Substrate processing device and substrate processing method

Abstract

A substrate processing device according to the present invention comprises a weight prediction unit. The weight prediction unit uses a weight change prediction model to predict a change in weight of a resin member with which a processing liquid comes into contact, from a processing condition when substrate processing using the processing liquid is carried out in a processing unit that includes the resin member. The weight change prediction model is an inference model that has been obtained by machine learning of a training data set which includes pairs of the processing condition of the substrate and a change in weight of the resin member after carrying out the substrate processing under the processing condition by the processing unit.

Description

Substrate processing apparatus and substrate processing method

[0001] The present invention relates to a substrate processing apparatus and a substrate processing method.

[0002] Substrate processing apparatuses are used to process substrates such as semiconductor wafers, glass substrates for liquid crystal displays, glass substrates for photomasks, and glass substrates for optical disks using processing liquids such as developing liquid, cleaning liquid, rinsing liquid, and photoresist liquid. For example, in a cleaning apparatus described in Patent Document 1, the substrate is held horizontally by a spin chuck. A cup is disposed so as to surround the sides and below the substrate W held by the spin chuck. A cleaning liquid supply nozzle is disposed above the substrate held by the spin chuck.

[0003] While the substrate is held and rotated by the spin chuck, a cleaning liquid is supplied from a cleaning liquid supply nozzle to the center of the substrate surface. In this case, the resist liquid at the center of the surface is spread over the entire surface by centrifugal force caused by the rotation of the substrate. This cleans the surface of the substrate. Furthermore, the cleaning liquid that splashes from the surface of the substrate to the surrounding area is caught in a cup and then collected in a drainage section.

[0004] Patent No. 4347785

[0005] In substrate processing apparatuses, various components such as cups, nozzles, and piping are made of resin. These resin components gradually corrode or deteriorate when in contact with processing liquids. Therefore, maintenance, such as replacing the resin components or replacing the substrate processing apparatus, is required before the corrosion or deterioration of the resin components becomes severe. However, because the degree of corrosion or deterioration of the resin components varies depending on the operating period of the substrate processing apparatus and the type of substrate processing, it is not easy to perform maintenance on the substrate processing apparatus at the appropriate time.

[0006] An object of the present invention is to provide a substrate processing apparatus and a substrate processing method that are capable of maintaining a resin member at an appropriate timing.

[0007] A substrate processing apparatus according to one aspect of the present invention includes a weight prediction unit that predicts a weight change of a resin member that comes into contact with a processing liquid in a processing unit, using a weight change prediction model, from processing conditions when substrate processing is performed using a processing liquid, and the weight change prediction model is an inference model obtained by machine learning of a learning dataset that includes a pair of substrate processing conditions and weight changes of the resin member after substrate processing is performed by the processing unit under those processing conditions.

[0008] A substrate processing method according to another aspect of the present invention predicts a weight change of a resin member using a weight change prediction model from processing conditions when substrate processing is performed using a processing liquid in a processing unit including a resin member that comes into contact with the processing liquid, and is executed by a processor, wherein the weight change prediction model is an inference model obtained by machine learning from a learning dataset including a pair of substrate processing conditions and weight changes of the resin member after substrate processing is performed by the processing unit under those processing conditions.

[0009] According to the present invention, it is possible to perform maintenance on the resin member at an appropriate time.

[0010] FIG. 1 is a diagram showing an example of the configuration of a substrate processing apparatus according to a first embodiment of the present invention. FIG. 2 is a diagram showing an example of the configuration of an information processing apparatus. FIG. 3 is a schematic diagram showing the interior of a processing unit viewed horizontally. FIG. 4 is a diagram showing a discrete distribution of force field parameters. FIG. 5 is a diagram showing a continuous distribution of force field parameters. FIG. 6 is a diagram showing an example of extraction of molecular descriptors. FIG. 7 is a block diagram showing the functional configuration of the learning apparatus of FIG. 1. FIG. 8 is a diagram showing an example of a data set acquired by a data acquisition unit. FIG. 9 is a block diagram showing the functional configuration of the information processing apparatus of FIG. 2. FIG. 10 is a diagram showing an example of substrate processing conditions and liquid contact times determined by a condition determination unit. FIG. 11 is a diagram showing an example of a prediction of weight change. FIG. 12 is a diagram showing an example of a calculation of a contribution rate. FIG. 13 is a diagram showing an example of a screen displayed on a display device by a presentation unit. FIG. 14 is a diagram showing another example of a screen displayed on a display device by a presentation unit. FIG. 15 is a diagram showing yet another example of a screen displayed on a display device by a presentation unit. FIG. 16 is a diagram showing an example of changing the substrate transport conditions by a transport mechanism. FIG. 17 is a flowchart showing an example of the flow of a learning process. FIG. 18 is a flowchart showing an example of the flow of a life evaluation process.

[0011] 1. First Embodiment (1) Substrate Processing Apparatus A substrate processing apparatus and a substrate processing method according to an embodiment of the present invention will be described below with reference to the drawings. In the following description, the term "substrate" refers to a semiconductor substrate (wafer), a substrate for an FPD (Flat Panel Display) such as a liquid crystal display device or an organic EL (Electro Luminescence) display device, a substrate for an optical disk, a substrate for a magnetic disk, a substrate for a magneto-optical disk, a substrate for a photomask, a ceramic substrate, or a substrate for a solar cell. FIG. 1 is a diagram showing an example of the configuration of a substrate processing apparatus according to a first embodiment of the present invention. As shown in FIG. 1, a substrate processing apparatus 500 includes a processing unit 100, a database storage device 200, an information processing device 300, and a learning device 400.

[0012] The processing unit 100, the database storage device 200, the information processing device 300, and the learning device 400 are connected to a network 501 and are capable of transmitting and receiving data to and from each other. The network 501 may be, for example, a local area network (LAN) or a wide area network (WAN). Alternatively, the network 501 may be the Internet. The connection form of the network 501 may be a wired connection or a wireless connection. Furthermore, the processing unit 100 and the information processing device 300 may be connected by a dedicated network instead of the network 501.

[0013] The processing unit 100 includes a plurality of processing sections 110 and a transport mechanism 120. The transport mechanism 120 sequentially transports substrates to be processed to the plurality of processing sections 110 according to predetermined transport conditions. Each processing section 110 includes various resin members that come into contact with the processing liquid. The structure of the processing section 110 will be described later. Each processing section 110 sequentially performs a series of processes using the processing liquid on a plurality of substrates according to a processing recipe that describes the processing details for the substrates. In this example, each processing section 110 performs a series of processes to form a solid or liquid film on the substrate by supplying the processing liquid to the substrate under predetermined processing conditions, and then removes the solid or liquid film from the substrate and dries the substrate.

[0014] Specifically, each processing unit 110 applies a processing liquid, which is a mixture of a sublimation agent and a solvent at a predetermined concentration, to the surface of the substrate on which the pattern has been formed, and then performs a series of processes to deposit the sublimation agent on the substrate by rotating the substrate and evaporating the solvent. Then, each processing unit 110 sprays an inert gas onto the substrate to dry it by sublimation. However, the substrate processing in each processing unit 110 is not limited to sublimation drying of the substrate, and may also be, for example, development, cleaning, or coating film formation.

[0015] The database storage device 200 includes a large-capacity storage device such as a server. Molecular descriptors of various resin materials and various solutions are stored in the database storage device 200. The molecular descriptors will be described in detail later.

