JP2022112078A - Management system and management method for epitaxial deposition equipment - Google Patents

Abstract

To provide a technique capable of obtaining stable quality as quality of a SiC epitaxial film on a system, etc., for controlling an epitaxial film deposition apparatus.SOLUTION: A control system 1 for creating a recipe of processing an epitaxial film deposition apparatus 20 includes a processor and a memory. The processing of the epitaxial film deposition apparatus 20 includes processing of depositing an epitaxial film using epitaxial growth on a substrate (a SiC substrate 24), and the processor creates a first processing recipe right after maintenance of the epitaxial film deposition apparatus 20 and a recipe of inputting information on a quality evaluation value of the processed epitaxial film 25 to construct or update a processing model as an initial condition and form a quality evaluation value in an allowable range including a target value as a recipe for the second and following processing right after the maintenance on the basis of the model.SELECTED DRAWING: Figure 1

Description

The present invention relates to technology such as a management system for an epitaxial film deposition apparatus.

The most important issues in realizing a sustainable society are depletion of energy resources and excessive emission of greenhouse gases such as CO2. For this reason, power converters and the like that are excellent in energy efficiency and emit little CO2 have become important. Most power converters are composed of a power module in which an insulated gate bipolar transistor (IGBT), which is a switching element, and a PiN diode (PND), which is a rectifying element, are connected in parallel. For this reason, the loss reduction of the semiconductor element is directly linked to the energy saving of the power converter. As a technique for reducing the loss of semiconductor devices, a method of forming devices with 4H-type silicon carbide (4H—SiC: hereinafter also referred to as SiC) has attracted attention. A high-quality and inexpensive SiC epitaxial substrate (hereinafter also referred to as a substrate) is required to improve the reliability of SiC devices and reduce the cost.

2. Description of the Related Art As one of semiconductor manufacturing apparatuses, there is an epitaxial film forming apparatus. As an epitaxial film forming apparatus for SiC, there is a SiC epitaxial growth apparatus. Epitaxial growth is one of thin film crystal growth techniques, and is a technique for growing crystals on a crystal that serves as an underlying substrate and arranging them in alignment with the crystal plane of the substrate. SiC epitaxial growth is a technique for depositing SiC on an offcut SiC substrate. In general, SiC substrates have a high donor concentration, so it is necessary to adjust the donor concentration and film thickness according to the withstand voltage application, and epitaxial growth (in other words, epitaxial film formation) is performed to fabricate SiC devices. Demands on epitaxial growth technology are wide-ranging, such as increasing the diameter of epitaxial growth accompanying the increasing diameter of substrates, securing uniformity of donor concentration, securing uniformity of epitaxial film thickness, high-speed growth, and reduction of crystal defects. .

A management system for an epitaxial film forming apparatus is one of the systems for performing evaluation, management, control, etc. related to semiconductor manufacturing. This management system manages film formation conditions (sometimes referred to as “recipes”, etc.), which are conditions for controlling the film formation process, to be set in the film formation apparatus. This management system comprehends and evaluates the state and result of the film forming process based on the film forming conditions of the film forming apparatus, and generates suitable film forming conditions.

As an example of the prior art related to the above, Japanese Patent Application Laid-Open No. 2020-123675 (Patent Document 1) can be cited. Japanese Patent Application Laid-Open No. 2002-200002 describes a technique for determining a recipe according to a change in a semiconductor manufacturing apparatus over time as a management system for the semiconductor manufacturing apparatus.

JP 2020-123675 A

However, in the conventional semiconductor device manufacturing technology, the quality of the SiC epitaxial film (which may be abbreviated as "epitaxial film") produced in the film forming process of the film forming apparatus is affected by changes over time in the film forming apparatus. It turns out that it fluctuates due to maintenance, etc. Therefore, in order to obtain a stable epitaxial film, there is a problem that the film formation cost increases.

In particular, the inventors have found that even if film formation conditions suitable for an epitaxial film formation apparatus are once set, the quality of the resulting epitaxial film varies each time the film formation apparatus is maintained. . Before and after maintenance, although it is generally difficult to quantify, there are fluctuations in the state of the equipment. For example, substances associated with film formation are deposited on the walls of a vacuum chamber (also called a chamber) of a film formation apparatus. Immediately after maintenance, the past film forming conditions may no longer be suitable due to such fluctuations in the state of the apparatus. Therefore, the quality of the epitaxial film formed as a result of film formation immediately after maintenance sometimes deviates from the range of the predetermined quality conditions determined by the manufacturing specifications and the like.

An object of the present invention is to provide a technique that can obtain stable quality as the quality of the SiC epitaxial film in relation to the technique such as the management system of the epitaxial film forming apparatus.

A representative embodiment of the present invention has the following configuration. A management system for an epitaxial deposition apparatus according to an embodiment is a management system for generating a recipe for processing an epitaxial deposition apparatus, comprising a processor, wherein the processing for the epitaxial deposition apparatus uses epitaxial growth on a substrate to form an epitaxial film. including a process of forming a film, the processor inputs a recipe for a first process immediately after maintenance of the epitaxial film forming apparatus and information on an evaluation value of the quality of the epitaxial film as a process result, and performs the process as an initial condition; is constructed or updated, and based on the model, as a recipe for the second and subsequent processes immediately after the maintenance, a recipe is generated in which the quality evaluation value is within an allowable range including the target value.

According to the representative embodiments of the present invention, stable quality can be obtained as the quality of the SiC epitaxial film with respect to the technology such as the management system of the epitaxial film forming apparatus. Problems and effects other than those described above are described in [Mode for Carrying Out the Invention].

1 shows the configuration of a management system for the epitaxial film deposition apparatus of Embodiment 1. FIG. 1 shows the configuration of a management system in Modification (1) of Embodiment 1. FIG. 2 shows the configuration of a management system in modification (2) of Embodiment 1. FIG. 1 shows a detailed configuration example of a management system according to Embodiment 1. FIG. In Embodiment 1, a configuration example of an epitaxial film forming apparatus is shown. 1 shows a functional block configuration of a management system according to Embodiment 1; 4 shows a processing flow of the management system of Embodiment 1. FIG. In Embodiment 1, explanatory diagrams such as the configuration of the model are shown. In the first embodiment, an example of changes in recipe and epi quality over time is shown. 1 shows the configuration of a management system according to a second embodiment; 10 shows a processing flow of the management system of Embodiment 2. FIG. 11 shows a processing flow of the management system of Embodiment 3. FIG. FIG. 10 shows an example of correspondence between a film forming apparatus, a chamber, and a model in the management system of Embodiment 4. FIG. FIG. 10 shows a GUI display example in a modified example of the first to fourth embodiments; FIG. FIG. 5 shows a continuation of the GUI display example in the modified examples of the first to fourth embodiments. FIG. 10 shows an example of changes over time of recipes and epi quality in a management system of a comparative example with respect to the embodiment.

BEST MODE FOR CARRYING OUT THE INVENTION Hereinafter, embodiments of the present invention will be described in detail with reference to the drawings. In the drawings, in principle, the same parts are denoted by the same reference numerals, and repeated explanations are omitted. In the drawings, the representation of each component may not represent the actual position, size, shape, range, etc. in order to facilitate the understanding of the invention, and the present invention is based on the positions disclosed in the drawings. , size, shape, range, and the like.

