substrate. The epitaxial growth chamber 22d is an example of the processing chamber 22D and is a chamber for performing epitaxial growth film formation processing on the SiC substrate. The analysis and evaluation chamber 22C is a chamber for analyzing and evaluating the substrate processed in the processing chamber 22D. The epitaxial analysis chamber 22b is an example of the analysis and evaluation chamber 22C and is a chamber for evaluating, for example, the film thickness and impurity concentration of the epitaxial film 25 in FIG. 1. The defect analysis chamber 22c is an example of the analysis and evaluation chamber 22C and is a chamber for analyzing and evaluating defects in the epitaxial film 25.

[0030] The management system 1 may also be provided with a regeneration chamber 22E or a CMP chamber 22e. The regeneration chamber 22E is a chamber for regenerating substrates processed in the processing chamber 22D. The CMP chamber 22e is an example of the regeneration chamber 22E and is a chamber for regenerating substrates by CMP (chemical mechanical polishing). This allows the substrate (SiC substrate 24 in FIG. 1) to be reused by CMP or the like even if the desired epitaxial results are not obtained as a result of the processing (for example, if the epitaxial quality of the processing result using the generated recipe does not satisfy the tolerance range described below).

[0031] The substrate analysis and evaluation device includes, for example, a substrate shape measurement device and a defect and surface roughness evaluation device. The substrate shape measurement device measures the substrate shape, such as warpage, wafer edge shape, and plate thickness. The evaluation device evaluates surface defects, internal defects, and surface roughness using, for example, X-rays, PL light, CL, radiation topo, laser, or a microscope.

[0032] The epitaxial film analysis and evaluation device includes, for example, an epitaxial concentration evaluation device, an epitaxial film thickness evaluation device, and a defect evaluation device. The defect evaluation device evaluates defects using, for example, X-rays, PL light, CL, radiation topo, laser, or a microscope.

[0033] [Device status change] In the SiC epitaxial film formation apparatus, which is the epitaxial film formation apparatus 20 of FIG. 1, the following material deposition occurs as a result of film formation, as an example of changes in the state of the apparatus. During SiC epitaxial growth, by-products derived from the source gas firmly adhere to components other than the SiC substrate 24 (e.g., the inner wall of the chamber 22 and the susceptor). These by-products (i.e., deposits) are exposed to high temperatures during epitaxial growth and evaporate, causing deterioration in epitaxial quality over time. In other types of semiconductor manufacturing equipment, such as CVD equipment, by-products can be removed relatively easily by gas cleaning or other methods. However, effective gas cleaning or other methods have not yet been established for SiC epitaxial film formation equipment. As a result, frequent maintenance, such as opening the chamber, is required to remove the by-products, which increases costs.

[0034] Furthermore, in a SiC epitaxial film formation apparatus, as described above, maintenance of the chamber or the like can result in an inherent change in the state of the apparatus, and even if the same recipe is applied before and after maintenance, the epitaxial quality of the film formation result may change and no longer satisfy the tolerance range. Therefore, the management system of the first embodiment finds an optimal recipe for the second and subsequent processes based on the input information of the first process immediately after maintenance. In other words, the management system updates the model so that the quality of the epitaxial film resulting from the process using the estimated recipe becomes an optimal recipe that is as close as possible to the target value within the tolerance range.

[0035] [Management system - Functions] The main functions and processing overview of the management system 1 are as follows. The processing overview is also shown in the flow chart of Figure 7, which will be described later. The management system 1 inputs the first recipe and epitaxial quality (analysis evaluation result values are used for quality) immediately after maintenance of the film formation apparatus 20, compares the epitaxial quality estimated using a model of the conventional technology example or a model before maintenance (f(x) in Figure 8, which will be described later) with the latest epitaxial quality input above, and calculates a compensation coefficient (A in Figure 8). The management system 1 uses the calculated compensation coefficient to update the model to a new model (F(x) in Figure 8). This model is a machine learning model (described later (Figure 8)) for recipe generation and estimation. In other words, this model update is an update to compensate for changes in the apparatus state before and after maintenance.

[0036] The management system 1 uses the updated new model to estimate an optimal recipe for the target SiC substrate 24, and estimates the epitaxial quality of the processing result for each estimated candidate recipe. The management system 1 generates an optimal recipe that ensures that the estimated epitaxial quality satisfies an acceptable range. The management system uses the generated recipe to perform film formation from the second onwards immediately after maintenance, and stores data such as the recipe and epitaxial quality at the time of implementation as history (in other words, processing history information) in a database (DB).

[0037] Furthermore, the management system 1 of the first embodiment may perform the following control when the epitaxial quality estimated from the optimal recipe does not satisfy the tolerance range. That is, the management system 1 notifies the user U1 of information about the optimal recipe calculated at that time via a graphical user interface (GUI) and informs the user U1 that the recipe does not satisfy the tolerance range. The user U1 confirms this notification, determines whether or not to allow the film formation using the recipe, and makes an input according to the decision. Even if the recipe does not satisfy the tolerance range, if an input allowing the execution is made, the film formation using the recipe is allowed. If an input not allowing the execution is made, it is possible to take measures such as revising the set value of the tolerance range. This function corresponds to steps S105 and S106 in the flow of FIG. 7 described below.

[0038] Furthermore, even if the epitaxial quality of the film formation result according to the recipe does not satisfy the tolerance range in each film formation, the management system notifies the user U1 via the GUI and terminates the flow in response to the input of the user U1's decision. This function corresponds to steps S111 and S112 in the flow of FIG. 7, which will be described later.

