Substrate processing apparatus and substrate processing method

Abstract

A substrate processing apparatus (100) includes a substrate holding section (120), a processing liquid supply section (160), an input information acquisition section (22a), a sublimation drying processing condition information acquisition section (22b), and a control section (22). The input information acquisition section (22a) acquires input information including information of at least one of substrate information and processing liquid information. The sublimation drying processing condition information acquisition section (22b) acquires sublimation drying processing condition information for a processing target substrate from a learned model based on the input information, the sublimation drying processing condition information indicating sublimation drying processing conditions. The control section (22) controls the substrate holding section (120) and the processing liquid supply section (160) based on the sublimation drying processing condition information acquired in the sublimation drying processing condition information acquisition section (22b) to perform sublimation drying processing on the processing target substrate.

Description

TECHNICAL FIELD

[0001] The present application relates to a substrate processing apparatus, a substrate processing method, a learning data generation method, a learning method, a learning apparatus, a learning completed model generation method, and a learning completed model. BACKGROUND

[0002] A substrate processing apparatus that processes a substrate is suitably used for manufacturing of semiconductor devices and the like. In the manufacturing process of semiconductor devices, there are cases where a pattern of a trench and a structure is formed on a substrate. For example, in a stacked structure supported by a base material of a substrate, a plurality of trenches are formed in a pattern, and a plurality of memory elements are formed for each trench, whereby a semiconductor device provided with memory elements formed in three dimensions can be manufactured.

[0003] When a substrate provided with a pattern of a trench and a structure is processed with a chemical liquid, there are cases where the pattern of the substrate collapses when the chemical liquid dries. Therefore, a method is being studied in which, after processing a substrate with a chemical liquid, a processing liquid containing a sublimation substance is supplied to the substrate to prevent the pattern of the substrate from collapsing (see Patent Literature 1).

[0004] In the method of Patent Literature 1, after a solution containing a sublimation substance and a solvent is filled in a recess of a substrate, the solvent is evaporated, and the sublimation substance in a solid state fills the recess. Thereafter, the substrate is heated at a temperature higher than the sublimation temperature of the sublimation substance, and the sublimation substance is sublimated to be removed from the substrate, whereby the pattern of the substrate is prevented from collapsing and the substrate is dried.

[0005] PRIOR ART DOCUMENTS

[0006] PATENT LITERATURE

[0007] Patent Literature 1: Japanese Patent Application Laid-Open No. 2018-139331 SUMMARY

[0008] PROBLEMS TO BE SOLVED BY THE INVENTION

[0009] However, in the method of Patent Literature 1, there are cases where the supply amount of the solution of the sublimation substance cannot be appropriately set. When the supply amount of the solution of the sublimation substance is too small, the thickness of the sublimation substance becomes small with respect to the depth of the recess, and there is a possibility that the pattern of the substrate collapses. In addition, when the supply amount of the solution of the sublimation substance is too large, the cost and the processing time increase.

[0010] The present application has been achieved in view of the above problems, and aims to provide a substrate processing apparatus, a substrate processing method, a learning data generation method, a learning method, a learning apparatus, a learning completed model generation method, and a learning completed model that can suppress collapse of a pattern of a structure of a processing target substrate and can appropriately dry the processing target substrate.

[0011] Means for solving the problem

[0012] According to an aspect of the present application, a substrate processing apparatus includes: a substrate holding section that rotatably holds a processing target substrate provided with a pattern of a trench and a structure; a processing liquid supply section that supplies a processing liquid containing a sublimation substance and a solvent to the processing target substrate; an input information acquisition section that acquires input information for the processing target substrate, the input information including information of at least one of substrate information and processing liquid information, the substrate information showing attributes or formation conditions of the trench and the structure, the processing liquid information showing attributes of the processing liquid; a sublimation drying processing condition information acquisition section that acquires sublimation drying processing condition information for the processing target substrate from a learned model based on the input information, the sublimation drying processing condition information showing sublimation drying processing conditions; and a control section that controls the substrate holding section and the processing liquid supply section based on the sublimation drying processing condition information acquired in the sublimation drying processing condition information acquisition section to perform sublimation drying processing on the processing target substrate. The learned model is constructed by machine learning of learning data for a learning target substrate provided with a pattern of a trench and a structure, the learning data being given information associated with at least one of substrate information showing attributes or formation conditions of the trench and the structure and processing liquid information showing attributes of a processing liquid containing a sublimation substance and a solvent used when the learning target substrate is subjected to sublimation drying processing, sublimation drying processing condition information showing conditions for the learning target substrate to be subjected to sublimation drying processing, and processing result information showing a result of the learning target substrate being subjected to sublimation drying processing with the processing liquid.