For the purpose of explanation, when explaining the processing by the program, there are cases where the program, function, processing unit, etc. are mainly explained, but the main body as hardware for them is the processor or the controller composed of the processor etc. , devices, computers, systems, etc. A computer executes processing according to a program read out on a memory by a processor while appropriately using resources such as a memory and a communication interface. As a result, predetermined functions, processing units, and the like are realized. The processor is composed of, for example, a semiconductor device such as a CPU or GPU. A processor is composed of devices and circuits capable of performing predetermined operations. The processing can be implemented not only by software program processing but also by dedicated circuits. FPGA, ASIC, etc. can be applied to the dedicated circuit. The program may be installed as data in the target computer in advance, or may be distributed to the target computer as data from the program source 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 consist of a plurality of program modules. A computer system may be configured by a plurality of devices. The computer system may consist of a client-server system, a cloud computing system, an IoT system, or the like. For the sake of explanation, various data and information may be described in terms of, for example, tables and lists, but the structure and format are not limited to these. Data and information for identifying various elements may be described using expressions such as identification information, identifiers, IDs, names, numbers, etc. These expressions can be replaced with each other.

[Problems, etc.]
A supplementary explanation of the above-mentioned problems will be given. A management system and a management method for a semiconductor manufacturing apparatus as comparative examples for the embodiments will be outlined below. The management system of the comparative example records and holds processing history information of the film forming apparatus. The processing history information is, in other words, time-series data of processing, and includes information such as recipes used for processing and evaluation results of epitaxial quality as processing results. The management system autonomously estimates changes over time in the chamber from the processing history information, and determines whether maintenance is necessary. A recipe is a film formation condition for realizing a film formation result close to a target value using a model. In the technique of this comparative example, since the temporal change of the device state is estimated from the processing history information including the recipe and the epitaxial quality, film formation for confirming the temporal change is not involved.

FIG. 16 shows a graph of an example of temporal changes of (a) film formation conditions and (b) epitaxial quality in a comparative example. The horizontal axis is the film formation timing (corresponding point or time). For example, during the period from time t1 to time t2, the film formation conditions are set such that the epitaxial quality has a substantially constant value v1 within the target range V0. The predetermined target range V0 is a range from the lower limit value V2 to the upper limit value V3 centering on the value V1. For example, at time tm1, maintenance of the film forming apparatus was performed. As a result, the device state is inherently fluctuating. As a result, in the period from time t3 to time t4 after maintenance, under the same film formation conditions as in the period before maintenance, the epitaxial quality has a value v2 that is outside the target range V0. Similarly, maintenance of the film forming apparatus was performed at time tm2. As a result, in the period from time t5 to time t6 after maintenance, the epitaxial quality has a value v3 outside the target range V0 under the same film forming conditions as in the period before maintenance.

As described above, the comparative example does not take into consideration the internal fluctuations in the device state before and after maintenance, and determines the recipe even immediately after maintenance, assuming the same constant device state before and after maintenance. Therefore, as described above, the epitaxial quality may deviate from the target range V0 for multiple film formations immediately after maintenance (for example, each film formation indicated by multiple plots from time t3).

Fluctuations in the state of the deposition apparatus are generally difficult to quantify due to various factors. In the embodiment, as a mechanism, it is not necessary to quantify the state of the deposition apparatus, and for the recipe as the input of the model, the evaluation value of the epitaxial quality of the deposition result as the output is used. It controls the generation and adjustment of suitable recipes.

<Embodiment 1>
Techniques such as a management system for an epitaxial film forming apparatus according to a first embodiment of the present invention will be described with reference to FIGS. 1 to 9. FIG. The management system of Embodiment 1 is a system having a function of generating and proposing a recipe, which is an optimum film forming condition suitable for the state of the epitaxial film forming apparatus, and is mainly realized by a computer system. The management method of the first embodiment is a method having steps executed by the management system of the first embodiment. The management system of the first embodiment uses the epitaxial film quality (sometimes referred to as epitaxial quality) value of the epitaxial film formation result immediately after the maintenance of the epitaxial film forming apparatus as an initial condition to set the film forming conditions related to the epitaxial film forming. A model for generating is built and updated by machine learning. In other words, "immediately after maintenance" refers to the time when the film formation process is performed for the first time after maintenance.

[Management system]
FIG. 1 shows a schematic configuration of a management system for an epitaxial film deposition apparatus according to a first embodiment. A management system 1 of Embodiment 1 is a management system for an epitaxial film deposition apparatus 20 and is implemented by a computer system 10 . The computer system 10 includes an integrated management section 11, an apparatus control section 12, a recipe search section 13, an analysis evaluation section 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. The user U1 operates and utilizes the management system 1 by the computer system 10 via the input device 15A and the output device 15B. A user U1 is, for example, a person who manages a manufacturing process (especially a film forming process). The computer system 10 is connected to the control section 21 of the epitaxial film forming apparatus 20 through communication 200 such as wired or wireless communication.

The epitaxial deposition apparatus 20 includes a controller 21 and a chamber 22 . The control unit 21 controls the driving of each unit of the film forming apparatus. The chamber 22 is configured by, for example, a vacuum chamber or the like, and includes a stage 23 inside the vacuum chamber. The stage 23 is controlled, for example, in horizontal movement (for example, parallel movement and rotation). A SiC substrate 24 or the like is placed on the stage 23 as a target substrate. The film forming apparatus 20 forms an epitaxial film 25 by performing an epitaxial film forming process on the surface of the SiC substrate 24 . A predetermined substance is deposited in the chamber 22 in accordance with such an epitaxial film formation process. Deposition point 26 is, as an example, a side wall surface of chamber 22 . The film forming apparatus 20 includes a sensor 27 which will be described later. The sensor 27 measures predetermined physical quantities such as thickness and amount of the deposits at the deposition location 26 . The mode of the chamber 22 and the mode of deposition are not limited to this example, and there are various modes.

The integrated management unit 11 is a part that performs input, output, storage, management, etc. of necessary data/information. The apparatus control unit 12 is a part that controls the operation of the film forming apparatus 20 while communicating with the film forming apparatus 20 . The recipe search unit 13 is a part that searches for an optimum recipe to be set in the film forming apparatus based on machine learning or the like. The analysis/evaluation unit 14 is a part that analyzes and evaluates the result of film formation according to the recipe, and obtains an evaluation value, etc. of the quality of the epitaxial film. The analysis/evaluation unit 14 or the user U1 obtains values of the analysis/evaluation results, such as the epitaxial film thickness, impurity concentration, crystal defect density, and the like.

[Management system - modified example (1)]
FIG. 2 shows an implementation example of the management system 1 in a modified example of the first embodiment. In the implementation example of FIG. 2, components similar to those of FIG. In the configuration example of FIG. 2, each unit such as the integrated management unit 11 is implemented by a computer system. That is, it has a computer system CS1 of the integrated management unit 11, a computer system CS2 of the device control unit 12, a computer system CS3 of the recipe search unit 13, and a computer system CS4 of the analysis evaluation unit 14. It is connected to the. Each computer system may be configured by connecting a plurality of computers, or may apply a machine in which a plurality of GPUs are connected in parallel, for example. In the configuration example of FIG. 2, rapid processing is possible by parallel distributed processing using a plurality of computer systems.