[0039] [Management system - functional block configuration] Fig. 6 shows an example of a functional block configuration of the management system 1 according to the first embodiment of Fig. 1. The computer system 10 in Fig. 6 includes a central processing unit 104, a DB 105, an input device 15A, an output device 15B, a model configuration unit 107, a recipe estimation unit 108, an equipment control unit 110, a process processing unit 111, an analysis and evaluation unit 112, and a convergence determination unit 113.

[0040] The input information 101 via the input device 15A includes target values, tolerance ranges, analytical evaluation results, the first recipe immediately after maintenance, and epitaxial quality, etc. The epitaxial quality is an evaluation value of concentration, etc. The target value is a target value for epitaxial quality. The tolerance range is a range determined according to specifications, etc., using the target values. The analytical evaluation results are information that can be acquired from the analytical evaluation device. The information on the first recipe and epitaxial quality immediately after maintenance is the recipe applied in the first film formation immediately after the maintenance, and the evaluation value of the epitaxial quality of the film formation result.

[0041] The user U1 can operate the input device 15A (for example, a keyboard or a mouse) to input each piece of information and make it part of the input information 101. Alternatively, the integrated management unit 11 of the management system 1 may automatically acquire and input input information that can be acquired automatically.

[0042] The output information 102 via the output device 15B includes the optimum recipe, analysis and evaluation results, etc. The optimum recipe is set in the epitaxial film formation apparatus 20 (also referred to as the film formation apparatus 20) as a recipe to be applied from the second time onwards immediately after maintenance. In other words, the recipe set in the film formation apparatus 20 is updated to the new recipe.

[0043] The user U1 can check the contents of the output information 102 on the display screen of a display device that is one of the output devices 15B. The user U1 can give instructions, make settings, and input the input information 101 according to the GUI on the display screen.

[0044] The management system 1 includes a model configuration unit 107, a recipe estimation unit 108, an equipment control unit 110, a process processing unit 111, an analysis and evaluation unit 112, and a convergence determination unit 113, which are implemented based on program processing by a central processing unit 104. The central processing unit 104 is configured with a processor, etc., as shown in FIG. 4, and performs processing while appropriately using resources such as memory and a communication interface. Control programs and the like are stored in a ROM or a secondary storage device (not shown). The DB 105 can be configured with a memory, a secondary storage device, a DB server, or the like. The central processing unit 104 performs processing while reading and writing various data and information stored in the memory, the DB 105, etc. The DB 105 organizes and stores various data and information, such as input information 101, output information 102, related information, and setting information. The data and information being processed is stored in a memory, such as a RAM.

[0045] The model configuration unit 107 configures and manages a machine learning model. The recipe estimation unit 108 estimates a recipe based on the model. These units (107, 108) correspond to the recipe search unit 13 in FIG. 1. The equipment control unit 110 controls the film formation equipment 20. The process processing unit 111 performs processing related to the film formation process in the film formation equipment 20. These units (110, 111) correspond to the equipment control unit 12 in FIG. 1. The analysis and evaluation unit 112 corresponds to the analysis and evaluation unit 14 in FIG. 1.

[0046] [Processing flow (1)] 7 shows a flow of main processing by the management system 1 (particularly the computer system 10, the central processing unit 104, the processor, etc.) of the first embodiment. This flow has steps S101 to S112. In step S101, the management system 1 inputs, as input information 101, information such as the first recipe immediately after maintenance and the epitaxial quality (corresponding evaluation value) of the film formation result using that recipe.

[0047] In step S102, the management system 1 causes the recipe estimation unit 108 to calculate compensation coefficients for a model for recipe generation (FIG. 8, described later) in the model configuration unit 107.

[0048] In step S103, the management system 1 inputs information on a desired target value and an allowable range as part of the input information 101. For example, the user U1 sets the target value and the allowable range via the input device 15A and the output device 15B.

[0049] In step S104, the management system 1 generates an optimal recipe for achieving an epitaxial quality value as close as possible to the target value within the tolerance range using the model compensated using the compensation coefficient in step S102, and estimates the epitaxial quality value resulting from epitaxial film formation based on this recipe. This estimation of the epitaxial quality can be realized, for example, by simulation.

[0050] In step S105, the management system 1 determines and confirms whether the epitaxial quality value resulting from the above estimation satisfies the tolerance range. If it does (Y), the process proceeds to step S107; if it does not (N), the process proceeds to step S106.

[0051] In step S106, the management system 1 notifies the user U1 through the GUI of the output device 15B that the recipe generated in step S104 does not satisfy the allowable range (specified in step S103). Then, the management system 1 receives an input from the user U1 to confirm and determine whether or not to allow the film formation to be performed using that recipe. If an input indicating that the film formation using that recipe is allowed is entered (Y), the process proceeds to step S107; if an input indicating that the film formation is not allowed is entered (N), the process returns to step S103. In step S103, the setting of the allowable range is reviewed.

[0052] In step S107, the management system 1 causes the process processing unit 111 to cause the film forming apparatus 20 to perform the epitaxial film forming process according to the optimum recipe. Then, the management system 1 inputs information about the recipe used at that time into the DB 105 and stores it as part of the history.

[0053] In step S108, the management system 1 causes the analysis and evaluation unit 112 to analyze and evaluate the epitaxial quality of the film formation process result according to the recipe, and creates analysis and evaluation result information (in other words, processing result information) including the epitaxial quality.

[0054] In step S109, the management system 1 receives and acquires the processing result information by the central processing unit 104. Alternatively, the user U1 may input the processing result information through the input device 15A. The central processing unit 104 associates the processing result information with information on the recipe used for the processing and stores it in the DB 105 as part of the history.

[0055] In step S110, the management system 1 uses the updated information of the DB 105 in the model configuration unit 107 to update the model for recipe generation.