[0013] In one of the embodiments, the substrate processing apparatus further includes a storage section that stores the learned model.

[0014] In one of the embodiments, the substrate information includes information showing any one of a depth of the trench, a width of the trench, and an aspect ratio of the trench for each of the processing target substrate and the learning target substrate.

[0015] In one of the embodiments, the processing liquid information includes information showing any one of a concentration of the sublimation substance with respect to the processing liquid and a temperature for each of the processing target substrate and the learning target substrate.

[0016] In one embodiment, the sublimation drying processing condition information includes information indicating any one of a supply amount of the processing liquid, a spray pattern of the processing liquid, and a rotation speed of the substrate, for each of the processing target substrate and the learning target substrate.

[0017] In one embodiment, the input information includes temperature information indicating a temperature of the processing target substrate in the sublimation drying processing.

[0018] According to another aspect of the present application, a substrate processing method includes the steps of: rotatably holding a processing target substrate provided with a pattern of grooves and structures; acquiring input information for the processing target substrate, the input information including information of at least one of substrate information indicating properties or formation conditions of the grooves and structures, and processing liquid information indicating properties of a processing liquid including a sublimation substance and a solvent; acquiring sublimation drying processing condition information from a learned model based on the input information, the sublimation drying processing condition information indicating sublimation drying processing conditions of the processing target substrate; and sublimation drying processing the processing target substrate in accordance with the sublimation drying processing condition information. In the step of acquiring the sublimation drying processing condition information, the learned model is constructed by machine learning of learning data for a learning target substrate provided with a pattern of grooves and structures, the learning data being given in association with: information of at least one of substrate information indicating properties or formation conditions of the grooves and structures, and processing liquid information indicating properties of a processing liquid including a sublimation substance and a solvent used when sublimation drying processing the learning target substrate; sublimation drying processing condition information indicating conditions of sublimation drying processing performed on the learning target substrate; and processing result information indicating a result of the sublimation drying processing performed on the learning target substrate.

[0019] According to another aspect of the present application, a learning data generation method includes the steps of: acquiring information of at least one of substrate information and processing liquid information from time series data output from a substrate processing apparatus that processes a learning target substrate on which a pattern of a groove and a structure is provided, the substrate information showing an attribute or a formation condition of the groove and the structure, the processing liquid information showing an attribute of a processing liquid containing a sublimable substance and a solvent used when the learning target substrate is subjected to a sublimation drying process; acquiring sublimation drying process condition information showing a sublimation drying process condition when the learning target substrate is subjected to a sublimation drying process in the substrate processing apparatus from the time series data; acquiring processing result information showing a result of the sublimation drying process of the learning target substrate in the substrate processing apparatus from the time series data; and associating the information of at least one of the substrate information and the processing liquid information, the sublimation drying process condition information, and the processing result information with respect to the learning target substrate and storing them as learning data in a storage unit.

[0020] According to another aspect of the present application, a learning method includes the steps of: acquiring learning data generated according to the above-described learning data generation method; and inputting the learning data to a learning program and performing machine learning on the learning data.

[0021] According to another aspect of the present application, a learning device includes: a storage unit that stores learning data generated according to the above-described learning data generation method; and a learning unit that inputs the learning data to a learning program and performs machine learning on the learning data.

[0022] According to another aspect of the present application, a learned model generation method includes the steps of: acquiring learning data generated according to the above-described learning data generation method; and generating a learned model constructed by performing machine learning on the learning data.

[0023] According to another aspect of the present application, a learned model is constructed by performing machine learning on learning data generated according to the above-described learning data generation method.

[0024] According to another aspect of the present application, a substrate processing apparatus includes: a substrate holding section that rotatably holds a substrate on which a pattern of a trench and a structure is provided; a processing liquid supply section that supplies a processing liquid containing a sublimation substance and a solvent to the substrate; a storage section that stores a conversion table to which information associated with at least one of substrate information and processing liquid information and sublimation drying processing condition information are given, the substrate information showing an attribute or a formation condition of the trench and the structure, the processing liquid information showing an attribute of the processing liquid containing the sublimation substance and the solvent, the sublimation drying processing condition information showing a condition of sublimation drying processing; an input information acquisition section that acquires information of at least one of the substrate information showing the attribute or the formation condition of the trench and the structure and the processing liquid information showing an attribute of the processing liquid supplied in the processing liquid supply section as input information; a sublimation drying processing condition information acquisition section that acquires sublimation drying processing condition information for the substrate based on the input information using the conversion table, the sublimation drying processing condition information showing a condition of sublimation drying processing; and a control section that controls the substrate holding section and the processing liquid supply section based on the sublimation drying processing condition information acquired in the sublimation drying processing condition information acquisition section to subject the substrate to sublimation drying processing.