[Management system - modified example (2)]
FIG. 3 shows an implementation example of the management system 1 in another modified example of the first embodiment. In the implementation example of FIG. 3, the recipe search unit 13 is provided in the form of a server or the like on the cloud computing system 202 via the wide area communication network as part of the same components as in FIG. A computer system CS5 including the integrated management unit 11 and the like is connected to a recipe searching unit 13 such as a server on the cloud computing system 202 via a wide area network. Since the processing of the recipe search unit 13 is related to machine learning and the like, the load is relatively high and may require high performance and a large amount of computational resources. Therefore, in this implementation example, rapid processing is possible by implementing the recipe search unit 13 on the cloud computing system 202 . Also, in this implementation example, one recipe searching unit 13 can be shared among a plurality of computer systems associated with a plurality of film forming apparatuses.

[Management System - Details]
FIG. 4 shows a detailed configuration example of the management system 1 of the first embodiment shown in FIG. 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 a bus or the like for interconnecting them. The function of each part may be realized by processing by a computer program, or may be implemented by a dedicated circuit such as FPGA. Further, the configuration of FIG. 4 includes a controller 211 connected to the device control unit 12 and a controller 212 connected to the analysis evaluation unit 14, and these controllers are set or instructed by the control unit or the user U1. corresponds to an operation unit for inputting

[Epitaxial deposition equipment]
FIG. 5 shows a mounting configuration example of the epitaxial film forming apparatus 20 of FIG. (A) of FIG. 5 shows an example of an implementation configuration in the form of a cluster device. The cluster apparatus comprises a substrate analysis chamber 22A, a cleaning chamber 22B, an assay chamber 22C, a processing chamber 22D, a regeneration chamber 22E, and a load lock/transfer chamber 22F.

FIG. 5B shows a mounting configuration example in the form of a SiC epitaxial device. This SiC epitaxial device is a SiC epitaxial cluster device comprising analysis/regeneration/growth chambers, 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. - A transfer chamber 22f is provided.

The management system 1 of the first embodiment manages semiconductor manufacturing apparatuses including an epitaxial film forming apparatus 20 as targets. In addition, the management system 1 requires, as a component, an analytical evaluation device for substrates and epitaxial films, or a means for acquiring analytical evaluation result data from the analytical evaluation device. 1, the analytical evaluation chamber 22C in FIG. 5A, and the epi-analysis chamber 22b in FIG. 5B. The necessary components may be integrated into one device as a cluster device as shown in FIG. 5A, or may be a SiC epitaxial device as shown in FIG. 5B. Alternatively, substrate analysis chambers 22A and 22a may be provided, and the management system 1 may automatically acquire and use substrate information from the substrate analysis chambers or the like as in Embodiment 3 described later.

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

Moreover, the management system 1 may be provided with a regeneration chamber 22E and a CMP chamber 22e. The regeneration chamber 22E is a chamber for regenerating the substrate processed in the processing chamber 22D. The CMP chamber 22e is an example of the regeneration chamber 22E, and is a chamber for regeneration of the substrate by CMP (Chemical Mechanical Polishing). As a result, even if the desired epitaxial result is not obtained as a result of the processing (for example, if the epitaxial quality of the processing result with the generated recipe does not satisfy the allowable range described later), the substrate (SiC substrate in FIG. 1 24) can be reused by CMP or the like.

The substrate analysis/evaluation apparatus includes, for example, a substrate shape measurement apparatus and a defect/surface roughness evaluation apparatus. A substrate shape measuring device measures a 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 topography, lasers, or microscopes.

The epitaxial film analysis and evaluation apparatus includes, for example, an epitaxial concentration evaluation apparatus, an epitaxial film thickness evaluation apparatus, and a defect evaluation apparatus. Defect evaluation equipment evaluates defects using, for example, X-rays, PL light, CL, radiation topography, lasers, or microscopes.

[Device status change]
In the SiC epitaxial film forming apparatus, which is the epitaxial film forming apparatus 20 of FIG. 1, substances are deposited as follows as an example of apparatus state changes during film formation. During SiC epi-growth, by-products derived from the material gas adhere firmly to members other than the SiC substrate 24 (for example, the inner wall of the chamber 22 and the susceptor). These by-products (in other words, deposits) are exposed to high temperatures during the epitaxial growth and evaporate, causing changes in epitaxial quality over time. In other types of semiconductor manufacturing equipment, such as CVD equipment, it is relatively easy to remove by-products by gas cleaning or the like. However, in the SiC epitaxial film forming apparatus, at present, effective methods such as gas cleaning have not been established. Therefore, in order to remove the by-products, maintenance such as frequent opening of the chamber is required, which causes an increase in cost.

In addition, in the SiC epitaxial deposition system, as a result of maintenance of the chamber, etc., as described above, the state of the system inherently changes. may change and the allowable range may not be satisfied. Therefore, the management system according to the first embodiment finds the optimum recipe for the second and subsequent processes immediately after maintenance based on the input information for the first process. In other words, the management system updates the model so that the epitaxial film quality resulting from processing with the estimated recipe is the optimum recipe that is as close to the target value as possible within the allowable range.

[Management system - functions]
The main functions and processing overview of the management system 1 are as follows. The outline of the processing is also shown in the flow of FIG. 7, which will be described later. The management system 1 inputs the first recipe and epitaxial quality (analysis evaluation result value is used as the quality) immediately after the maintenance of the film forming apparatus 20, and the model of the prior art example or the model before maintenance (to be described later). The epitaxial quality estimated using f(x)) in FIG. 8 is compared with the input latest epitaxial quality to calculate a compensation coefficient (A in FIG. 8). The management system 1 updates the model to a new model (F(x) in FIG. 8) using the calculated compensation coefficient. This model is a machine learning model described later (FIG. 8) for recipe generation/estimation. This model update is, in other words, an update for compensating for device state changes before and after maintenance.

The management system 1 uses the updated new model to estimate the optimum recipe for the target SiC substrate 24, and estimates the epitaxial quality of the processing result for each of the estimated candidate recipes. The management system 1 generates an optimum recipe such that the estimated epi quality satisfies the allowable range. The management system uses the generated recipe to perform the second and subsequent film formations immediately after maintenance, and saves data such as the recipe and epitaxial quality at the time of execution as a history (in other words, processing history information) in a database (DB). do.

Moreover, the management system 1 of Embodiment 1 may perform control as follows when the epitaxial quality estimated from the optimum recipe does not satisfy the allowable range. That is, the management system 1 notifies the user U1 of information on the optimum recipe calculated at that time via a graphical user interface (GUI), and informs the user U1 that the recipe cannot satisfy the allowable range. The user U1 confirms this notification, determines whether or not to allow the film formation according to the recipe, and makes an input according to the determination. Even if the allowable range cannot be satisfied with that recipe, if an input permitting execution is entered, execution of film formation with that recipe is permitted. If an input is made that does not allow execution, it is possible to take measures such as reviewing the set value of the allowable range. This function corresponds to steps S105 and S106 in the flow of FIG. 7, which will be described later.