[0056] In step S111, the management system 1 determines and confirms whether the epitaxial quality of the film formation result using the above-mentioned optimal recipe satisfies the tolerance range, and if it does (Y), the flow ends, and if it does not (N), the flow proceeds to step S112.

[0057] In step S112, the management system 1 notifies the user U1 via the GUI of the output device 15B that the recipe used in that film formation process did not satisfy the tolerance range. The management system 1 then receives an input from the user U1 to confirm and decide whether or not to end the process with that film formation result. If an input indicating that the process should be ended with that film formation result is made (Y), the flow ends, and if an input indicating that the process should not be ended is made (N), the process returns to step S104, for example. In step S104, the recipe is reviewed. Thereafter, the same process is repeated.

[0058] Based on the above flow, epitaxial film formation is performed using an optimal or suitable recipe for each subsequent maintenance immediately after maintenance, thereby stably achieving optimal or suitable epitaxial quality. In the example of the above flow, the management system 1 can collectively estimate and propose the above-mentioned optimal recipe for multiple maintenances from the second maintenance onwards. Alternatively, the management system 1 may estimate and propose the above-mentioned optimal recipe for each maintenance from the second maintenance onwards. The management system 1 may also estimate and propose the above-mentioned recipe for each set number of maintenances.

[0059] [Model] FIG. 8 is an explanatory diagram of a method for updating a model using machine learning in the first embodiment. A method for constructing a model for generating and estimating a recipe will be described below. (a) shows an overview of the model for a recipe. In the machine learning model, input information includes the recipe and epitaxial quality for each film formation, and output information includes an estimated result of the optimal recipe and epitaxial quality. In the first embodiment, the model input particularly includes the recipe and epitaxial quality for the first film formation immediately after maintenance, and the model output includes an estimated optimal recipe and epitaxial quality to be applied to the second and subsequent film formations immediately after maintenance. "Optimal" means that the estimated epitaxial quality satisfies the tolerance range and is as close as possible to the target value.

[0060] (b) shows an overview of the model update. Using a function expression, the previous or previous model is represented as f(x), and the new model after the update is represented as F(x). As shown in the figure, the new model F(x) is derived as F(x) = A * f(x) + W0. A is the compensation coefficient. W0 is the state of the equipment immediately after maintenance.

[0061] (c) shows a model updating method using machine learning. In this method, an optimal recipe is generated based on a model, taking into account the state (W0) of the film forming apparatus 20 immediately after maintenance. In this method, the concentration distribution C tThe latent variable representing the device state is X t The subscript t represents time.

[0062] Concentration distribution C, which is the output according to the device state t is written as shown in the figure. Equation 1 is t =g(X t , θ)+δ, where θ is a value depending on the recipe, substrate information (described later), etc. δ is the variation. Equation 2 is t =f1(X t-1 ,Z t-1 )+ε. Z t-1 is a value depending on whether maintenance is performed, the thickness of the deposit, the elapsed time, etc. ε is the variation. g(X t ,θ) and f1(X t-1 ,Z t-1 ) are functions. The functional forms of these functions are unknown. Therefore, the management system 1 derives the functional forms of these functions by performing regression analysis.

[0063] The concentration distribution C from the second time onward after the maintenance, excluding the first time immediately after the maintenance of the film forming apparatus 20, is expressed by the following formula: t +C0. Equation 4 is A=f2(C0,C -1 ) C0 is the first measured concentration distribution value immediately after maintenance. C -1 is the actual measured value of the concentration distribution before maintenance. As mentioned above, the compensation coefficient A is calculated by multiplying the actual measured value of the concentration distribution (C0, C -1 Therefore, the concentration distribution C immediately after maintenance can be derived from the concentration distribution C t It can be derived using

[0064] Regarding the details of the recipe generation or machine learning method, known methods such as neural networks and support vector machines can be applied, for example, methods such as RNN (Recurrent Neural Network) and LSTM (Long-Short Term Memory) can be applied.

[0065] [Recipe Adjustment] FIG. 9 is an explanatory diagram showing a specific example of epitaxial quality fluctuations and recipe adjustments in response to the execution of each epitaxial film formation and appropriate maintenance in the film formation apparatus 20 in the management system 1 of the first embodiment. The graph in FIG. 9 shows the film formation time (corresponding time point or number of times) on the horizontal axis, and shows the film formation conditions (recipes) of (a) and the time change in epitaxial quality (b). In other words, the contents of FIG. 9 show an example of adjusting the recipe to an optimum one immediately after maintenance in response to fluctuations in the apparatus state and epitaxial quality due to maintenance. The recipe in (a) shows one parameter value (gas flow rate as an example, but not limited to) of the film formation conditions applied to the film formation process for SiC epitaxial growth. The epitaxial quality in (b) is an evaluation value of the quality of the SiC epitaxial film (epitaxial film 25 in FIG. 1) resulting from the film formation process using the recipe in (a), and is, for example, the concentration distribution C in FIG. t is the value.

[0066] The allowable range 900 is a range having a lower limit V2 and an upper limit V3 centered around a target value V1 of epitaxial quality. Furthermore, time tm1 and the like indicate examples of dates and times when maintenance was performed on the film forming apparatus 20. An example of such maintenance is adjusting the condition of the chamber 22 by removing deposits from the chamber 22, as described above.

[0067] For example, at time t1 (first time), film formation is performed using a recipe value of p1, resulting in an epitaxial quality value of v1, which falls within the tolerance range of 900. At time t2 (second time), film formation is performed using an adjusted recipe value of p2, resulting in an epitaxial quality value of v2, which falls within the tolerance range of 900 and is close to the target value V1. From time t2 to time t10 (tenth time), film formation is performed while the recipe value is finely adjusted as appropriate, resulting in the epitaxial quality of each time being maintained at a value close to the target value V1, as in v2, within the tolerance range of 900.