[0016] Information processing device 300 is configured, for example, by a general-purpose computer. Fig. 2 is a diagram showing an example of the configuration of information processing device 300. As shown in Fig. 2, information processing device 300 includes a CPU (central processing unit) 310, a RAM (random access memory) 320, a ROM (read-only memory) 330, a storage device 340, an operation unit 350, a display device 360, an input / output I / F (interface) 370, and a bus 380. CPU 310, RAM 320, ROM 330, storage device 340, operation unit 350, display device 360, and input / output I / F 370 are connected to bus 380.

[0017] The RAM 320 is used as a working area for the CPU 310. The ROM 330 stores a system program. The storage device 340 includes a storage medium such as a hard disk or a semiconductor memory, and stores a life evaluation program for executing the life evaluation process described below. The life evaluation program may be stored in the ROM 330 or another external storage device. The storage device 340 may also store a processing recipe or transport conditions. Furthermore, the storage device 340 may store history information indicating the history of substrate processing in the processing unit 100.

[0018] The operation unit 350 is an input device such as a keyboard, mouse, or touch panel. A user can give predetermined instructions to the information processing device 300 by operating the operation unit 350. The display device 360 ​​is a display device such as a liquid crystal display device, and displays a GUI (Graphical User Interface) for accepting instructions from the user or the processing results of the lifespan evaluation process. The input / output I / F 370 is connected to the network 501 in FIG. 1 .

[0019] The learning device 400 is configured, for example, by a general-purpose computer. Therefore, the learning device 400 has a configuration basically similar to that of the information processing device 300 except for the following points, and therefore a detailed description of the configuration of the learning device 400 will be omitted. The storage device of the learning device 400 stores a learning program for executing the learning process described below. The learning program may be stored in the ROM of the learning device 400 or in another external storage device. Note that the information processing device 300 and the learning device 400 may be configured with the same hardware. In this case, the programs installed on the information processing device 300 and the learning device 400 are different. Therefore, the processes executed by the information processing device 300 and the learning device 400 are different from each other.

[0020] (2) Processing Unit The configuration of the processing section 110 of the processing unit 100 in FIG. 1 will now be described. FIG. 3 is a horizontal schematic diagram of the interior of the processing section 110. While FIG. 3 shows the configuration of one processing section 110, the configurations of the other processing sections 110 are similar to that shown in FIG. 3. As shown in FIG. 3, the processing section 110 includes a processing mechanism 1 and a control device 2. The control device 2 is capable of communicating with the information processing device 300 in FIG. 2 and controls the processing mechanism 1 based on commands from the information processing device 300. The processing mechanism 1 includes a box-shaped chamber 3, a spin chuck 10 that holds a single substrate W horizontally within the chamber 3 and rotates the substrate W about a vertical rotation axis A1 passing through the center of the substrate W, and a cylindrical processing cup 21 that surrounds the spin chuck 10 about the rotation axis A1.

[0021] The spin chuck 10 includes a disk-shaped spin base 12 held in a horizontal position, a plurality of chuck pins 11 that hold the substrate W in a horizontal position above the spin base 12, a spin shaft 13 that extends downward from the center of the spin base 12, and a spin motor 14 that rotates the spin shaft 13 to rotate the spin base 12 and the plurality of chuck pins 11.

[0022] The processing cup 21 includes a plurality of guards 24 that receive the processing liquid discharged outward from the substrate W, a plurality of cups 23 that receive the processing liquid guided downward by the plurality of guards 24, and a cylindrical outer wall member 22 that surrounds the plurality of guards 24 and the plurality of cups 23. The processing cup 21 is an example of a resin member. The plurality of guards 24 can be raised and lowered individually by a guard lifting unit 27.

[0023] The processing mechanism 1 includes a chemical solution nozzle 31 that discharges a chemical solution, a rinse solution nozzle 35 that discharges a rinse solution, a processing solution nozzle 39 that discharges a processing solution, and a substitution solution nozzle 43 that discharges a substitution solution. The chemical solution nozzle 31, the rinse solution nozzle 35, the processing solution nozzle 39, and the substitution solution nozzle 43 can be independently moved horizontally within the chamber 3 by nozzle movement units 34, 38, 42, and 46 provided corresponding to the nozzles.

[0024] The processing liquid nozzle 39 is connected to a processing liquid pipe 40 that guides the processing liquid to the processing liquid nozzle 39. The processing liquid nozzle 39 and the processing liquid pipe 40 are another example of a resin member. When a processing liquid valve 41 disposed in the processing liquid pipe 40 is opened, the processing liquid is continuously discharged downward from the discharge port of the processing liquid nozzle 39. The processing liquid is a mixed liquid containing a sublimation agent and a solvent that is soluble in the sublimation agent.

[0025] The processing liquid nozzle 39 is connected to a nozzle moving unit 42. The nozzle moving unit 42 moves the processing liquid nozzle 39 in at least one of the vertical and horizontal directions. The nozzle moving unit 42 moves the processing liquid nozzle 39 horizontally between a processing position where the processing liquid ejected from the processing liquid nozzle 39 is supplied to the upper surface of the substrate W, and a standby position where the processing liquid nozzle 39 is positioned around the processing cup 21 in a plan view.

[0026] The processing mechanism 1 includes a disk-shaped blocking member 51 arranged above the spin chuck 10. The blocking member 51 includes a disk portion 52 arranged horizontally above the spin chuck 10. The blocking member 51 is supported horizontally by a cylindrical support shaft 53 extending upward from the center of the disk portion 52. The center line of the disk portion 52 is arranged on the rotation axis A1 of the substrate W. The lower surface of the disk portion 52 corresponds to the lower surface 51L of the blocking member 51. The lower surface 51L of the blocking member 51 is parallel to the upper surface of the substrate W and has an outer diameter equal to or greater than the diameter of the substrate W.

[0027] The blocking member 51 is connected to a blocking member lifting unit 54 that vertically raises and lowers the blocking member 51. The blocking member lifting unit 54 moves the blocking member 51 to any position between an upper position (the position shown in FIG. 3 ) and a lower position.

[0028] A center nozzle 55 is disposed in a through-hole that passes vertically through the center of the blocking member 51. The center nozzle 55 rises and falls together with the blocking member 51. The center nozzle 55 is connected to an upper gas pipe 56 that guides inert gas to the center nozzle 55. The processing unit 110 includes an upper temperature regulator 59 that heats or cools the inert gas discharged from the center nozzle 55. When an upper gas valve 57 disposed in the upper gas pipe 56 is opened, the inert gas is continuously discharged downward from the discharge port of the center nozzle 55 at a flow rate that corresponds to the opening of a flow rate adjustment valve 58 that changes the flow rate of the inert gas. The inert gas discharged from the center nozzle 55 is nitrogen gas.

[0029] (3) Database Storage Device The molecular descriptors of the resin material stored in the database storage device 200 of FIG. 1 will be described below. The molecular descriptors of the solvent stored in the database storage device 200 are the same as the molecular descriptors of the resin material. In this example, the molecular descriptors of the resin material are molecular descriptors included in the force field descriptor or the mass of the resin material. The force field descriptors are force field parameters used in MD (molecular dynamics) calculations that describe the behavior of molecules in the resin material.

[0030] Specifically, the force field potential used in MD calculations is expressed by the following formula (1). The first term in formula (1) is the van der Waals interaction energy, and the second term is the Coulomb interaction energy. The third term in formula (1) is the interaction energy of stretching vibration bonds, the fourth term is the interaction energy of bending vibration bonds, and the fifth term is the interaction energy of bonds associated with dihedral angle changes.