[0025] Effects of the Invention

[0026] According to the present application, it is possible to suppress collapse of a pattern of a structure of a processing target substrate and to appropriately dry the processing target substrate. BRIEF DESCRIPTION OF DRAWINGS

[0027] Figure 1 is a schematic view of a substrate processing learning system including the substrate processing apparatus of the present embodiment.

[0028] Figure 2 is a schematic view of a substrate processing system including the substrate processing apparatus of the present embodiment.

[0029] Figure 3 is a schematic view of the substrate processing apparatus of the present embodiment.

[0030] Figure 4 is a block diagram of a substrate processing system including the substrate processing apparatus of the present embodiment.

[0031] Figure 5 In (a), (b) is a flowchart of a sublimation drying processing in the substrate processing method of the present embodiment.

[0032] Figure 6 In (a), (b) is a flowchart of a sublimation drying processing in the substrate processing method of the present embodiment.

[0033] Figure 7 In (a) to (c), schematic diagrams showing a substrate processing method in the substrate processing apparatus of the present embodiment are shown.

[0034] Figure 8 is a block diagram of a substrate processing system equipped with the substrate processing apparatus of the present embodiment and a learning data generation apparatus.

[0035] Figure 9 is a flowchart showing a learning data generation method of the present embodiment.

[0036] Figure 10 is a block diagram of a learning data generation apparatus and a learning apparatus of the present embodiment.

[0037] Figure 11 is a flowchart showing a learning method and a generation method of a learned model of the present embodiment.

[0038] Figure 12 is a diagram showing learning data input to the learning apparatus of the present embodiment.

[0039] Figure 13 is a diagram showing learning data input to the learning apparatus of the present embodiment.

[0040] Figure 14 In (a) to (e), schematic diagrams showing a substrate processing method in the substrate processing apparatus of the present embodiment are shown.

[0041] Figure 15 is a diagram showing learning data input to the learning apparatus of the present embodiment.

[0042] Figure 16 In (a) to (c), schematic diagrams showing a substrate processing method in the substrate processing apparatus of the present embodiment are shown.

[0043] Figure 17 is a diagram showing learning data input to the learning apparatus of the present embodiment.

[0044] Figure 18 is a diagram showing learning data input to the learning apparatus of the present embodiment.

[0045] Figure 19 In (a), a schematic diagram of a processing target substrate having a pattern of a trench and a structure which is a processing target of the substrate processing apparatus of the present embodiment is shown, (b) is a schematic diagram of a processing liquid supply portion in the substrate processing apparatus, and (c) is a diagram showing a sublimation drying processing condition obtained based on substrate information and processing liquid information.

[0046] Figure 20 is a graph showing learning data input to a learning device of the present embodiment.

[0047] Figure 21 is a schematic view of a substrate processing device of the present embodiment.

[0048] Figure 22 is a graph showing learning data input to a learning device of the present embodiment.

[0049] Figure 23 is a schematic view of a substrate processing learning system provided with a substrate processing device of the present embodiment.

[0050] Figure 24 is a schematic view of a substrate processing device of the present embodiment.

[0051] Figure 25 is a graph showing learning data input to a learning device of the present embodiment.

[0052] Figure 26 is a block diagram of a substrate processing device of the present embodiment.

[0053] Figure 27 is a graph showing a conversion table in a substrate processing device of the present embodiment. DETAILED DESCRIPTION

[0054] Embodiments of a substrate processing device, a substrate processing method, a learning data generation method, a learning method, a learning device, a learned model generation method, and a learned model will be described below with reference to the accompanying drawings. In the drawings, the same reference numerals are attached to the same or equivalent portions, and the description will not be repeated. In the present specification, for the sake of easy understanding of the present application, cases where an X direction, a Y direction, and a Z direction orthogonal to each other are described. Typically, the X direction and the Y direction are parallel to a horizontal direction, and the Z direction is parallel to a vertical direction.

[0055] First, referring to Figure 1 a substrate processing learning system 200 provided with a substrate processing device 100 of the present embodiment will be described. Figure 1 is a schematic view of a substrate processing learning system 200.

[0056] As shown in Figure 1 , the substrate processing learning system 200 is provided with the substrate processing device 100, a substrate processing device 100L, a learning data generation device 300, and a learning device 400.