[0068] Here, after time t10 (the 10th time in total), at time tm1, maintenance is performed on the film forming apparatus 20. Immediately after this maintenance, time t11 corresponds to the first film formation immediately after the maintenance (the 11th time in total), time t12 corresponds to the second film formation immediately after the maintenance (the 12th time in total), and so on.

[0069] In the first deposition run immediately after maintenance at time t11, recipe value p3 is the same as the recipe value at time t10 immediately before maintenance, and as a result of this deposition, the epitaxial quality value fluctuates to value v3, as indicated by the black circle. This value v3 is below the lower limit V2 and falls outside the allowable range 900.

[0070] 7 based on the recipe value p3 and epitaxial quality value v3 at time t11 (the first time immediately after maintenance), the management system 1 proposes changing the recipe value p3 to an optimal value p4 estimated based on the new model at the next time t12 (the second time immediately after maintenance), and performs adjustment 901. As a result of this adjustment 901, the epitaxial quality value resulting from film formation using the recipe value p4 at time t12 becomes a value v2 that is close to the target value V1 within the tolerance range 900.

[0071] The recipe is adjusted in the same way for each run after time t13. For example, at time t12, the recipes for each run after time t13 are also proposed together, and are adjusted so that the value increases slightly each time. During the period from time t12 to time t20, the epitaxial quality value is maintained at value v2 within the tolerance range of 900 as a result of each run.

[0072] Furthermore, after time t20, at time tm2, maintenance of the film forming apparatus 20 is performed. As a result, at time t21 (the 21st time in total) immediately after the first maintenance, the epitaxial quality of the film formation result at recipe value p5 is v4. Value v4 exceeds upper limit value V3 and falls outside the allowable range 900. Therefore, the management system 1 performs adjustment 902, similar to the previous adjustment 901. In this adjustment 902, the recipe value p5 at time t21, the first time immediately after the maintenance, is changed to recipe value p6 at time t22, the second time immediately after the maintenance. From time t22 onwards, the recipe is fine-tuned and the epitaxial quality of the film formation result at each maintenance is maintained at value v2. Thereafter, the recipe is similarly adjusted according to the epitaxial quality immediately after the maintenance. Note that the example of FIG. 9 illustrates an example of adjustment of one parameter constituting the recipe and one evaluation value of the epitaxial quality, but this is not limiting and the same can be applied to multiple other parameters.

[0073] [Effects etc. (1)] As described above, the management system for the epitaxial film formation apparatus according to the first embodiment can provide stable quality for the SiC epitaxial film. According to the first embodiment, the user's effort in creating and setting recipes can be reduced, thereby improving the efficiency of the semiconductor device manufacturing process. According to the first embodiment, as shown in the example of FIG. 9 , the recipe to be applied from the second onward is adjusted to be an optimal recipe depending on the epitaxial quality of the result of the first film formation immediately after maintenance. This allows the epitaxial quality to be quickly converged to and maintained at an appropriate value within the allowable range in each subsequent film formation except for the first film formation after maintenance.

[0074] <Embodiment 2> 10 and 11, a management system for an epitaxial film formation apparatus according to the second embodiment will be described. The basic configuration of the second embodiment is the same as that of the first embodiment, and the following mainly describes the components of the second embodiment that are different from those of the first embodiment. The management system according to the second embodiment has a function of issuing a maintenance notification when the epitaxial quality of a recipe estimated to be applied to the film formation process immediately after maintenance does not satisfy the allowable range.

[0075] [Management System (2)] 10 shows a functional block configuration of the computer system 10 in the management system 1 of the second embodiment. In addition to the components of the first embodiment shown in FIG. 6, this configuration includes a maintenance determination unit 109 and a maintenance effect evaluation unit 106 as processing units related to the maintenance notification function. In addition to the information described above, the output information 102 includes a maintenance notification and a maintenance evaluation result.

[0076] If the epitaxial quality of the recipe estimated by the recipe estimation unit 108 to be applied to the film formation process immediately after maintenance does not satisfy the tolerance range, the management system 1 inputs information indicating "maintenance required" from the maintenance determination unit 109 to the central processing unit 104. This information indicates that maintenance of the film formation apparatus 20 is required to satisfy the tolerance range. Upon receiving this input, the central processing unit 104 proceeds to a maintenance notification step (see FIG. 11, which will be described later). The central processing unit 104 causes the recipe estimation unit 108 to start generating an optimal maintenance method required to generate a suitable recipe. The recipe estimation unit 108 generates the optimal maintenance method based on the model and returns the information to the central processing unit 104. The central processing unit 104 notifies the user U1 of the optimal maintenance method, etc., via the GUI.

[0077] There are multiple types of maintenance methods for the film forming apparatus 20. For example, assume that there are three types of methods, Methods A, B, and C. The recipe estimation unit 108 selects the optimal maintenance method required to generate a suitable recipe (i.e., a recipe in which the epitaxial quality satisfies an allowable range) from the multiple candidate maintenance methods.

[0078] A maintenance method can be generated, for example, as follows: The management system 1 manages the processing history information in the DB 105 so that it includes maintenance history information including information such as the date and time of maintenance implementation. When the maintenance determination unit 109 determines that maintenance is required, the management system 1 determines and derives, based on a model including the maintenance history information, which of multiple candidate maintenance methods is the most suitable maintenance method, as well as detailed parameter values for the maintenance method.