In addition, in each film formation, even if the epitaxial quality of the film formation result in the recipe does not satisfy the allowable range, the management system notifies the user U1 through the GUI and responds to the user U1's judgment input. , terminate the flow. This function corresponds to steps S111 and S112 in the flow of FIG. 7, which will be described later.

[Management system - functional block configuration]
FIG. 6 shows a functional block configuration example of the management system 1 according to the first embodiment shown in FIG. 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 construction unit 107, a recipe estimation unit 108, an apparatus control unit 110, a process processing unit 111, an analysis evaluation unit 112, and It has a convergence determination unit 113 .

The input information 101 through the input device 15A includes target values, allowable ranges, analysis evaluation results, first recipes immediately after maintenance, epi quality, and the like. Epitaxial quality is an evaluation value such as concentration. The target value is the epitaxial quality target value. The allowable range is a range defined according to specifications and the like, which is configured using target values. The analysis evaluation result is information that can be acquired from the analysis evaluation device. The first recipe and epitaxial quality information immediately after maintenance is the recipe applied in the first film formation immediately after maintenance and the evaluation value of the epitaxial quality of the film formation result.

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

The output information 102 through the output device 15B includes the optimum recipe, analysis evaluation results, and the like. The optimum recipe is set in the epitaxial film forming apparatus 20 (also referred to as the film forming apparatus 20) as a recipe to be applied after the second time immediately after maintenance. In other words, the recipe set in the film forming apparatus 20 is updated to the new recipe.

The user U1 can check the contents of the output information 102 on the display screen of the display device, which is one of the output devices 15B. The user U1 can instruct, set, and input the input information 101 according to the GUI on the display screen.

The management system 1 includes a model configuration unit 107, a recipe estimation unit 108, an apparatus control unit 110, a process processing unit 111, an analysis evaluation unit 112, and a convergence determination unit as units realized based on program processing by the central processing unit 104. It has a part 113 . The central processing unit 104 is composed of the processor shown in FIG. 4 and the like, and performs processing while appropriately using resources such as memory and communication interfaces. Control programs and the like are stored in the ROM or a secondary storage device (not shown). The DB 105 can be composed of 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, and the like. The DB 105 organizes and stores various data and information such as input information 101, output information 102, related information, and setting information. A memory such as a RAM stores data and information being processed.

The model configuration unit 107 configures and manages a machine learning model. A recipe estimation unit 108 estimates a recipe based on the model. These (107, 108) correspond to the recipe searching unit 13 in FIG. The device control unit 110 controls the film forming device 20 . The process processing unit 111 performs processing related to the film forming process in the film forming apparatus 20 . These (110, 111) correspond to the device controller 12 in FIG. The analysis evaluation section 112 corresponds to the analysis evaluation section 14 in FIG.

[Processing flow (1)]
FIG. 7 shows a flow of main processing by the management system 1 (especially the computer system 10, central processing unit 104, processors, 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 with that recipe.

In step S<b>102 , the management system 1 uses the recipe estimation unit 108 to calculate the compensation coefficient of the recipe generation model (see FIG. 8 to be described later) in the model construction unit 107 .

In step S<b>103 , the management system 1 inputs information on the desired target value and the allowable range as part of the input information 101 . For example, user U1 sets the target value and allowable range through input device 15A and output device 15B.

At step S104, the management system 1 uses the model compensated using the compensation factor of step S102 to generate an optimal recipe for achieving an epi quality value as close as possible to the target value within an acceptable range; Estimate the epi quality value of the resulting epi deposition based on this recipe. This epi quality estimation can be realized, for example, as a simulation.

In step S105, the management system 1 judges and confirms whether the epitaxial quality value obtained as a result of the above estimation satisfies the allowable range. If satisfied (Y), proceed to step S107. If not satisfied (N), proceed to step S106.

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 accepts an input based on confirmation/judgment by the user U1 as to whether or not the film formation according to the recipe is permitted. If the input permitting the execution of film formation with the recipe is input (Y), the process proceeds to step S107, and if the input is not permissible (N), the process returns to step S103. In step S103, the setting of the allowable range is reviewed.

In step S<b>107 , the management system 1 causes the film forming apparatus 20 to perform the epitaxial film forming process according to the optimum recipe by the process processing section 111 . Then, the management system 1 inputs the information of the recipe used at that time into the DB 105 and saves it as part of the history.

In step S108, the management system 1 causes the analysis/evaluation unit 112 to perform an analysis/evaluation of the epitaxial quality of the film formation process result with the above recipe, and obtains analysis/evaluation result information (in other words, process result information) including the epitaxial quality. create.

In step S<b>109 , 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 stores the processing result information in the DB 105 as part of the history in association with the recipe information used in the processing.

In step S<b>110 , the management system 1 updates the recipe generation model in the model configuration unit 107 using the updated information in the DB 105 .

In step S111, the management system 1 determines and confirms whether the epitaxial quality of the film formation result with the optimum recipe satisfies the allowable range. At (N), the process proceeds to step S112.

In step S112, the management system 1 notifies the user U1 via the GUI of the output device 15B that the recipe used for the film formation process at that time did not satisfy the permissible range. Then, the management system 1 receives an input by the user U1 for confirmation/judgment as to whether or not to end with the film formation result. If an input to end the film formation is made (Y), the flow ends, and if an input not to end is made (N), the process returns to step S104, for example. In step S104, the recipe is reviewed. After that, the same processing is repeated.

Based on the flow described above, epitaxial film formation is performed with an optimum or suitable recipe every time after the second time immediately after maintenance, and optimum or suitable epitaxial quality can be stably obtained. Note that, in the example of the flow described above, the management system 1 can collectively estimate and propose the optimum recipe as described above in a plurality of times after the second time after maintenance. Not limited to this, the management system 1 may estimate and propose the optimum recipe as described above every time after the second time after maintenance. The management system 1 may estimate and propose the above recipe in units of set times.

[model]
FIG. 8 is an explanatory diagram of a method for updating a model using machine learning, etc., in the first embodiment. A method of constructing a model for generating and estimating a recipe will be described below. (a) shows an overview of the model for the recipe. In the machine learning model, the input information includes the recipe and epitaxial quality for each deposition, and the output information includes the optimal recipe and epitaxial quality estimation results. In Embodiment 1, in particular, the model input includes the first recipe and epi quality immediately after maintenance, and the model output is the estimated optimal recipe and epi quality for the second and subsequent applications immediately after maintenance. Including epi quality. Optimal means that the estimated epi quality satisfies the acceptable range and is as close as possible to the target value.

(b) shows an overview of model update. Using functional representation, we denote the old or previous model by f(x) and the new model after update by F(x). A new model, F(x), is derived from F(x)=A*f(x)+W0, as shown. A is a compensation factor. W0 is the device state immediately after maintenance.

(c) shows a model update method using machine learning. In this method, the optimum recipe is generated based on the model, taking into consideration the apparatus state (W0) immediately after maintenance of the film forming apparatus 20 . In this method, the concentration distribution _Ct is used as one of the epitaxial qualities (corresponding evaluation values) of film formation results up to maintenance. Let _Xt be a latent variable representing the device state. The subscript t represents time.