[0057] The substrate processing apparatus 100 processes a processing target substrate. Here, a pattern of a trench and a structure body is provided in the processing target substrate, and the substrate processing apparatus 100 performs sublimation drying processing on the processing target substrate. In addition, the substrate processing apparatus 100 can perform processing other than sublimation drying processing on the processing target substrate. The substrate processing apparatus 100 is of a single-wafer type, and processes the processing target substrate on a wafer-by-wafer basis. Typically, the processing target substrate is a substantially circular plate.

[0058] The substrate processing apparatus 100L processes a learning target substrate. Here, a pattern of a trench and a structure body is provided in the learning target substrate, and the substrate processing apparatus 100L performs sublimation drying processing on the learning target substrate. In addition, the substrate processing apparatus 100L can perform processing other than sublimation drying processing on the learning target substrate. The learning target substrate has the same configuration as the processing target substrate. The substrate processing apparatus 100L is of a single-wafer type, and processes the processing target substrate on a wafer-by-wafer basis. Typically, the processing target substrate is a substantially circular plate. The substrate processing apparatus 100L has the same configuration as the substrate processing apparatus 100. The substrate processing apparatus 100L can be the same substrate processing apparatus as the substrate processing apparatus 100. For example, the same substrate processing apparatus can have processed the learning target substrate in the past, and then processes the processing target substrate. Alternatively, the substrate processing apparatus 100L can be another article having the same configuration as the substrate processing apparatus 100.

[0059] In the following description of the present specification, the learning target substrate is referred to as a "learning target substrate WL", and the processing target substrate is referred to as a "processing target substrate Wp". In addition, when the learning target substrate WL and the processing target substrate Wp are not distinguished, the learning target substrate WL and the processing target substrate Wp are referred to as a "substrate W".

[0060] The substrate W is, for example, a semiconductor wafer, a substrate for a liquid crystal display device, a substrate for a plasma display, a substrate for a field emission display (FED), a substrate for an optical disc, a substrate for a magnetic disc, a substrate for an optical magnetic disc, a substrate for a photomask, a ceramic substrate, or a substrate for a solar cell.

[0061] The substrate processing apparatus 100L outputs time series data TDL. The time series data TDL is data to show a time change of a physical quantity in the substrate processing apparatus 100L. The time series data TDL shows a time change of a physical quantity (value) that changes time series in a predetermined period. For example, the time series data TDL is data to show a time change of a physical quantity related to a process performed by the substrate processing apparatus 100L on a learning target substrate. Alternatively, the time series data TDL is data to show a time change of a physical quantity related to a characteristic of a learning target substrate processed by the substrate processing apparatus 100L. Alternatively, the time series data TDL can also include data to show a manufacturing process before the learning target substrate is processed by the substrate processing apparatus 100L.

[0062] Further, the value shown in the time series data TDL can also be a value directly measured in a measuring machine. Alternatively, the value shown in the time series data TDL can also be a value obtained by performing an arithmetic process on a value directly measured in a measuring machine. Alternatively, the value shown in the time series data TDL can also be a value obtained by performing an arithmetic process on values measured in a plurality of measuring machines.

[0063] The learning data generating apparatus 300 generates learning data LD based on the time series data TDL or at least a part of the time series data TDL. The learning data generating apparatus 300 outputs the learning data LD.

[0064] The learning data LD includes information of at least one of substrate information and processing liquid information of the learning target substrate WL, sublimation drying process condition information, and process result information. The substrate information of the learning target substrate WL shows attributes or formation conditions of a trench and a structure of the learning target substrate WL. The substrate information can also be information of the learning target substrate WL measured before the learning target substrate WL is subjected to the sublimation drying process. For example, the substrate information can also be information obtained from the learning target substrate WL before being carried into the substrate processing apparatus 100L. Alternatively, the substrate information can also be information related to a process performed on the learning target substrate WL before the sublimation drying process.

[0065] The processing liquid information shows attributes of a processing liquid including a sublimation substance and a solvent used for the sublimation drying process performed on the learning target substrate WL. The sublimation drying process condition information shows a process condition of the sublimation drying process performed on the learning target substrate WL. The process result information shows a result of the sublimation drying process performed on the learning target substrate WL. The process result information includes collapse information showing collapse of a pattern of a structure.

[0066] The learning device 400 performs machine learning on the learning data LD, thereby generating a learned model LM. The learning device 400 outputs the learned model LM.

[0067] The substrate processing device 100 outputs time-series data TD. The time-series data TD is data to show a time change of a physical quantity in the substrate processing device 100. The time-series data TD shows a time change of a physical quantity (value) that changes in time series in a predetermined period. For example, the time-series data TD is data to show a time change of a physical quantity related to a process performed by the substrate processing device 100 on a process target substrate. Alternatively, the time-series data TD is data to show a time change of a physical quantity related to a characteristic of a process target substrate processed by the substrate processing device 100.