[0079] In the second embodiment, the effect of maintenance performed on the film forming apparatus 20 is quantitatively evaluated. The maintenance effect evaluation unit 106 quantitatively evaluates the effect of the maintenance and inputs the maintenance effect evaluation results to the central processing unit 104. The maintenance effect can be evaluated, for example, by measuring the amount and thickness of by-products or deposits on the wall inside the chamber 22 in FIG. 1 . For example, the sensor 27 in FIG. 1 measures a predetermined physical quantity, such as the amount and thickness of deposits at the deposition location 26. The management system 1 acquires this measurement value and uses it as an evaluation value for the maintenance effect. Examples of the sensor 27 include a film thickness sensor and an optical film thickness evaluation device. The user U1 may input the measurement results to the management system 1. The management system 1 may automatically acquire and input the signal from the sensor 27. The management system 1 may monitor and record the signal from the sensor 27 over time.

[0080] As in the first embodiment, the second embodiment also has a function (step S106 in FIG. 7) that allows the user U1 to execute processing at his / her discretion even when the optimum recipe does not satisfy the allowable range.

[0081] [Processing flow (2)] Fig. 11 shows a processing flow of the management system 1 in the second embodiment. The flow in Fig. 11 adds step S200 relating to a maintenance notification function to the flow in Fig. 7. If the estimated epi quality according to the estimated recipe does not satisfy the allowable range in step S105 described above (N), the process proceeds to step S200. Step S200 includes steps S201 to S204.

[0082] In step S201, the management system 1 uses the model in the model construction unit 107, via the recipe estimation unit 108, to generate the maintenance method (in other words, the maintenance content, etc.) required to generate a recipe for achieving an epitaxial quality value close to the target value.

[0083] In step S202, the management system 1 notifies the user U1 via the GUI to perform maintenance using the above-mentioned maintenance method, and the user U1 performs maintenance on the film forming apparatus 20. The maintenance effect evaluation unit 106 evaluates the maintenance effect of the maintenance implementation result and sets it as maintenance effect evaluation result information. Alternatively, the user U1 evaluates the maintenance effect and inputs the maintenance effect evaluation result information.

[0084] In step S203, the central processing unit 104 acquires the maintenance effect evaluation result information, associates the maintenance implementation details (including the date and time, method, etc.) with the maintenance effect evaluation result, and stores them in the DB 105 as part of the maintenance history information.

[0085] In step S204, the management system 1 updates the model using the updated information in the DB 105 via the model configuration unit 107. After step S204, the flow ends.

[0086] [Effects etc. (2)] As described above, according to the second embodiment, when the recipe without maintenance does not satisfy the tolerance range, a suitable maintenance method is notified and performed to obtain a suitable recipe that satisfies the tolerance range. As a result, in the first film formation immediately after the maintenance, the suitable recipe is applied, resulting in good epitaxial quality that satisfies the tolerance range. According to the second embodiment, the efficiency of the maintenance work can also be improved in response to the maintenance notification.

[0087] The effects will be explained further using FIG. 9 . In the first embodiment, for example, it is assumed that time points tm1, tm2, etc. were periodic maintenance events. In contrast, in the second embodiment, it is assumed that no such periodic maintenance events are performed, or that additional maintenance events are appropriately performed between periodic maintenance events. For example, after time point t10, the management system 1 determines that the epitaxial quality of the film deposition at the next time point t11 will not meet the acceptable range, assuming that no maintenance events are performed. Therefore, the management system 1 generates an appropriate recipe and maintenance method, taking into account the maintenance events and their results, as described above, and notifies the user U1. As a result, maintenance events are performed between time points t10 and t11. As a result, the epitaxial quality of the first film deposition at time point t11 immediately after the maintenance event can be within the acceptable range.

[0088] <Third Embodiment> A management system for an epitaxial film formation apparatus according to a third embodiment will be described with reference to FIG. 12. The third embodiment is a configuration in which functions are added to the second embodiment. The management system according to the third embodiment inputs information (referred to as substrate information) about a substrate (SiC substrate 24 in FIG. 1) that is the target of the film formation process, and generates an optimal recipe by taking the substrate information into consideration. After placing a substrate (e.g., a semiconductor wafer) on stage 23 in chamber 22, management system 1 first evaluates the substrate and obtains substrate evaluation result information.

[0089] [substrate] Generally, SiC substrates have more defects such as dislocations than Si substrates. For example, dislocation density affects the defect density after epitaxial film formation. Therefore, when using defect density as an output (the output of the above-mentioned model, an evaluation value of epitaxial quality), information on defects in the substrate must be taken into consideration. Furthermore, warpage of the substrate also affects epitaxial quality. Therefore, in the third embodiment, substrate information including such information is taken into account to generate an optimal recipe with high accuracy. In the third embodiment, substrate information is included in the input information of the training data for machine learning. The management system 1 generates an optimal recipe so that the epitaxial quality of the film formation result using the target substrate is as close as possible to the target value within an acceptable range.

[0090] When generating the recipe, the board information, which is one of the inputs, must be constrained to satisfy pre-specified board specifications (i.e., board quality). For example, if the target board has many defects, the estimated optimal recipe may not satisfy the tolerance range. For this reason, the management system 1 calculates and determines the maximum number of defects (i.e., defect density) required for the target board to generate the optimal recipe. For example, the management system 1 gradually reduces the number of defects and calculates whether a recipe that satisfies the tolerance range can be generated under each defect number condition. The management system 1 then notifies the user U1 of the defect number and defect density conditions (i.e., board specification conditions) that make it possible to generate the recipe via the GUI. The management system 1 then notifies the user U1 whether the installed target board satisfies the conditions. If the target board does not satisfy the conditions, the management system 1 then prompts the user U1 via the GUI to confirm and decide whether to change the target board to one that satisfies the conditions. The user U1 then decides and inputs whether to change the board or not. If the user U1 decides not to change the board, the management system 1 transitions to the maintenance execution flow.