[0031]

[0032] The force field parameter ε in Eq. (1) ij , σ ij , q i q j , K bond , r0, K angle , θ 0,i , K dihedral and polarity are used as molecular descriptors, where ε ij is the van der Waals interaction energy depth. σ ij is the equilibrium distance of the van der Waals interactions. i q j is the magnitude of the electrostatic interaction. bond is the spring constant of the chemical bond. r0 is the equilibrium length of the chemical bond. K angle is the force constant for bond bending. θ 0,i is the equilibrium bond angle. dihedral is the rotation barrier height of the dihedral angle. Also, the polarity is the absolute value of the charge difference |q i -q j In this example, the force field parameter K bond , K angle , K dihedral is used as the molecular descriptor.

[0033] An example of a molecular descriptor for polypropylene as a resin material will be described. The following formula (2) shows the chemical structure of polypropylene in SMILES notation. As shown in formula (2), polypropylene is a polymer compound in which many structural units each consisting of a carbon atom and a hydrogen atom are bonded together.

[0034]

[0035] From equation (2), the force field parameter ε ij , σ ij , q i q j , K bond , r0, K angle , θ 0,i , K dihedral A discrete distribution (probability distribution) which is a histogram of the polarity and the charge of the bonded carbon atoms is calculated. FIG. 4 is a diagram showing the discrete distribution of the force field parameters. In the example of FIG. 4, the force field parameter is polarity. Here, Equation (2) includes bonds of eight atoms. There are two bonds between carbon atoms, and six bonds between carbon atoms and hydrogen atoms. Therefore, the absolute value of the difference in charge between the bonded carbon atoms is defined as λ. １ The absolute value of the difference in charge between the bonded carbon atom and the hydrogen atom is taken as λ ２ Then, as shown in FIG. １ , λ ２ The probabilities of the force field parameters in are 2 / 8 and 6 / 8, respectively.

[0036] Next, kernel mean embedding is performed on the force field parameter distribution in Figure 4, thereby converting the force field parameter distribution into a continuous distribution. Figure 5 is a diagram showing the continuous distribution of force field parameters. As shown in Figure 5, multiple kernel functions (Gaussian functions in this example) are calculated using multiple probabilities appearing in the discrete distribution as weights. Next, the discrete distribution is replaced with a continuous function that is the sum of the calculated kernel functions.

[0037] Then, the probabilities of a predetermined number of force field parameters are extracted as molecular descriptors from the continuous distribution curve of the force field parameters in Fig. 5. Fig. 6 is a diagram showing an example of molecular descriptor extraction. As shown in Fig. 6, the probabilities of 10 force field parameter values ​​g determined at equal intervals are extracted from the continuous distribution curve of the force field parameters. １ ~g １０ The probabilities of the 10 force field parameters in are extracted as components of the molecular descriptor. Therefore, in the example of Fig. 6, the molecular descriptor has a 10-dimensional vector structure.

[0038] In the above example, polarity is used as a molecular descriptor. However, other force field parameters such as ε ij , σ ij , q i q j , K bond , r0, K angle , θ 0,i , K dihedral Alternatively, even when the mass of a resin material is used, a molecular descriptor with an arbitrary number of dimensions can be extracted by a similar process. The molecular descriptors extracted from various resin materials are stored in the database storage device 200 in association with the resin material.

[0039] (4) Learning Device Fig. 7 is a block diagram showing the functional configuration of the learning device 400 in Fig. 1. As shown in Fig. 7, the learning device 400 includes, as functional units, a data acquisition unit 401, a descriptor acquisition unit 402, and a model generation unit 403. The functional units of the learning device 400 are realized by the CPU of the learning device 400 executing a learning program. Some or all of the functional units of the learning device 400 may be realized by hardware such as electronic circuits.

[0040] The data acquisition unit 401 acquires multiple data sets that have been generated in advance. Each data set is generated by a user of the processing unit 100 through a preliminary experiment. The user may input the generated data set to the data acquisition unit 401 using an operation unit (not shown). Alternatively, the user may store the generated multiple data sets in a storage device (not shown) of the learning device 400. In this case, the multiple data sets are acquired from the storage device.

[0041] 8 is a diagram showing an example of a data set acquired by the data acquisition unit 401. As shown in Fig. 8, each data set includes a pair of processing conditions for the substrate and a weight change of the resin member corresponding to the processing conditions. The processing conditions for the substrate include, for example, the resin member, processing liquid, temperature of the processing liquid, or concentration of the processing liquid in the processing unit 110 of Fig. 3.

[0042] The weight change of the resin member corresponding to each processing condition is a value obtained by subtracting the weight of the resin member at the time when the processing solution under that processing condition contacts the resin member for a predetermined time from the initial weight of the resin member. In this example, the weight of the resin member is normalized so that the initial weight is 1 (100%). Therefore, the weight of the resin member in this example is a relative value based on the initial weight and is treated as a dimensionless quantity without units.

[0043] In the example of Fig. 8, the data set includes weight changes of the resin member when the time the resin member is in contact with the treatment liquid (hereinafter referred to as the liquid contact time) is one day, one week, and two weeks. The weight of the resin member decreases due to corrosion or deterioration as the liquid contact time increases. Therefore, the weight change of the resin member increases as the liquid contact time increases.

[0044] In a preliminary experiment for generating a data set, a resin member having a known initial weight and made of the same material as the resin member in the processing unit 110 is immersed in the same processing liquid as the processing liquid under the processing conditions. The weight of the resin member is then measured when the immersion time (wetted time) of the resin member reaches a predetermined time. This determines the weight change of the resin member for each wetted time. A set of each processing condition and the weight change of the resin member for each wetted time determined under the processing condition is generated as a data set.

[0045] The descriptor acquisition unit 402 acquires the molecular descriptor of the resin material used as the resin component in the data set from the molecular descriptors of various resin materials stored in the database storage device 200. In this case, the data acquisition unit 401 associates the molecular descriptor of the resin material acquired by the descriptor acquisition unit 402 with the resin component in the data set.

[0046] Similarly, the descriptor acquisition unit 402 acquires the molecular descriptor of the solution used as the treatment liquid in the data set from the molecular descriptors of various solutions stored in the database storage device 200. In this case, the data acquisition unit 401 associates the molecular descriptor of the solution acquired by the descriptor acquisition unit 402 with the treatment liquid in the data set.

[0047] The model generation unit 403 trains a predetermined machine learning model using the substrate processing conditions and wetted time in each data set acquired by the data acquisition unit 401 as explanatory variables and the weight change of the resin member as a target variable. This generates a weight change prediction model that predicts the weight change of the resin member from the substrate processing conditions and wetted time. The machine learning model used to generate the weight change prediction model may be, for example, a decision tree or a neural network. The generated weight change prediction model is installed in the information processing device 300 by being stored in the storage device 340 or the like of the information processing device 300 in FIG. 2 .

[0048] (5) Information Processing Device Fig. 9 is a block diagram showing the functional configuration of the information processing device 300 shown in Fig. 2. As shown in Fig. 9, the information processing device 300 includes, as functional units, a condition determination unit 301, a weight prediction unit 302, a calculation unit 303, a lifespan evaluation unit 304, a presentation unit 305, and a process control unit 306. The functional units of the information processing device 300 are realized by the CPU 310 of the information processing device 300 shown in Fig. 2 executing a lifespan evaluation program. Some or all of the functional units of the information processing device 300 may be realized by hardware such as electronic circuits.

[0049] The condition determination unit 301 determines the processing conditions and the liquid contact time for the substrate. FIG. 10 is a diagram showing an example of the processing conditions and the liquid contact time for the substrate determined by the condition determination unit 301. As shown in FIG. 10 , the processing conditions for the substrate include a descriptor of the processing liquid, a descriptor of the resin member, the temperature of the processing liquid, and the concentration of the processing liquid. The temperature and the concentration of the processing liquid are described in the processing recipe. Therefore, the temperature or the concentration of the processing liquid may be determined based on the processing recipe.