The concentration distribution _Ct , which is the output according to the device state, is described as shown. Equation 1 is C _t =g(X _t , θ)+δ. θ is a value according to the recipe, substrate information (described later), and the like. δ is the dispersion. Equation 2 is X _t =f ₁ (X _t−1 , Z _t−1 )+ε. Z _t-1 is a value that depends on the presence or absence of maintenance, film thickness of deposits, elapsed time, and the like. ε is the scatter. g(X _t , θ) and f ₁ (X _t−1 , Z _t−1 ) are functions. The functional forms of these functions are unknown. Therefore, the management system 1 derives functional forms of these functions by performing regression analysis.

The concentration distribution C for the second and subsequent times after maintenance, excluding the first time immediately after maintenance of the film forming apparatus 20, is described by the following equation. Equation 3 is C=A*C _t +C ₀ . Equation 4 is A=f ₂ (C ₀ ,C ₋₁ ). _C0 is the first measured value of the concentration distribution immediately after maintenance. C ₋₁ is the measured value of the concentration distribution before maintenance. The compensation factor A can be derived from the measured concentration distribution (C ₀ , C ₋₁ ), as described above. Therefore, the concentration distribution C immediately after maintenance can be derived using the concentration distribution _Ct .

In addition, for the details of recipe generation or machine learning methods, known methods such as neural networks and support vector machines can be applied. Applicable.

[Recipe adjustment]
FIG. 9 shows a specific example of changes in epitaxial quality and adjustment of recipes in accordance with each epitaxial film formation in the film forming apparatus 20 and appropriate maintenance in the management system 1 of the first embodiment. It is an explanatory diagram showing. In the graph of FIG. 9, the abscissa indicates the film formation time (corresponding time point or number of times), and shows the time change of the film formation conditions, which are recipes (a), and the epitaxial quality (b). In other words, the content of FIG. 9 shows an example of adjusting the optimum recipe immediately after maintenance in accordance with fluctuations in the device state and epitaxial quality that accompany maintenance. The recipe of (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 of SiC epitaxial growth. The epitaxial quality of (b) is an evaluation value of the quality of the SiC _epitaxial film (the epitaxial film 25 in FIG. 1) as a result of the film formation process in the recipe of (a). is the value of

The permissible range 900 is a range having a lower limit value V2 and an upper limit value V3 centering on the epitaxial quality target value V1. Time tm1 and the like are examples of dates and times when maintenance of the film forming apparatus 20 is performed. An example of maintenance is conditioning the chamber 22, such as by removing deposits in the chamber 22 as described above.

For example, at time t1 (the first time in total), as a result of film formation with the recipe value p1, the epitaxial quality value is v1, which satisfies the allowable range 900. FIG. At time t2 (the second time in total), as a result of film formation with the recipe value adjusted to p2, the epitaxial quality value is v2, which is close to the target value V1 within the allowable range of 900. there is During the period from time t2 to time t10 (the 10th time in total), the epitaxial quality of each time was within the allowable range 900 as shown by the value v2 as a result of performing film formation each time while finely adjusting the recipe value. is maintained at a value close to the target value V1.

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

In the first film formation immediately after maintenance at time t11, the recipe value p3 is the same as the recipe value at time t10 immediately before maintenance. As shown, it fluctuates to the value v3. This value v3 is outside the allowable range 900 below the lower limit value V2.

Therefore, the management system 1 performs the above-described processing of FIG. 7 based on the recipe value p3 and the epitaxial quality value v3 at time t11 (first time immediately after maintenance). 2), adjustment 901 is performed by proposing to change the value p3 of the recipe to the optimum value p4 estimated based on the new model. As a result of this adjustment 901, the value of the epitaxial quality resulting from the film formation with the recipe value p4 at time t12 is a value v2 within the allowable range 900 and close to the target value V1.

The recipe is similarly adjusted each time after time t13. For example, at time t12, recipes for each time after time t13 are also proposed collectively, and the values are adjusted to increase little by little for each time. During the period from time t12 to time t20, the value of the epitaxial quality is maintained at the value v2 within the allowable range 900 as a result of each film formation.

After time t20, at time tm2, maintenance of the film forming apparatus 20 is performed. As a result, at the first time point t21 (21st time in total) immediately after the maintenance, the epitaxial quality of the film formation result with the recipe value p5 is the value v4. The value v4 is out of the allowable range 900 so as to exceed the upper limit value V3. Therefore, the management system 1 performs adjustment 902 in the same manner as adjustment 901 described above. In this adjustment 902, the recipe value p5 at the first time point t21 immediately after maintenance is changed to the recipe value p6 at the second time point t22 immediately after maintenance. After the time t22, the epitaxial quality of each film formation result is maintained at the value v2 along with the fine adjustment of the recipe. Thereafter, the recipe is similarly adjusted according to the epitaxial quality immediately after maintenance is performed. Note that although the example of FIG. 9 shows an adjustment example for one parameter that constitutes a recipe and one evaluation value for epitaxial quality, the present invention is not limited to this, and can be similarly applied to a plurality of other parameters. .

[Effects (1)]
As described above, according to the epitaxial film forming apparatus management system and the like of the first embodiment, stable quality can be obtained as the quality of the SiC epitaxial film. According to the first embodiment, it is possible to reduce the trouble of creating and setting recipes by the user, and to improve 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 for the second and subsequent times is set to be the optimum recipe according to the epitaxial quality of the result of the first film formation immediately after maintenance. adjust. As a result, it is possible to immediately converge and maintain the epitaxial quality at a suitable value within the allowable range in each of the second and subsequent film formations except the first after maintenance.

<Embodiment 2>
A management system and the like of the epitaxial film forming apparatus according to the second embodiment will be described with reference to FIGS. 10 and 11. FIG. The basic configuration of the second embodiment and the like is the same as that of the first embodiment, and the components of the second embodiment and the like that are different from the first embodiment will be mainly described below. The management system of the second embodiment has a function of notifying maintenance when the epitaxial quality in the recipe estimated for application to the film formation process immediately after maintenance does not satisfy the allowable range.

[Management system (2)]
FIG. 10 shows the functional block configuration of the computer system 10 in the management system 1 of the second embodiment. This configuration has 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 components of the first embodiment shown in FIG. The output information 102 includes maintenance notifications and maintenance evaluation results in addition to the above information.

In the management system 1, when the epitaxial quality in the recipe estimated by the recipe estimating unit 108 to be applied to the film formation process immediately after maintenance does not satisfy the allowable range, the maintenance determining unit 109 instructs the central processing unit 104 to Enter the information to the effect that maintenance is required. This is information indicating that maintenance of the film forming apparatus 20 is required to satisfy the allowable range. Upon receiving this input, the central processing unit 104 shifts to the maintenance notification step (see FIG. 11 to be described later). The central processing unit 104 causes the recipe estimating unit 108 to start generating the optimum maintenance method necessary for generating a suitable recipe. The recipe estimator 108 generates an optimum 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 optimum maintenance method and the like through the GUI.

There are several types of maintenance methods for the film forming apparatus 20 . For example, assume that there are three types of methods A, B, and C. The recipe estimation unit 108 selects the optimum maintenance method necessary for generating a suitable recipe (that is, a recipe whose epitaxial quality satisfies the allowable range) from such candidate maintenance methods.