[0091] [Management System (3)] The functional block configuration of the management system 1 in embodiment 3 is similar to the configuration of embodiment 2 in Figure 10, with the difference being that it includes substrate information as one of the pieces of input information 101, and information regarding substrate specifications such as the above-mentioned defect density and substrate change notifications as one of the pieces of output information 102.

[0092] [Processing flow (3)] FIG. 12 shows a processing flow of the management system 1 in the third embodiment. The flow in FIG. 12 adds a processing step related to substrate information to the flows in FIGS. 7 and 11. Step S301 is provided between step S103 and step S104 in FIG. 7. In step S301, the management system 1 inputs and acquires substrate information related to the target substrate (SiC substrate 24) on the stage 23 and registers it in DB 105. At this time, user U1 may input the substrate information via a GUI, or the management system 1 may acquire the substrate information from another device.

[0093] The processing content of step S104 is partially different from that described above (referred to as S104c). In step S104c, the management system 1 uses the above-described model and the above-described substrate information to generate an optimal recipe and estimate the corresponding epitaxial quality.

[0094] The processing content of step S107 is partially different from that described above (referred to as S107c). In step S107c, the management system 1 inputs the recipe and the substrate information to the BD 105 in association with each other.

[0095] In the third embodiment, if the tolerance is not satisfied in step S105 (N), the process proceeds to step S302. In step S302, the management system 1 uses the model of the model configuration unit 107, via the recipe estimation unit 108, to derive substrate specifications, such as the above-mentioned defect density, required to generate a recipe that achieves an epitaxial quality value close to the target value within the tolerance.

[0096] Next, in step S303, the management system 1 determines whether the target substrate on the stage 23 satisfies the substrate specification conditions based on the substrate specifications and conditions and the substrate information. If not, the management system 1 determines whether to change the target substrate on the stage 23 to another substrate that satisfies the substrate specifications and conditions. At this time, the management system 1 notifies the user U1 via the GUI of information such as the defect density of the target substrate and the substrate specifications and conditions, and accepts a decision and input regarding whether to change to another substrate if the target substrate does not satisfy the conditions. The user U1 confirms the notification and inputs a decision regarding whether to change the substrate. If the substrate is to be changed (Y), the process returns to step S301. In that case, in step S301, substrate information regarding the changed substrate is input. If the substrate is not to be changed (N), the process proceeds to step S200 related to the maintenance notification described above, and similar processing is performed.

[0097] [Board and model] In the third embodiment, as described above, a plurality of substrates are handled as target substrates, and a suitable recipe is generated in accordance with the differences between the individual substrates. In the third embodiment, the model for generating the recipe described above is a model that assumes a single standard SiC substrate as the target substrate. Based on this model and the substrate information of each substrate, the management system 1 can generate a suitable recipe in accordance with the substrate specifications and conditions of each substrate.

[0098] [Effects etc. (3)] As described above, according to the third embodiment, in addition to the same effects as those of the first and second embodiments, stable epitaxial quality can be obtained as a result of film formation using a suitable recipe according to the characteristics of each individual substrate applied in each film formation. In the example of Fig. 9, the film formation in each film formation at each point in time is performed on a substrate having individual characteristics as the target substrate, and suitable epitaxial quality can be obtained in each film formation, including immediately after maintenance.

[0099] <Fourth Embodiment> A management system for an epitaxial film formation apparatus according to a fourth embodiment will be described with reference to FIG. 13. The fourth embodiment is a configuration in which functions are added to the third embodiment. The management system according to the fourth embodiment further has a function of generating an optimal recipe, taking into consideration the characteristics of each of the film formation apparatuses and chambers in the multiple film formation apparatuses 20 and multiple chambers 22. In the fourth embodiment, multiple epitaxial film formation apparatuses and multiple chambers in the epitaxial film formation apparatuses are provided as candidates for processing a substrate. The processor inputs information on the epitaxial film formation apparatus and chamber that are to be used to process the substrate from among the candidates as one piece of input information for the model, and generates a processing recipe for the chamber of the target epitaxial film formation apparatus based on the model.

[0100] The configuration of the management system 1 of the fourth embodiment is similar to that of Fig. 10 described above, except that an equipment number and a chamber number are included as part of the input information 101. The processing flow of the fourth embodiment differs from that of Fig. 12 in that, in step S301, an equipment number and a chamber number are further input and registered in DB 105 in association with other information.

[0101] The equipment number is an ID that identifies each individual film formation equipment 20, and the chamber number is an ID that identifies each individual chamber 22. User U1 inputs, via the GUI, the equipment number of the film formation equipment 20 to be used for film formation and the chamber number of the chamber 22 to be used in that film formation equipment 20 (for example, the epitaxial growth chamber 22d in FIG. 5B). Alternatively, the management system 1 may automatically determine the equipment number and chamber number of the target, or may acquire them from another device, etc. If the equipment number or chamber number is not input or specified, the management system 1 uses default settings for the equipment number and chamber number. These default settings can also be set by user U1 via the GUI.

[0102] The management system 1 performs processes such as recipe generation in step S104 using a model associated with the film forming apparatus 20 and chamber 22 identified by the input apparatus number and chamber number. In step S107, the management system 1 associates information such as the substrate information, apparatus number, chamber number, model, and recipe, and stores the information in DB 105.

[0103] [Film forming equipment and chambers] When different film formation apparatuses 20 and chambers 22 are used for film formation, even if the same recipe is applied, the epitaxial quality of the resulting film may differ. Therefore, the management system 1 of the fourth embodiment identifies each film formation apparatus 20 and its chamber 22 based on the apparatus number and chamber number, and generates an optimal recipe corresponding to the characteristics of the film formation apparatus 20 and chamber 22. Specifically, the characteristics of each film formation apparatus 20 and chamber 22 are reflected in the machine learning model. Alternatively, a model may be constructed and used for each individual film formation apparatus 20 and chamber 22, or a technique such as transfer learning may be used. In transfer learning, a model is constructed assuming a certain film formation apparatus 20 and chamber 22 as fixed, and corresponding models are constructed for other film formation apparatuses 20 and chambers 22 based on that model through transfer learning. This allows for the construction of highly accurate models corresponding to each film formation apparatus 20 and chamber 22 based on limited fixed information.