[0050] The descriptor of the treatment liquid is obtained from molecular descriptors of various treatment liquids stored in the database storage device 200 based on information about the treatment liquid (type, temperature, or concentration). Alternatively, it is obtained by calculation by the information processing device 300 based on information about the treatment liquid (type, temperature, or concentration). Similarly, the descriptor of the resin member is obtained from molecular descriptors of various resin members stored in the database storage device 200 based on information about the resin member. Alternatively, it is obtained by calculation by the information processing device 300 based on information about the resin member. As described above, in this example, the force field parameter K is used as the descriptor of the resin member. bond , K angle , K dihedral is used.

[0051] The wetted time is proportional to the supply time of the processing liquid to the substrate. The supply time of the processing liquid to the substrate is specified in the processing recipe. Therefore, the wetted time can also be determined based on the processing recipe. For example, the wetted time may be determined by multiplying the supply time of the processing liquid by a predetermined proportionality coefficient. Alternatively, the user can specify the processing conditions and the wetted time for the substrate to the condition determination unit 301 by operating the operation unit 350. Therefore, some or all of the processing conditions and the wetted time for the substrate may be determined based on the user's instructions. The wetted time shown in FIG. 10 is the cumulative wetted time when substrates are processed multiple times using a certain processing recipe.

[0052] The weight prediction unit 302 uses a weight change prediction model installed in the information processing device 300 to predict a weight change of the resin member after a series of substrate processing operations based on the processing conditions and wet time determined by the condition determination unit 301. Furthermore, if a processing recipe scheduled to be executed in the processing unit is changed, the weight prediction unit 302 updates the predicted weight change based on the changed processing conditions or wet time. Furthermore, the weight prediction unit 302 may calculate the remaining weight of the resin member after a series of substrate processing operations. In this example, similar to the weight change, the remaining weight is treated as a relative value based on the initial weight of the resin member. In this case, the remaining weight is calculated by subtracting the predicted weight change from 1 (100%).

[0053] FIG. 11 is a diagram showing an example of weight change prediction. In this example, processing units 110A to 110C are provided as three processing units 110. Individual processing conditions and liquid contact times are determined for each of the processing units 110A to 110C by the condition determination unit 301. Therefore, as shown in FIG. 11 , the weight change of the resin member in processing unit 110A is predicted from the processing conditions and liquid contact time determined for processing unit 110A. The weight change of the resin member in processing unit 110B is predicted from the processing conditions and liquid contact time determined for processing unit 110B. The weight change of the resin member in processing unit 110C is predicted from the processing conditions and liquid contact time determined for processing unit 110C.

[0054] The calculation unit 303 calculates reference information for the prediction of the weight change of the resin member by the weight prediction unit 302. The reference information includes the contribution rate and prediction accuracy of each processing condition. The contribution rate may be calculated by calculating a SHAP (Shapely Additive Explanations) value or the importance of an explanatory variable of an extra tree. FIG. 12 is a diagram showing an example of calculation of the contribution rate. In the example of FIG. 12, the processing condition with the highest contribution rate among the multiple processing conditions of each processing unit 110 is highlighted. Specifically, in processing unit 110A, the resin member has the highest contribution rate. In processing unit 110B, the processing liquid has the highest contribution rate. In processing unit 110C, the concentration of the processing liquid has the highest contribution rate.

[0055] The life evaluation unit 304 evaluates the life of the resin member against the treatment liquid based on the weight change of the resin member predicted by the weight prediction unit 302. In this example, the time when the cumulative weight change predicted by the weight prediction unit 302 reaches a preset threshold value is evaluated as the life. For example, the threshold value is set to 0.1. That is, the time when the remaining weight of the resin member reaches 0.9 (90%) is evaluated as the life of the resin member. In this example, if the weight change is predicted to be 0.01 for 100 hours of contact time, the life is evaluated as 100 hours x 0.1 / 0.01 = 1000 hours. The evaluated life serves as an indicator of when to replace the resin member.

[0056] The presentation unit 305 presents to the user various information such as the weight change of the resin member predicted by the weight prediction unit 302, the reference information calculated by the calculation unit 303, or the lifespan of the resin member evaluated by the lifespan evaluation unit 304. In this example, a predetermined screen is displayed on the display device 360, thereby presenting various information to the user.

[0057] Fig. 13 is a diagram showing an example of a screen displayed on the display device 360 ​​by the presentation unit 305. The screen in Fig. 13 is referred to as a life prediction screen 361. As shown in Fig. 13, the life prediction screen 361 displays the processing conditions used for predicting the weight change by the weight prediction unit 302, the change over time in the remaining weight of the resin member calculated by the life evaluation unit 304, and the evaluation result of the life of the resin member predicted by the life evaluation unit 304.

[0058] Fig. 14 is a diagram showing another example of a screen displayed on the display device 360 ​​by the presentation unit 305. The screen in Fig. 14 is referred to as a history screen 362. As shown in Fig. 14, the history screen 362 displays the history of substrate processing in the processing unit 110 selected by the user, based on the history information stored in the storage device 340 in Fig. 2. In the example of Fig. 14, the history of substrate processing includes sets of the resin member and processing liquid under the processing conditions, the period, and the remaining weight.

[0059] Fig. 15 is a diagram showing yet another example of a screen displayed on the display device 360 ​​by the presentation unit 305. The screen in Fig. 15 is referred to as a reference information screen 363. As shown in Fig. 15, the reference information screen 363 displays the contribution rate and prediction accuracy of the processing conditions calculated by the calculation unit 303 for the processing unit 110 selected by the user. By appropriately viewing a screen such as the life prediction screen 361, the history screen 362, or the reference information screen 363, the user can recognize the processing conditions of the desired processing unit 110, the life of the resin member, the substrate processing history, or reference information.

[0060] The process control unit 306 controls the operation of the transport mechanism 120 in FIG. 1 to transport the substrate to be processed to one of the multiple processing units 110 based on the transport conditions. Furthermore, the process control unit 306 controls the operation of the processing mechanism 1 of each processing unit 110 via the control device 2 in FIG. 3 of that processing unit 110 based on the process recipe. The process control unit 306 may change the transport conditions of the substrates by the transport mechanism 120 based on the life evaluated by the life evaluation unit 304 so as to change the transport order of the substrates. When the transport conditions are changed, the processing unit to which the substrate is loaded changes. Since the processing recipes are linked to the substrates, if the processing recipes linked to each substrate are different, changing the processing unit to which the substrate is loaded results in a change in the processing content performed in the processing unit. Furthermore, the process control unit 306 may change the processing conditions of the substrate based on the life evaluated by the life evaluation unit 304. When the processing conditions are changed, the processing recipe related to the processing conditions is changed.

[0061] Fig. 16 is a diagram showing an example of changes in the transport conditions for substrates by the transport mechanism 120. In the example of Fig. 16, initially, in the substrate transport order, the processing unit 110A is set as number 1, the processing unit 110B is set as number 2, and the processing unit 110C is set as number 3. Furthermore, the cumulative liquid contact time in each of the processing units 110A to 110C according to the initial processing recipe is 1500 hours.

[0062] In such a processing unit 100, as shown in FIG. 16, the lifespans of the resin members in processing sections 110A to 110C evaluated at a certain point in time are assumed to be 2 months, 1 month, and 1.5 months, respectively. That is, the resin members in processing section 110A have the longest lifespan, and the resin members in processing section 110B have the shortest lifespan. In this case, the order of substrate transport is switched between processing section 110B and processing section 110C. Therefore, in the substrate transport order, processing section 110B is changed from second to third, and processing section 110C is changed from third to second. This extends the lifespan of the resin members in processing section 110B.