Generation of a maintenance method can be realized, for example, as follows. The management system 1 manages the DB 105 so that maintenance history information including information such as the date and time and method of performing maintenance is included in the processing history information. When the maintenance determination unit 109 determines that maintenance is required, the management system 1 determines which maintenance method is the most suitable among a plurality of candidate maintenance methods, based on a model including maintenance history information, and determines whether maintenance is required. Determine and derive detailed parameter values, etc. in the method.

Moreover, in Embodiment 2, the effect of the maintenance when the film forming apparatus 20 is maintained is quantitatively evaluated. The maintenance effect evaluation unit 106 quantitatively evaluates the effect of the maintenance and inputs the maintenance effect evaluation result to the central processing unit 104 . The evaluation of the maintenance effect can be carried out, for example, by measuring the deposition amount and thickness of by-products or deposits on the inner wall of the chamber 22 in FIG. For example, the sensor 27 in FIG. 1 measures predetermined physical quantities such as the deposition amount and thickness of the deposition location 26 . The management system 1 acquires this measured value and uses it as an evaluation value of the maintenance effect. Examples of the sensor 27 include a film thickness sensor and an optical film thickness evaluation device. The results of this measurement may be entered into the management system 1 by the user U1. The management system 1 may automatically acquire and input the signal of the sensor 27 . The management system 1 may monitor and record the signal of the sensor 27 over time.

As in the case of the first embodiment, the second embodiment and the like also have a function (step S106 in FIG. 7) that allows the user U1 to perform processing even when the optimum recipe does not satisfy the allowable range.

[Processing flow (2)]
FIG. 11 shows the processing flow of the management system 1 according to the second embodiment. The flow of FIG. 11 has a step S200 related to the maintenance notification function added to the flow of FIG. In step S105 described above, if the epi quality estimated by the estimated recipe does not satisfy the allowable range (N), the process transitions to step S200. Step S200 has steps S201 to S204.

In step S201, the management system 1 causes the recipe estimating unit 108 to use the model in the model configuration unit 107 to generate a recipe for realizing an epitaxial quality value close to the target value. In other words, maintenance contents, etc.) are generated.

In step S<b>202 , the management system 1 notifies the user U<b>1 through the GUI to perform maintenance according to the above maintenance method, and the user U<b>1 performs maintenance of the film forming apparatus 20 . The maintenance effect evaluation unit 106 evaluates the maintenance effect of the maintenance implementation result and uses it as maintenance effect evaluation result information. Alternatively, the user U1 evaluates the maintenance effect and inputs maintenance effect evaluation result information.

In step S203, the central processing unit 104 acquires maintenance effect evaluation result information, associates the maintenance implementation details (including 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. do.

In step S<b>204 , the management system 1 updates the model using the updated information in the DB 105 by the model configuration unit 107 . After step S204, the flow ends.

[Effects (2)]
As described above, according to the second embodiment, when the allowable range cannot be satisfied with a recipe without maintenance, a suitable maintenance method is notified and executed so that the allowable range can be satisfied. As a result, in the first film formation immediately after the maintenance is performed, good epitaxial quality that satisfies the allowable range can be obtained as a result of film formation using the suitable recipe. According to the second embodiment, it is possible to improve the efficiency of maintenance work according to the maintenance notification.

The effect will be supplemented using FIG. 9 described above. In the case of Embodiment 1, for example, it is assumed that time points tm1, tm2, etc. are regular maintenance. On the other hand, in the case of the second embodiment, as an operation, there is no such regular maintenance, or additional maintenance is carried out as appropriate between regular maintenances. For example, after the time point t10, the management system 1 determines that the epitaxial quality does not satisfy the allowable range in the film formation at the next time point t11, assuming that there is no maintenance. Therefore, the management system 1 generates a suitable recipe and maintenance method in consideration of the maintenance implementation and its implementation result as described above, and notifies the user U1 of them. As a result, maintenance is performed between times t10 and t11, for example. As a result, the epitaxial quality can be kept within the allowable range in the first film formation at time t11 immediately after the maintenance.

<Embodiment 3>
A management system and the like of the epitaxial film forming apparatus according to the third embodiment will be described with reference to FIG. The third embodiment is a form in which functions are added to the second embodiment. The management system according to the third embodiment inputs information (referred to as substrate information) of the substrate (SiC substrate 24 in FIG. 1) that is the target of the film formation process, and generates an optimum recipe in consideration of the substrate information. etc. After setting a substrate (for example, a semiconductor wafer) on the stage 23 in the chamber 22, the management system 1 first evaluates the substrate and obtains substrate evaluation result information.

[substrate]
SiC substrates generally have more defects such as dislocations than Si substrates. For example, dislocation density affects defect density after epi deposition. Therefore, when the defect density is used as an output (the output of the model described above, an evaluation value of the epitaxial quality), it is necessary to consider the information on the defects of the substrate. In addition, substrate warpage and the like also affect the epitaxial quality. Therefore, in the third embodiment, by adding substrate information including such information, an optimal recipe is generated with high accuracy. In the third embodiment, board information is included in the input information in the teacher data for machine learning. The management system 1 generates an optimum 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 the allowable range.

When generating the recipe, the board information, which is one of the inputs, must be constrained so that it satisfies the board specifications (in other words, board quality) specified in advance. For example, if the target substrate has many defects, the estimated optimal recipe may not satisfy the allowable range. For this reason, the management system 1 calculates and judges the number of defects (in other words, defect density) in the target substrate at which the optimum recipe can be generated. The management system 1, for example, reduces the number of defects step by step and calculates whether or not a recipe that satisfies the allowable range can be generated under each condition of the number of defects. Then, the management system 1 notifies, through the GUI, the conditions of the number of defects and the defect density (in other words, the conditions of the substrate specifications) for which the recipe can be generated. The management system 1 notifies whether or not the installed target substrate satisfies the conditions. After that, if the target board does not satisfy the conditions, the management system 1 confirms with the user U1 through the GUI whether to change the target board to a board that satisfies the conditions, and asks for a decision. User U1 determines and inputs whether or not to change the board accordingly. Further, when it is decided not to change the board, the management system 1 transitions to the maintenance execution flow.

[Management system (3)]
The functional block configuration of the management system 1 in Embodiment 3 is the same as the configuration in Embodiment 2 shown in FIG. One of them includes information on board specifications such as the defect density, board change notification, and the like.

[Processing flow (3)]
FIG. 12 shows the processing flow of the management system 1 according to the third embodiment. The flow of FIG. 12 is obtained by adding processing steps related to board information to the flows of FIGS. 7 and 11 . Step S301 is provided between steps S103 and S104 in FIG. In step S<b>301 , the management system 1 inputs/obtains substrate information about the target substrate (SiC substrate 24 ) on the stage 23 and registers it in the DB 105 . At this time, the user U1 may input the substrate information through the GUI, or the management system 1 may acquire the substrate information from another device.

The processing contents of step S104 are partially different from those described above (referred to as S104c). In step S104c, the management system 1 uses the model described above and the substrate information described above to generate an optimal recipe and estimate the epitaxial quality corresponding thereto.

Also, the processing contents of step S107 are partly different from those described above (referred to as S107c).

In the third embodiment, if the permissible range is not satisfied in step S105 (N), the process proceeds to step S302. In step S302, the management system 1 causes the recipe estimating unit 108 to use the model of the model constructing unit 107 to generate a recipe that realizes an epitaxial quality value close to the target value within an allowable range. Conditions such as the defect density are derived.