[0104] FIG. 13 supplements the fourth embodiment by showing an example of the relationship between multiple film formation apparatuses 20, multiple chambers 22, and the models to be applied. Table (a) shows an example of management information for configuring individual models for each film formation apparatus 20 and chamber 22, as a first approach. Table (a) has columns for apparatus number, chamber number, and model. For example, model M111 is applied to chamber number 11 in a film formation apparatus 20 with apparatus number 1, and another model M112 is applied to another chamber with chamber number 12. Another film formation apparatus 20 with apparatus number 2 applies different models (M221, M222). Table (b) shows an example of management information for configuring multiple film formation apparatuses 20 and multiple chambers 22 based on a single common model, as a second approach. For example, two film formation apparatuses 20 with apparatus numbers 1 and 2 use the same model M100. The film forming apparatus 20 with the apparatus number 3 uses the same model M300.

[0105] [Effects etc. (4)] As described above, according to the fourth embodiment, in addition to the same effects as those of the third embodiment, stable epitaxial quality can be obtained as a result of film formation using a suitable recipe according to the characteristics of the individual film formation apparatus 20 and chamber 22 applied in each film formation. For example, the state including the influence of maintenance may differ for each chamber 22 of the film formation apparatus 20. Even in such a case, an optimal recipe can be generated based on a model that takes into account the characteristics of each chamber 22, and the epitaxial quality of film formation immediately after maintenance can be stabilized.

[0106] <Modifications of Embodiments 1 to 4> The following modifications of the first to fourth embodiments are also possible. The configuration of the management system 1 of this modification is similar to that shown in FIG. 10, for example, except that it includes hyperparameters as one piece of input information 101. These hyperparameters refer to parameters set in a known machine learning algorithm, in other words, configuration information for a machine learning algorithm model. The processing flow of this modification differs from the flow of FIG. 12 of the third embodiment, for example, in that in step S301, hyperparameters are input as one piece of input information. User U1 inputs and sets the hyperparameters via a GUI.

[0107] In machine learning, when a huge amount of data is used, the amount of calculation, time, and load increase when searching for an optimal recipe. While the time required for the search can be shortened by increasing the search calculation step (such as the range of value changes), accuracy decreases. Therefore, in the fourth embodiment, appropriate hyperparameters are input and set for the model used to generate the recipe, thereby achieving both high accuracy and short calculation times. For example, the management system 1 sets the starting value for the recipe search as one of the hyperparameters to a recipe value close to the optimal solution. The optimal solution is, for example, a recipe (corresponding film formation conditions) in which the epitaxial quality reaches the target value V1 in FIG. 9. Thus, according to the fourth embodiment, even when the search calculation step for the recipe search is small, the time required for the search may be shortened. In other words, the efficiency of processing and work can be improved.

[0108] [GUI] In the first to fourth embodiments or their modified examples, the following GUI examples can be applied: In this example, a display example of a screen including a GUI for user U1 in a modified example having the above-mentioned hyperparameter function is shown.

[0109] 14 and 15 show examples of the GUI. In FIG. 14, field g1 is a recipe search function field, where the enable / disable state of the functions described in the first embodiment and the like can be set. Field g2 is a maintenance notification message field, where the aforementioned maintenance notification message is displayed. Field g3 is an equipment / chamber information input field, where the aforementioned equipment number and chamber number can be entered and confirmed. Field g4 is a substrate information input field, where the aforementioned substrate information can be entered and confirmed. For example, values such as the substrate ID and thickness of the substrate for each slot can be entered.

[0110] Column g5 is a target value / tolerance input column, where the aforementioned target value, upper limit, and lower limit can be input and confirmed for each parameter representing epitaxial quality. Column g6 is a hyperparameter input column, where the aforementioned hyperparameter values can be input and confirmed. Column g7 is a data set input column immediately after maintenance, where the recipe and epitaxial quality can be input and confirmed as the data set for the first deposition immediately after the aforementioned maintenance. Column g8 is a maintenance evaluation result input column, where the evaluation results of the maintenance effect when the aforementioned maintenance was performed (for example, time tm1 in Figure 9) (such as the maintenance method for the maintenance content and the evaluation value for each maintenance item) can be input and confirmed. The columns in Figure 14 mainly correspond to input columns for setting information.

[0111] 15, each column mainly corresponds to a column for outputting results. Column g9 is an optimal recipe output column, which displays, for example, the contents of an optimal recipe generated to be applied from the second time onward immediately after the maintenance in the first embodiment, and an estimated result of epitaxial quality using the optimal recipe. Column g10 is an epitaxial quality output column immediately after the maintenance, which displays, for example, an evaluation value (described as an actual measurement value) of the epitaxial quality resulting from the second or subsequent film formation immediately after the maintenance in the first embodiment, and a corresponding estimated value (a value estimated by the management system 1 in accordance with the optimal recipe in column g9). Column g11 is a history information output column, which displays history information stored in DB105. Column g11 includes, for example, a processing information column, a recipe content column, and an epitaxial quality evaluation result column. The processing information column includes a processing number, a processing date and time, a substrate ID, etc. The processing information column may also include an equipment number, a chamber number, etc.