[0063] Furthermore, the wetted time in processing section 110A is changed from 1,500 hours to 2,000 hours, and the wetted time in processing section 110B is changed from 1,500 hours to 1,000 hours. This shortens the lifespan of the resin members in processing section 110A and lengthens the lifespan of the resin members in processing section 110B. These changes in transport conditions can equalize the lifespan of the resin members in processing sections 110A to 110C. In the example of Figure 16, the lifespan of the resin members in processing sections 110A to 110C evaluated after the change in processing conditions is 1.5 months for all of them.

[0064] (6) Learning Process Fig. 17 is a flowchart showing an example of the flow of the learning process. The learning process is performed by a CPU included in the learning device 400 as the CPU executes a learning program. The learning process of Fig. 17 will be described below with reference to the learning device 400 of Fig. 7. First, the data acquisition unit 401 acquires a data set that has been generated in advance (step S1).

[0065] Next, the descriptor acquisition unit 402 acquires the molecular descriptor of the resin material used as the resin component in the data set acquired in step S1 from the database storage device 200 (step S2). Subsequently, the data acquisition unit 401 associates the molecular descriptor of the resin material acquired in step S2 with the resin component in the data set acquired in step S1 (step S3).

[0066] Similarly, the descriptor acquisition unit 402 acquires from the database storage device 200 a molecular descriptor of the solution used as the treatment liquid in the data set acquired in step S1 (step S4). Subsequently, the data acquisition unit 401 associates the molecular descriptor of the solution acquired in step S4 with the treatment liquid in the data set acquired in step S1 (step S5). Steps S2 and S3 and steps S4 and S5 may be executed first, or may be executed simultaneously.

[0067] Thereafter, the data acquisition unit 401 determines whether to end the acquisition of the data sets (step S6). If a sufficient number of data sets have been acquired, the data acquisition unit 401 may determine to end the acquisition of the data sets. Alternatively, the data acquisition unit 401 may determine to end the acquisition of the data sets based on a user instruction. The user can instruct the data acquisition unit 401 to end the acquisition of the data sets by operating an operation unit (not shown).

[0068] If the acquisition of the dataset is not completed, the process returns to step S1. Steps S1 to S5 are repeated until the acquisition of the dataset is completed. If the acquisition of the dataset is completed, the model generation unit 403 generates a weight change prediction model by training a machine learning model using the dataset in which the molecular descriptors of the resin material and the molecular descriptors of the solution are associated in steps S3 and S5 (step S7). This completes the learning process.

[0069] (7) Lifespan Evaluation Process Fig. 18 is a flowchart showing an example of the flow of the lifespan evaluation process. The lifespan evaluation process is performed by the CPU 310 of Fig. 2 provided in the information processing device 300 as the CPU 310 executes a lifespan evaluation program. The lifespan evaluation process of Fig. 18 will be described below with reference to the information processing device 300 of Fig. 9. First, the condition determination unit 301 determines the processing conditions and liquid contact time for the substrate in each processing unit 110 (step S11).

[0070] Next, the weight prediction unit 302 predicts a weight change of the resin member after a series of substrate processing in each processing unit 110 from the processing conditions and liquid contact time determined in step S11 using the weight change prediction model generated in step S7 of the learning process (step S12). Furthermore, the calculation unit 303 calculates reference information for predicting the weight change of the resin member in step S12 (step S13).

[0071] Next, the life evaluation unit 304 evaluates the life of the resin member of each processing unit 110 based on the weight change of the resin member predicted in step S12 (step S14). Furthermore, the presentation unit 305 presents to the user various information such as the weight change of the resin member predicted in step S12, the reference information calculated in step S13, and the life of the resin member evaluated in step S14 (step S15).

[0072] Thereafter, the process control unit 306 determines whether the evaluated lifespans of the resin members are uniform for all processing units 110 (step S16). If the difference between the longest and shortest lifespans among the evaluated lifespans is equal to or less than a predetermined threshold, it is determined that the lifespans of the resin members are uniform. If the lifespans of the resin members are uniform, the process control unit 306 determines whether the process recipe has been changed (step S17).

[0073] If the processing recipe has not been changed, the process returns to step S16. In this case, steps S16 and S17 are repeated until the lifespans of the resin members become uneven or the processing recipe is changed. On the other hand, if the processing recipe has been changed, the process returns to step S11. In this case, in step S11, the condition determination unit 301 determines the processing conditions and liquid contact time for the substrate in each processing unit 110 so as to correspond to the changed processing recipe. Thereafter, the processing from step S12 onwards is executed again.

[0074] If the lifespans of the resin members are not uniform in step S16, the process control unit 306 changes the substrate transport conditions or the substrate processing conditions in any of the processing units 110 based on the lifespans evaluated in step S16 (step S18). Specifically, the substrate transport conditions are changed so that the lifespans of the resin members evaluated by the life evaluation unit 304 become more uniform. Because the processing recipes are linked to the substrates, if the processing recipes linked to each substrate are different, the processing unit 110 including the resin members evaluated to have a longer lifespan than the other processing units 110 may be changed so that the substrates to be transported are those associated with a processing recipe that provides a longer liquid contact time. Furthermore, the processing unit 110 including the resin members evaluated to have a shorter lifespan than the other processing units 110 may be changed so that the substrates to be transported are those associated with a processing recipe that provides a shorter liquid contact time.

[0075] Even if the processing recipes associated with the respective substrates are the same, the substrate transport conditions may be changed so that the substrate is moved up in the transport order for a processing unit 110 that includes a resin member that is evaluated to have a longer lifespan than the other processing units 110. Alternatively, the substrate transport conditions may be changed so that the substrate is moved down in the transport order for a processing unit 110 that includes a resin member that is evaluated to have a shorter lifespan than the other processing units 110.

[0076] Furthermore, the substrate processing conditions may be changed so that the liquid contact time is longer than before the change for a processing section 110 including a resin member that is evaluated to have a longer lifespan than the other processing sections 110. Alternatively, the substrate processing conditions may be changed so that the liquid contact time is shorter than before the change for a processing section 110 including a resin member that is evaluated to have a shorter lifespan than the other processing sections 110.

[0077] After step S18 is executed, the process returns to step S12. In this case, in step S12, the weight prediction unit 302 predicts the weight change of the resin member after a series of substrate processing in each processing unit 110 based on the changed substrate processing conditions. Then, step S13 and subsequent steps are executed again.

[0078] (8) Effects In the substrate processing apparatus 500 according to this embodiment, a resin member that comes into contact with a processing liquid is included in the processing unit 110. The weight prediction unit 302 predicts a weight change of the resin member using a weight change prediction model based on processing conditions when a substrate is processed using the processing liquid in the processing unit 110. The weight change prediction model is an inference model obtained by machine learning a learning dataset that includes a pair of substrate processing conditions and weight changes of the resin member after the substrate is processed by the processing unit 110 under the processing conditions.

[0079] According to this configuration, the weight change prediction model is used to predict the weight change of the resin member when substrate processing using the processing liquid is performed in the processing unit 110. In this case, by recognizing the weight change of the resin member, the user can uniformly determine the timing for maintenance of the resin member. This allows the resin member to be maintained at an appropriate timing.