Next, in step S303, the management system 1 determines whether or not the target board on the stage 23 satisfies the conditions of the board specifications based on the board specifications/conditions and the board information. determines whether to change the target substrate on the stage 23 to another substrate that satisfies the above substrate specifications and conditions. At this time, the management system 1 notifies the user U1 of the information such as the defect density of the target board and the board specifications/conditions through the GUI. Accept judgments and inputs about The user U1 confirms the notification and determines/inputs whether or not to change the board. If the board is to be changed (Y), the process returns to step S301. In that case, in step S301, board information about the changed board is input. If the substrate is not to be changed (N), the process proceeds to step S200 relating to the above-described maintenance notification, and similar processing is performed.

[Substrate and model]
In the third embodiment, as described above, a plurality of substrates are treated as target substrates, and suitable recipes are generated corresponding to differences among individual substrates. In the third embodiment, the recipe generation model described above is a model that assumes one standard SiC substrate as a target substrate. The management system 1 can generate a suitable recipe according to the board specifications/conditions of each board based on this model and the board information of each board.

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

<Embodiment 4>
A management system and the like of the epitaxial film forming apparatus according to the fourth embodiment will be described with reference to FIG. The fourth embodiment is a form in which functions are added to the third embodiment. The management system of Embodiment 4 further has a function of generating an optimum recipe, etc., in consideration of the characteristics of each film forming apparatus and chamber in the plurality of film forming apparatuses 20 and the plurality of chambers 22 . The fourth embodiment has a plurality of epitaxial deposition apparatuses and a plurality of chambers in the epitaxial apparatus as candidates for processing the substrate. The processor inputs, as one of the input information of the model, the information of the epitaxial deposition apparatus and the chamber to be processed on the substrate among the candidates, and based on the model, processes in the chamber of the target epitaxial deposition apparatus. Generate a recipe for

The configuration of the management system 1 of Embodiment 4 is the same as that of FIG. 10 described above, except that input information 101 includes an apparatus number and a chamber number. The processing flow in the fourth embodiment differs from that of FIG. 12, for example, in that, in step S301, an apparatus number and a chamber number are input and registered in the DB 105 in association with other information.

The device number is an ID that identifies each film forming device 20 , and the chamber number is an ID that identifies each chamber 22 . Through the GUI, the user U1 can enter the device number of the target film forming device 20 to be used for film formation, and the chamber number of the chamber 22 to be used in the film forming device 20 (for example, the epitaxial growth chamber 22d in FIG. 5B). Enter the number and Alternatively, the management system 1 may automatically grasp the target device number and chamber number, or acquire them from another device or the like. If the device number and chamber number are not input/designated, the management system 1 uses default setting values for the device number and chamber number. This default setting value can also be set by the user U1 through the GUI.

The management system 1 uses a model corresponding to the film forming apparatus 20 and chamber 22 identified by the input apparatus number and chamber number to perform processes such as recipe generation in step S104. In step S<b>107 , the management system 1 associates information such as substrate information, equipment number, chamber number, model, and recipe, and saves the information in the DB 105 .

[Deposition apparatus and chamber]
When the film forming apparatus 20 and the chamber 22 used for film formation are different, even if the same recipe is applied, the epitaxial quality of each film formation result may be different. Therefore, the management system 1 of Embodiment 4 grasps the individual film forming apparatuses 20 and their chambers 22 used for film formation based on the apparatus number and the chamber number, and determines the characteristics of the film forming apparatuses 20 and chambers 22 to generate a suitable recipe. Specifically, the machine learning model reflects the characteristics of each film forming apparatus 20 and chamber 22 . A model for each film forming apparatus 20 and chamber 22 may be constructed and used, or a technique such as transfer learning may be used. In transfer learning, one model is constructed assuming that one film deposition apparatus 20 and chamber 22 are fixed, and based on the model, other film deposition apparatuses 20 and chambers 22 are handled by transfer learning. configure the model; As a result, a highly accurate model corresponding to each film forming apparatus 20 and chamber 22 can be configured based on less fixed information.

FIG. 13 shows, as a supplement, an example of the relationship between a plurality of film forming apparatuses 20, a plurality of chambers 22, and applicable models, etc., in the fourth embodiment. Table (a) shows an example of management information in the case of configuring individual models for individual film deposition apparatuses 20 and chambers 22 as a first method. The table in (a) has as columns the device number, the chamber number, and the model. For example, the model M111 is applied to the chamber of chamber number=11 in the film deposition apparatus 20 of apparatus number=1, and another model M112 is applied to another chamber of chamber number=12. Another model (M221, M222) is applied to the film forming apparatus 20 with another apparatus number=2. Table (b) shows an example of management information in the case of configuring a plurality of film forming apparatuses 20 and a plurality of chambers 22 based on one common model as a second method. For example, the same model M100 is used in two film deposition apparatuses 20 with apparatus numbers=1 and 2. FIG. The same model M300 is used in one film deposition apparatus 20 with the apparatus number=3.

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

<Modifications of Embodiments 1 to 4>
As modifications of the first to fourth embodiments, the following are also possible. The configuration of the management system 1 of the modified example is the same as, for example, that of FIG. The hyperparameters refer to parameters set for a known machine learning algorithm, in other words, setting information for a machine learning algorithm model. The processing flow in this modified example differs from the flow in FIG. 12 of the third embodiment, for example, in that hyperparameters are input as one of the input information in step S301. The user U1 inputs and sets hyperparameters through the GUI.

In machine learning, when using a huge amount of data, the amount of calculation, time, and load increase when searching for the optimal recipe. It is possible to shorten the time required for the search by increasing the search calculation step (the width of changing the value, etc.), but the precision is lowered. Therefore, in the fourth embodiment, by inputting and setting appropriate hyperparameters for the model for recipe generation, both high accuracy and short time calculation are achieved. For example, the management system 1 sets a recipe search start value as one of the hyperparameters to a recipe value close to the optimum solution. The optimum solution is, for example, a recipe (corresponding film formation conditions) that makes the epitaxial quality equal to the target value V1 in FIG. As a result, according to the fourth embodiment, even when the search calculation steps of recipe search are small, the time required for search may be shortened. That is, it is possible to improve the efficiency of processing and work.

[GUI]
The following GUI examples are applicable to the first to fourth embodiments or their modifications. This example shows a display example of a screen including a GUI for the user U1 in the modified example having the function of the hyperparameters.

14 and 15 show GUI examples. In FIG. 14, column g1 is a recipe search function column, in which the enable/disable state of the functions described in the first embodiment and the like can be set. A field g2 is a maintenance notification message field, in which the aforementioned maintenance notification message is displayed. A field g3 is an apparatus/chamber information input field, in which the above-mentioned apparatus number and chamber number can be input/confirmed. A field g4 is a board information input field, in which the aforementioned board information can be input and confirmed. For example, values such as the board ID and thickness of the board for each slot can be input.