[0112] The recipe content field contains the parameter values of the recipe applied to the process. The evaluation result field contains the evaluation values of the epitaxial quality. Column g12 is a maintenance evaluation result output field, and displays the evaluation results of the maintenance effect when, for example, "maintenance required" is detected in the second embodiment and maintenance is performed.

[0113] The management system 1 stores data such as tables corresponding to the respective fields of the above GUI example in the DB 105 or memory. Based on the data, the management system 1 generates and displays screen data (which may be, for example, a Web page) to be displayed on the display screen of the output device 15B.

[0114] As another GUI, the management system 1 may display a graph of the results of detecting and monitoring changes over time in the physical quantity of the deposits using the sensor 27 in FIG. 1 in association with other information.

[0115] [Note] The present invention has been specifically described above based on the embodiments. However, the present invention is not limited to the above-described embodiments and can be modified in various ways without departing from the spirit and scope of the present invention. Unless otherwise specified, each component may be singular or plural. Combinations of the embodiments are also possible. Except for essential elements, components of the embodiments can be added, deleted, or replaced. In the embodiments, recipe generation and adjustment have been described as being applied to SiC epitaxial film formation. However, the present invention can also be applied to other processes and apparatuses, such as lithography (exposure, electron beam lithography, X-ray lithography, etc.), other film formation (CVD, PVD, evaporation, sputtering, thermal oxidation, etc.), pattern processing (etching, electron beam, laser, etc.), ion implantation (plasma, etc.), and cleaning (liquid, ultrasonic, etc.). [Explanation of symbols]

[0116] 1...management system, 10...computer system, 20...epitaxial film formation apparatus, 11...integrated management unit, 12...equipment control unit, 13...recipe search unit, 14...analysis and evaluation unit, 15A...input device, 15B...output device, U1...user, 21...control unit, 22...chamber, 23...stage, 24...SiC substrate, 25...epitaxial film, 26...deposition location, 27...sensor

Claims

1. A management system for an epitaxial film formation apparatus, The management system includes: maintenance history information that associates details of past maintenance of the epitaxial film formation apparatus with maintenance effect evaluation results of the past maintenance is stored in a database as part of processing history information; generating the next maintenance implementation content by machine learning based on the processing history information stored in the database; Epitaxial deposition equipment management system.

2. 2. The epitaxial deposition apparatus management system according to claim 1, The maintenance effect evaluation result is obtained by using a measurement value of a physical quantity of deposits in a chamber of the epitaxial film formation apparatus. Epitaxial deposition equipment management system.

3. 2. The epitaxial deposition apparatus management system according to claim 1, The next maintenance implementation content is a maintenance implementation content necessary for the evaluation value of the quality of the epitaxial film to be within an allowable range. Epitaxial deposition equipment management system.

4. 2. The epitaxial deposition apparatus management system according to claim 1, The details of the maintenance include a date and time of the maintenance, a maintenance method, and parameter values constituting the maintenance method. Epitaxial deposition equipment management system.

5. 2. The epitaxial deposition apparatus management system according to claim 1, The management system quantitatively evaluates the effect of the past maintenance and generates an evaluation value representing the effect of the past maintenance. Epitaxial deposition equipment management system.

6. 2. The epitaxial deposition apparatus management system according to claim 1, The management system includes: a function of estimating a recipe to be applied to film formation immediately after maintenance of the epitaxial film formation apparatus; and a function of notifying a user when it is determined that the evaluation value of the quality of the epitaxial film in the estimated recipe is not within an allowable range. Epitaxial deposition equipment management system.

7. 7. The epitaxial film formation apparatus management system according to claim 6, The notification includes information indicating that maintenance is required. Epitaxial deposition equipment management system.

8. 7. The epitaxial film formation apparatus management system according to claim 6, The notification includes information about the next maintenance to be performed that is necessary to bring the quality evaluation value within the acceptable range. Epitaxial deposition equipment management system.

9. 7. The epitaxial film formation apparatus management system according to claim 6, the deposition is deposition of an epitaxial film on a SiC substrate, The notification is a notification including information on specifications of the SiC substrate necessary for the quality evaluation value to be within the acceptable range. Epitaxial deposition equipment management system.

10. 7. The epitaxial film formation apparatus management system according to claim 6, The management system includes: inputting substrate information about a substrate to be used for film formation by the epitaxial film formation apparatus into the database; estimating the recipe to be applied to the film formation based on the substrate information; generating conditions for the substrate quality required to generate the recipe within the tolerance range; If the substrate used for the film formation does not satisfy the conditions, a notification is given as to whether or not to change to a substrate that satisfies the conditions; If it is input that the substrate is not to be changed, a notification including information that maintenance is required is given. Epitaxial deposition equipment management system.

11. 11. The epitaxial deposition apparatus management system according to claim 10, The management system includes: inputting information on a plurality of epitaxial deposition apparatuses that are candidates for processing the substrate and chambers within each of the epitaxial deposition apparatuses into the database; estimating the recipe for processing in an epitaxial deposition apparatus and chamber to be used for the deposition based on information in the database; Epitaxial deposition equipment management system.

12. A management system for an epitaxial film formation apparatus, The management system includes: inputting information on a recipe for a first process immediately after maintenance of the epitaxial film forming apparatus and an evaluation value of the quality of the epitaxial film resulting from the process, and performing machine learning to output a recipe for a second or subsequent process in which the evaluation value of the quality falls within an acceptable range; a recipe for a first process immediately after the maintenance is set to the same values as a recipe for a process immediately before the maintenance, the first process immediately after the maintenance is executed, an evaluation value of the quality of the result of the first process immediately after the maintenance is obtained, and the obtained information is used for the machine learning; The machine learning is machine learning that handles fluctuations in the state of the epitaxial film formation apparatus before and after maintenance of the epitaxial film formation apparatus. Epitaxial deposition equipment management system.

Images