[0080] The lifespan evaluation unit 304 evaluates the lifespan of the resin member based on the weight change of the resin member predicted by the weight prediction unit 302. In this case, since the lifespan of the resin member is evaluated, maintenance of the resin member can be more easily performed at an appropriate time. Furthermore, the lifespan evaluation unit 304 integrates the weight change of the resin member predicted by the weight prediction unit 302, and evaluates the time when the integrated weight change reaches a predetermined threshold as the lifespan of the resin member. In this case, the lifespan of the resin member can be easily evaluated.

[0081] The process control unit 306 changes the substrate transport conditions of the transport mechanism 120 or the substrate processing conditions of any of the processing units 110 so that the lifetimes of the multiple resin members in the multiple processing units 110 evaluated by the lifetime evaluation unit 304 become closer to uniform. In this case, the timing for maintenance of the multiple resin members in the multiple processing units 110 can be synchronized. Therefore, it is not necessary to perform maintenance on the multiple resin members at different times. This improves maintenance efficiency. Also, it is possible to minimize the reduction in operating time of the substrate processing apparatus 500 due to maintenance.

[0082] Specifically, substrates to be processed are loaded into the multiple processing units 110 in an order determined by the transport conditions. The processing control unit 306 changes the transport conditions of the transport mechanism 120 so that the lifespans of the resin components evaluated by the life evaluation unit 304 become more uniform. When the process recipes associated with the substrates are different, the processing unit 110 including the resin components evaluated to have a longer lifespan than the other processing units 110 is changed so that a substrate associated with a process recipe that provides a longer liquid contact time is transported. Alternatively, the processing unit 110 including the resin components evaluated to have a shorter lifespan than the other processing units 110 is changed so that a substrate associated with a process recipe that provides a shorter liquid contact time is transported.

[0083] Even when the process recipes associated with the respective substrates are the same, the process control unit 306 changes the substrate transport conditions so that a processing unit 110 containing a resin member that has been evaluated by the life evaluation unit 304 to have a longer life than the other processing units 110 among the multiple processing units 110 is moved up in the transport order for the substrate. Alternatively, the process control unit 306 changes the substrate transport conditions so that a processing unit 110 containing a resin member that has been evaluated by the life evaluation unit 304 to have a shorter life than the other processing units 110 is moved down in the transport order for the substrate. In this case, the timing for maintenance of the multiple resin members in the multiple processing units 110 can be synchronized with simple control.

[0084] Furthermore, the process control unit 306 changes the substrate processing conditions for a processing unit 110 among the plurality of processing units 110, the processing unit 306 including a resin member evaluated by the life evaluation unit 304 to have a longer life than the other processing units 110, so that the contact time between the resin member and the processing liquid is longer than before the change. Alternatively, the process control unit 306 changes the substrate processing conditions for a processing unit 110 among the plurality of processing units 110, the processing unit 306 including a resin member evaluated by the life evaluation unit 304 to have a shorter life than the other processing units 110, so that the contact time between the resin member and the processing liquid is shorter than before the change. Even in this case, the timing for maintenance of the plurality of resin members in the plurality of processing units 110 can be synchronized with simple control.

[0085] The processing conditions for the substrate by the processing unit 110 include molecular descriptors of the resin member included in the processing unit 110. In this case, a weight change prediction model can be easily generated using the characteristics of the resin member as explanatory variables. Specifically, the molecular descriptors of the resin member are molecular descriptors included in the force field descriptor. With this configuration, a weight change prediction model can be appropriately generated even when the resin member is made of a polymer compound.

[0086] In this example, the force field descriptor is a force field parameter K used in molecular dynamics calculations that describes the behavior of molecules in a resin material. bond , K angle , K dihedral In this case, the weight change of the resin part can be predicted with high accuracy. Furthermore, the molecular descriptor of the resin part is extracted from a distribution curve in which the histogram of the force field parameters is made continuous. With this configuration, even if the type of resin part is different, the molecular descriptor can be extracted uniformly according to a common extraction rule.

[0087] 2. Second Embodiment (1) Data Set As described above, in the first embodiment, a machine learning model is trained using the substrate processing conditions and liquid contact time as explanatory variables and the weight change of the resin member as a target variable, thereby generating a weight change prediction model that predicts the weight change of the resin member from the substrate processing conditions and liquid contact time. The substrate processing conditions include, for example, the resin member, processing liquid, and the temperature or concentration of the processing liquid in the processing unit 110 of FIG. 3.

[0088] In this embodiment, a crystallinity label is added to the resin member, which is one of the substrate processing conditions in the explanatory variables. The crystallinity label is a label for identifying whether the polymer compound (polymer) constituting the resin member is crystalline or amorphous. If the polymer compound is crystalline, the crystallinity label is set to "1," and if the polymer compound is amorphous, the crystallinity label is set to "0."

[0089] Specifically, crystalline polymer compounds have a relatively high degree of crystallinity. Here, the degree of crystallinity is the ratio of the crystalline region to the sum of the crystalline region and the amorphous region of a polymer compound. The crystalline region and the amorphous region of a polymer compound can be identified by analyzing the components of the polymer compound. For example, when observing a polymer compound by XRD (X-ray diffraction), the area of ​​the peak due to the crystalline component is the crystalline region, and the area of ​​the peak due to the amorphous component is the amorphous region. Alternatively, crystalline polymer compounds have a melting point, whereas amorphous polymer compounds do not have a melting point.

[0090] Therefore, in this example, multiple data sets corresponding to multiple resin components are prepared. In each data set, a crystallinity label of "1" is assigned to a resin component formed from a polymer compound whose crystallinity is equal to or greater than a predetermined crystallinity threshold, or a polymer compound having a melting point. On the other hand, a crystallinity label of "0" is assigned to a resin component formed from a polymer compound whose crystallinity is less than the predetermined crystallinity threshold, a polymer compound whose crystallinity cannot be measured, or a polymer compound without a melting point. The crystallinity threshold is, for example, 0.1 (10%), but is not limited to this value.

[0091] 7 trains a predetermined machine learning model using the substrate processing conditions and wetted time in each of the above data sets acquired by the data acquisition unit 401 as explanatory variables and the weight change of the resin member as a response variable. This makes it possible to generate a weight change prediction model that predicts the weight change of the resin member with higher accuracy from the substrate processing conditions and wetted time.

[0092] (2) Verification Example Resin parts with small weight changes are less likely to corrode or deteriorate due to chemical solutions. In other words, resin parts with small weight changes have high chemical resistance. On the other hand, resin parts with large weight changes are more likely to corrode or deteriorate due to chemical solutions. In other words, resin parts with large weight changes have low chemical resistance. As such, there is a correlation between the weight change and chemical resistance of resin parts. Therefore, to facilitate verification of prediction accuracy, a chemical resistance prediction model that predicts chemical resistance was used instead of a weight change prediction model. Furthermore, the LOOCV (Leave-One-Out Cross-Validation) method was used to verify the prediction accuracy of the chemical resistance prediction model.

[0093] In the first verification example, a chemical resistance prediction model was generated using a plurality of data sets similar to those in the first embodiment, which predicts the chemical resistance of a resin member based on the substrate processing conditions and liquid contact time. Furthermore, based on the generated chemical resistance prediction model, the chemical resistance of each resin member under specific substrate processing conditions and liquid contact time was predicted. Furthermore, based on the predicted chemical resistance, each resin member was classified into either a high chemical resistance group or a low chemical resistance group. In the second verification example, a plurality of data sets similar to those in the second embodiment were used, and classification similar to that in the first verification example was performed.

[0094] In each of the first and second verification examples, the F1 score, which indicates the accuracy of the classification results for each resin component, was evaluated. As a result, the average F1 score in the first verification example was 0.72. On the other hand, the average F1 score in the second verification example was 0.75. From the comparison results between the first and second verification examples, it was confirmed that adding the crystallinity label of the resin component to the explanatory variables improved the prediction accuracy of the generated chemical resistance prediction model. Similar results were obtained for the weight change prediction model as for the chemical resistance prediction model.