A column g5 is a target value/allowable range input column, in which the aforementioned target value, upper limit value, and lower limit value can be input/confirmed for each parameter representing epitaxial quality. A field g6 is a hyperparameter input field, in which the values of the hyperparameters described above can be input/confirmed. A field g7 is a data set input field immediately after maintenance, in which the recipe and epitaxial quality can be input/confirmed as a data set for the first film formation immediately after the maintenance described above. A field g8 is a maintenance evaluation result input field, and the evaluation result of the maintenance effect (maintenance method of maintenance contents, evaluation value of each maintenance item, etc.) when the above-described maintenance is performed (for example, time tm1 in FIG. 9). can be entered and confirmed. Each column in FIG. 14 mainly corresponds to an input column for setting information.

Next, in FIG. 15, each column mainly corresponds to the result output column. A column g9 is an optimum recipe output column, and contains, for example, the contents of the optimum recipe generated for the second and subsequent times immediately after the maintenance in Embodiment 1, and the estimated result of the epitaxial quality with the optimum recipe. Is displayed. A column g10 is an epitaxial quality output column immediately after maintenance, and for example, an epitaxial quality evaluation value (denoted as an actual measurement value) in the result of film formation after the second time immediately after maintenance in Embodiment 1, and the corresponding value. Estimated values (values estimated by the management system 1 corresponding to the optimum recipe in column g9) are displayed. A field g11 is a history information output field in which the history information stored in the DB 105 is displayed. The column g11 includes, for example, a processing information column, a recipe content column, and an epi-quality evaluation result column. The processing information column has a processing number, processing date and time, board ID, and the like. The processing information column may also have an apparatus number, a chamber number, and the like.

The Recipe Content column has each parameter value of the recipe applied to the process. The evaluation result column has each evaluation value of epitaxial quality. A column g12 is a maintenance evaluation result output column, and displays the evaluation result of the maintenance effect when maintenance is carried out with "maintenance required" in the above-described second embodiment, for example.

The management system 1 holds in the DB 105 or memory data such as a table corresponding to each column of the above GUI example. Based on the data, the management system 1 generates and displays screen data (for example, a web page) to be displayed on the display screen of the output device 15B.

As another GUI, the management system 1 may display a graph or the like of the results of detection and monitoring of temporal changes in the physical quantity of deposits by the sensor 27 of FIG. 1 in association with other information.

[Appendix]
Although the present invention has been specifically described above based on the embodiments, the present invention is not limited to the above-described embodiments, and can be variously modified without departing from the scope of the invention. Each component may be singular or plural unless otherwise specified. Forms based on combinations of embodiments are also possible. Addition, deletion, replacement, etc. of the constituent elements of the embodiment are possible except for essential elements. In the embodiments, the recipe generation/adjustment is applied to SiC epitaxial film formation. (CVD, PVD, vapor deposition, sputtering, thermal oxidation, etc.), patterning (etching, electron beam, laser, etc.), ion implantation (plasma, etc.), cleaning (liquid, ultrasonic waves, etc.), etc. .

DESCRIPTION OF SYMBOLS 1... Management system, 10... Computer system, 20... Epitaxial film-forming apparatus, 11... Integrated management part, 12... Apparatus control part, 13... Recipe search part, 14... Analysis evaluation part, 15A... Input device, 15B... Output device, U1 -- User, 21 -- Control part, 22 -- Chamber, 23 -- Stage, 24 -- SiC substrate, 25 -- Epitaxial film, 26 -- Deposition point, 27 -- Sensor.

Claims (10)

A management system for generating recipes for epitaxial deposition equipment processing, comprising:
with a processor
The process of the epitaxial film forming apparatus includes a process of forming an epitaxial film using epitaxial growth on the substrate,
The processor
constructing or updating a model of the process as an initial condition by inputting a recipe for the first process immediately after maintenance of the epitaxial film deposition apparatus and information on the evaluation value of the quality of the epitaxial film as a process result;
Based on the model, as a recipe for the second and subsequent processes immediately after the maintenance, a recipe is generated in which the quality evaluation value is within an allowable range including a target value.
Management system for epitaxial deposition equipment. In the epitaxial deposition apparatus management system according to claim 1,
The model is a machine learning model,
Management system for epitaxial deposition equipment. In the epitaxial deposition apparatus management system according to claim 1,
The processor determines that maintenance is required when there is no recipe that satisfies the allowable range for the estimated quality evaluation value among the recipes generated based on the model, and indicates that the maintenance is required. Output,
Management system for epitaxial deposition equipment. In the epitaxial deposition apparatus management system according to claim 3,
When the processor determines that the maintenance is required, the processor uses the model to generate a maintenance method necessary to generate a recipe that satisfies the allowable range, and generates the maintenance method along with the fact that the maintenance is required. outputting and updating the model using the evaluation value of the maintenance effect obtained as a result of performing the maintenance by the maintenance method;
Management system for epitaxial deposition equipment. In the epitaxial deposition apparatus management system according to claim 4,
The processor uses, as an evaluation value of the maintenance effect, a detected value of a physical quantity of by-products in the chamber of the epitaxial deposition apparatus immediately after the maintenance.
Management system for epitaxial deposition equipment. In the epitaxial deposition apparatus management system according to claim 1,
The processor
inputting board information including a quality evaluation value for the board as one of the input information of the model;
If there is no recipe that satisfies the allowable range based on the model and the substrate information, generating conditions for the quality of the substrate required for generating a recipe that satisfies the allowable range;
output whether to change to a substrate that satisfies the above conditions;
If it is input that the substrate is not to be changed, it is determined that the maintenance is necessary, and outputting that the maintenance is necessary;
Management system for epitaxial deposition equipment. In the epitaxial deposition apparatus management system according to claim 1,
having a plurality of epitaxial deposition apparatuses and a plurality of chambers in the epitaxial deposition apparatus as candidates for performing the processing on the substrate;
The processor inputs, as one of the input information of the model, information of an epitaxial deposition apparatus and a chamber to be subjected to the processing on the substrate among the candidates, and performs the target epitaxial deposition based on the model. generating a recipe for the process in the chamber of a membrane device;
Management system for epitaxial deposition equipment. In the epitaxial deposition apparatus management system according to claim 1,
When the quality evaluation value in the recipe estimated as the recipe for the second and subsequent processes immediately after the maintenance does not satisfy the allowable range, the processor determines that the recipe does not satisfy the allowable range but the Output whether or not to execute the process, and if it is input to execute, cause the process in the recipe to be executed;
Management system for epitaxial deposition equipment. In the epitaxial deposition apparatus management system according to claim 1,
The epitaxial film forming apparatus comprises an analysis chamber for analyzing and evaluating the quality of the epitaxial film based on actual measurements,
The management system acquires analysis/evaluation results in the analysis chamber,
Management system for epitaxial deposition equipment. A management method in a management system for generating a recipe for processing an epitaxial deposition apparatus, comprising:
The management system comprises a processor,
The process of the epitaxial film forming apparatus includes a process of forming an epitaxial film using epitaxial growth on the substrate,
the processor
a step of constructing or updating a model of the process as an initial condition by inputting a recipe of the first process immediately after maintenance of the epitaxial deposition apparatus and information of the evaluation value of the quality of the epitaxial film as a process result;
a step of generating, based on the model, a recipe for the second and subsequent processes immediately after the maintenance, in which the quality evaluation value is within an allowable range including a target value;
A method of managing an epitaxial deposition apparatus, comprising:

Images