[0095] 3. Other Embodiments (1) In the above embodiment, the weight and remaining weight of the resin member are treated as relative values ​​based on the initial weight, but the embodiment is not limited to this. The weight and remaining weight of the resin member may be treated as absolute values ​​having units.

[0096] (2) In the above embodiment, the substrate processing apparatus 500 includes a plurality of processing sections 110, but the embodiment is not limited to this. The substrate processing apparatus 500 may include a single processing section 110. Even in this case, the processing control section 306 can adjust the timing for maintenance of the resin member in the single processing section 110 by changing the processing conditions for the substrate in the processing section 110 based on the life of the resin member evaluated by the life evaluation section 304.

[0097] (3) In the above embodiment, the substrate processing apparatus 500 includes the processing section 110, but the embodiment is not limited to this. As long as the information processing apparatus 300 is configured to be connectable to the processing section 110 including a resin member, the substrate processing apparatus 500 does not need to include the processing section 110.

[0098] (4) In the above embodiment, the substrate processing apparatus 500 includes the life evaluation unit 304, but the embodiment is not limited to this. By recognizing the weight change of the resin member predicted by the weight prediction unit 302, a user can uniformly determine the timing for maintenance of the resin member. Therefore, the substrate processing apparatus 500 does not need to include the life evaluation unit 304.

[0099] (5) In the above embodiment, the substrate processing apparatus 500 includes the process control unit 306. However, the embodiment is not limited to this. If the processing unit 110 or the transport mechanism 120 is not controlled based on the weight change of the resin member predicted by the weight prediction unit 302, the substrate processing apparatus 500 does not need to include the process control unit 306.

[0100] (6) In the above embodiment, the treatment liquid is a mixture containing a sublimation agent and a solvent that dissolves in the sublimation agent, but the embodiment is not limited to this. The treatment liquid may be sulfuric acid, nitric acid, hydrochloric acid, hydrofluoric acid, phosphoric acid, acetic acid, ammonia water, hydrogen peroxide water, an organic acid (e.g., citric acid or oxalic acid), an organic alkali (e.g., TMAH: tetramethylammonium hydroxide), or an organic solvent (e.g., IPA: isopropyl alcohol). Alternatively, the treatment liquid may be a mixture of any of these liquids.

[0101] (7) In the above embodiment, the substrate processing conditions include a processing liquid descriptor, a resin member descriptor, the temperature of the processing liquid, and the concentration of the processing liquid, but the embodiment is not limited to this. The processing conditions may also include, for example, other parameters described in the processing recipe. For example, the processing conditions may include a processing liquid supply time, a processing liquid discharge flow rate, a substrate rotation speed, a substrate rotation speed, a processing time, a supply nozzle behavior, a shield plate behavior, an inert gas supply time, or an inert gas discharge flow rate.

[0102] (8) In the above embodiment, the force field parameter K bond , K angle , K dihedral is used as a molecular descriptor, but the embodiment is not limited to this. ij , σ ij , q i q j , K bond , r0, K angle , θ 0,i , K dihedral One or more of the properties and polarity may be used as the molecular descriptor, or the mass of the resin member (resin material) may be used as the molecular descriptor.

[0103] 4. Correspondence between each element of the claims and each part of the embodiment Examples of correspondence between each element of the claims and each element of the embodiment are described below, but the present invention is not limited to the following examples. Various other elements having the configuration or function described in the claims can also be used as each element of the claims.

[0104] In the above-described embodiment, the processing cup 21, the processing liquid nozzle 39, or the processing liquid pipe 40 is an example of a resin member, the processing unit 110 is an example of a processing unit, and the weight prediction unit 302 is an example of a weight prediction unit. The substrate W is an example of a substrate, the substrate processing apparatus 500 is an example of a substrate processing apparatus, the lifetime evaluation unit 304 is an example of a lifetime evaluation unit, and the processing control unit 306 is an example of a processing control unit.

[0105] 5. Summary of Embodiments (Item 1) A substrate processing apparatus according to item 1 includes a weight prediction unit that predicts a weight change of a resin member that comes into contact with a processing liquid, using a weight change prediction model, from processing conditions when substrate processing using a processing liquid is performed in a processing unit including the resin member, and the weight change prediction model is an inference model obtained by machine learning of a learning dataset that includes a pair of processing conditions for a substrate and a weight change of the resin member after substrate processing is performed by the processing unit under the processing conditions.

[0106] In this substrate processing apparatus, a weight change prediction model is used to predict a weight change of a resin member when a substrate is processed using a processing liquid in a processing section. In this case, a user can uniformly determine the timing for maintenance of the resin member by recognizing the weight change of the resin member. This allows the resin member to be maintained at an appropriate timing.

[0107] (Item 2) The substrate processing apparatus described in item 1 may further include a life evaluation unit that evaluates a life of the resin member based on the weight change of the resin member predicted by the weight prediction unit.

[0108] In this case, since the life of the resin member is evaluated, maintenance of the resin member can be more easily performed at an appropriate timing.

[0109] (Clause 3) In the substrate processing apparatus described in clause 2, the life evaluation unit may accumulate the weight change of the resin member predicted by the weight prediction unit, and evaluate the point at which the accumulated weight change reaches a predetermined threshold value as the end of the life of the resin member.

[0110] In this case, the life of the resin member can be easily evaluated.

[0111] (4) The substrate processing apparatus described in paragraph 2 or 3 may further include a processing control unit that changes the processing conditions for the substrate by the processing unit based on the life of the resin member evaluated by the life evaluation unit.

[0112] In this case, the timing for maintenance of the resin member can be adjusted.

[0113] (Item 5) In the substrate processing apparatus described in Item 4, a plurality of the processing units may be provided, and the processing control unit may change the processing conditions for the substrate by any of the processing units so that the lifetimes of the plurality of resin members in the plurality of processing units evaluated by the lifetime evaluation unit become closer to uniform.

[0114] In this case, the timing for maintenance of the plurality of resin members in the plurality of processing sections can be synchronized. Therefore, it is not necessary to perform maintenance on the plurality of resin members at different times. This improves the efficiency of maintenance. Furthermore, it is possible to minimize the reduction in operating time of the substrate processing apparatus due to maintenance.

[0115] (Item 6) In the substrate processing apparatus described in Item 5, the processing control unit may change the substrate processing conditions for a processing unit among the plurality of processing units that includes the resin member that has been evaluated by the life evaluation unit to have a longer life than other processing units, so that the contact time between the resin member and the processing liquid is longer than before the change.

[0116] In this case, the timing for maintenance of a plurality of resin members in a plurality of processing sections can be synchronized with simple control.

[0117] (7) In the substrate processing apparatus described in paragraph 5 or 6, the processing control unit may change the substrate processing conditions for a processing unit among the plurality of processing units that includes the resin member that has been evaluated by the life evaluation unit to have a shorter life than other processing units, so that the contact time between the resin member and the processing liquid is shorter than before the change.

[0118] In this case, the timing for maintenance of a plurality of resin members in a plurality of processing sections can be synchronized with simple control.

[0119] (Item 8) The substrate processing apparatus described in any one of items 2 to 7 further includes a processing control unit, wherein a plurality of processing units are provided, and substrates to be processed are loaded into the plurality of processing units in an order determined by transport conditions, and the processing control unit may change the substrate transport conditions based on the lifespans of the resin members evaluated by the life evaluation unit so that the lifespans of the plurality of resin members in the plurality of processing units become closer to uniform.

[0120] In this case, the timing for maintenance of the resin member can be adjusted.

[0121] (Item 9) In the substrate processing apparatus described in Item 8, the processing control unit may change the substrate transport conditions so that a processing unit containing the resin member that has been evaluated by the life evaluation unit to have a longer life than other processing units among the multiple processing units is moved up in the transport order of the substrate.

[0122] In this case, the timing for maintenance of a plurality of resin members in a plurality of processing sections can be synchronized with simple control.

[0123] (Item 10) In the substrate processing apparatus described in item 8 or 9, substrates to be processed are loaded into the multiple processing units in an order determined by processing conditions, and the processing control unit may change the substrate transport conditions so that a processing unit among the multiple processing units that includes a resin member that has been evaluated by the life evaluation unit to have a shorter life than other processing units is moved down in the transport order of the substrate.

[0124] In this case, the timing for maintenance of a plurality of resin members in a plurality of processing sections can be synchronized with simple control.

[0125] (Item 11) In the substrate processing apparatus according to any one of Items 1 to 10, the processing conditions for the substrate by the processing section may include a molecular descriptor of the resin member included in the processing section.

[0126] In this case, a weight change prediction model can be easily generated using the characteristics of the resin member as explanatory variables.

[0127] (12) In the substrate processing apparatus according to the 11th aspect, the molecular descriptor of the resin member may be a molecular descriptor included in a force field descriptor.

[0128] According to this configuration, even when the resin member is made of a polymer compound, it is possible to appropriately generate a weight change prediction model.

[0129] (Item 13) In the substrate processing apparatus described in Item 12, the force field descriptor may include a spring constant of a chemical bond, a force constant for bond bending, and a rotational barrier height of a dihedral angle, which are used in a molecular dynamics calculation that describes the behavior of molecules in the resin member.

[0130] In this case, the change in weight of the resin member can be predicted with high accuracy.

[0131] (Item 14) In the substrate processing apparatus described in item 12 or 13, the force field descriptor includes force field parameters used in molecular dynamics calculations that describe the behavior of molecules in the resin member, and the molecular descriptor of the resin member may be extracted from a distribution curve in which a histogram of the force field parameters is made continuous.

[0132] According to this configuration, even when the types of resin members are different, molecular descriptors can be uniformly extracted according to a common extraction rule.

[0133] (Item 15) In the substrate processing apparatus described in any one of items 1 to 14, the processing conditions for the substrate in the learning dataset may include a crystallinity label for identifying whether the resin member is crystalline or non-crystalline.

[0134] According to this configuration, it is possible to generate a weight change prediction model that predicts with higher accuracy the weight change of the resin member when a substrate process using the process liquid is performed.

[0135] (Item 16) A substrate processing method according to Item 16 includes predicting a weight change of a resin member using a weight change prediction model based on processing conditions when substrate processing is performed using a processing liquid in a processing unit including a resin member that comes into contact with the processing liquid, and is executed by a processor, wherein the weight change prediction model is an inference model obtained by machine learning from a learning dataset including a pair of substrate processing conditions and weight changes of the resin member after substrate processing is performed by the processing unit under those processing conditions.

[0136] According to this substrate processing method, a weight change prediction model is used to predict a weight change of a resin member when a substrate is processed using a processing liquid in a processing section. In this case, a user can recognize the weight change of the resin member and uniformly determine the timing for maintenance of the resin member. This allows the resin member to be maintained at an appropriate timing.

Claims

1. A substrate processing apparatus comprising a processing unit including a resin member that comes into contact with a processing liquid, the processing unit using a weight change prediction model to predict weight changes in the resin member based on processing conditions when substrate processing is performed using a processing liquid, the weight change prediction model being an inference model generated by machine learning of a learning dataset including a pair of substrate processing conditions and weight changes in the resin member after substrate processing is performed by the processing unit under those processing conditions.

2. The substrate processing apparatus according to claim 1, further comprising a life evaluation unit that evaluates the life of said resin member based on the change in weight of said resin member predicted by said weight prediction unit.

3. A substrate processing apparatus as described in claim 2, wherein the life evaluation unit accumulates the weight change of the resin member predicted by the weight prediction unit, and evaluates the end of the life of the resin member as the point at which the accumulated weight change reaches a predetermined threshold value.

4. The substrate processing apparatus according to claim 2 or 3, further comprising a processing control section that changes processing conditions for the substrate by the processing section based on the life of the resin member evaluated by the life evaluation section.

5. A substrate processing apparatus as described in claim 4, wherein a plurality of the processing sections are provided, and the processing control section changes the processing conditions for the substrate by any of the processing sections so that the lives of the plurality of resin members in the plurality of processing sections evaluated by the life evaluation section become closer to uniform.

6. A substrate processing apparatus as described in claim 5, wherein the processing control unit changes the substrate processing conditions for a processing unit among the plurality of processing units that includes a resin member that has been evaluated by the life evaluation unit to have a longer life than other processing units, so that the contact time between the resin member and the processing liquid is longer than before the change.

7. A substrate processing apparatus as described in claim 5, wherein the processing control unit changes the substrate processing conditions for a processing unit among the plurality of processing units that includes a resin member that has been evaluated by the life evaluation unit to have a shorter life than other processing units, so that the contact time between the resin member and the processing liquid is shorter than before the change.

8. A substrate processing apparatus as described in claim 2 or 3, further comprising a processing control unit, wherein a plurality of processing units are provided, substrates to be processed are loaded into the plurality of processing units in an order determined by transport conditions, and the processing control unit changes the substrate transport conditions based on the lifespan of the resin members evaluated by the life evaluation unit so that the lifespans of the plurality of resin members in the plurality of processing units become closer to uniform.

9. A substrate processing apparatus as described in claim 8, wherein the processing control unit changes the substrate transport conditions so that a processing unit containing a resin member that has been evaluated by the life evaluation unit to have a longer life than other processing units among the plurality of processing units is moved up in the transport order of the substrate.

10. A substrate processing apparatus as described in claim 8, wherein the processing control unit changes the substrate transport conditions so that a processing unit containing a resin component that has been evaluated by the life evaluation unit to have a shorter life than other processing units among the plurality of processing units is moved down in the transport order of the substrate.

11. The substrate processing apparatus according to any one of claims 1 to 3, wherein the processing conditions for the substrate by the processing section include a molecular descriptor of the resin member included in the processing section.

12. The substrate processing apparatus according to claim 11, wherein the molecular descriptor of the resin member is a molecular descriptor included in a force field descriptor.

13. The substrate processing apparatus according to claim 12, wherein the force field descriptors include chemical bond spring constants, bond bending force constants, and dihedral angle rotation barrier heights used in molecular dynamics calculations that describe the behavior of molecules in the resin member.

14. A substrate processing apparatus according to claim 12, wherein the force field descriptor includes force field parameters used in molecular dynamics calculations that describe the behavior of molecules in the resin member, and the molecular descriptor of the resin member is extracted from a distribution curve in which a histogram of the force field parameters is made continuous.

15. A substrate processing apparatus according to any one of claims 1 to 3, wherein the substrate processing conditions in the learning dataset include a crystallinity label for identifying whether the resin member is crystalline or amorphous.

16. A substrate processing method, which is executed by a processor, predicts a weight change of a resin member using a weight change prediction model based on processing conditions when substrate processing is performed using a processing liquid in a processing unit including a resin member that comes into contact with the processing liquid, and the weight change prediction model is an inference model obtained by machine learning from a learning dataset including a pair of substrate processing conditions and weight changes of the resin member after substrate processing is performed by the processing unit under those processing conditions.

Images