[0068] Further, the value shown in the time-series data TD can be a value directly measured in a measurement machine. Alternatively, the value shown in the time-series data TD can be a value obtained by performing an arithmetic process on a value directly measured in a measurement machine. Alternatively, the value shown in the time-series data TD can be a value obtained by performing an arithmetic process on values measured in a plurality of measurement machines. Alternatively, the time-series data TD can include data to show a manufacturing process before the process target substrate is processed by the substrate processing device 100.

[0069] The object used by the substrate processing device 100 corresponds to the object used by the substrate processing device 100L. Therefore, the constitution of the object used by the substrate processing device 100 is the same as the constitution of the object used by the substrate processing device 100L. In addition, in the time-series data TD, the physical quantity of the object used by the substrate processing device 100 corresponds to the physical quantity of the object used by the substrate processing device 100L. Therefore, the physical quantity of the object used by the substrate processing device 100L is the same as the physical quantity of the object used by the substrate processing device 100.

[0070] The input information Cp for the process target substrate Wp is generated from the time-series data TD. The input information Cp for the process target substrate Wp includes at least one of substrate information of the process target substrate Wp and process liquid information. The substrate information of the process target substrate Wp shows attributes or formation conditions of a trench and a structure of the process target substrate Wp. The process liquid information shows attributes of a process liquid containing a sublimation substance and a solvent used for the sublimation drying process performed on the process target substrate Wp.

[0071] The input information Cp of the processing target substrate Wp can also be acquired by measuring the processing target substrate Wp before the processing target substrate Wp is subjected to the sublimation drying process. Alternatively, the input information Cp can also be acquired before the processing target substrate Wp is carried into the substrate processing apparatus 100. For example, the input information Cp can also be information related to a process performed on the processing target substrate Wp before the sublimation drying process.

[0072] The sublimation drying process condition information Rp is output from the learned model LM based on the input information Cp of the processing target substrate Wp, and the sublimation drying process condition information Rp shows a sublimation drying process condition suitable for the processing target substrate Wp in the substrate processing apparatus 100.

[0073] As described above with reference to Figure 1 , according to the present embodiment, the learning apparatus 400 performs machine learning. Therefore, a learned model LM with high accuracy can be generated from a time series data TDL that is very complex and has a large amount of analysis targets. Further, the input information Cp of the time series data TD from the processing target substrate Wp is input to the learned model LM, and the sublimation drying process condition information Rp for showing a sublimation drying process condition is output from the learned model LM. Therefore, a sublimation drying process suitable for the processing target substrate Wp can be performed.

[0074] Next, a substrate processing system 10 provided with the substrate processing apparatus 100 of the present embodiment will be described with reference to Figure 2 Figure 2 is a schematic plan view of the substrate processing system 10.

[0075] The substrate processing system 10 processes a substrate W. The substrate processing system 10 is provided with a plurality of substrate processing apparatuses 100. The substrate processing apparatus 100 processes the substrate W in such a manner that at least one of etching, surface treatment, property imparting, process film formation, removal of at least a part of a film, and cleaning is performed on the substrate W.

[0076] As shown in Figure 2 , the substrate processing system 10 is provided with a fluid cabinet 32, a fluid tank 34, a plurality of load ports LP, an index robot IR, a central robot CR, and a control apparatus 20 in addition to the plurality of substrate processing apparatuses 100. The control apparatus 20 controls the load ports LP, the index robot IR, and the central robot CR.

[0077] ​The plurality of load ports LP are stacked and accommodate a plurality of substrates W. An indexing robot IR transports the substrates W between the load ports LP and a central robot CR. In addition, a device can also be configured in which a setting table (passing table) for temporarily placing the substrates W is provided between the indexing robot IR and the central robot CR, and the substrates W are indirectly picked up and passed between the indexing robot IR and the central robot CR via the setting table. The central robot CR transports the substrates W between the indexing robot IR and the substrate processing apparatus 100. Each of the substrate processing apparatuses 100 ejects a liquid onto the substrate W and processes the substrate W. The liquid includes a chemical liquid, a rinse liquid, a replacement liquid, and / or a processing liquid. Alternatively, the liquid can include another liquid. The fluid cabinet 32 accommodates the liquid. In addition, the fluid cabinet 32 can also accommodate a gas.

[0078] Specifically, the plurality of substrate processing apparatuses 100 form a plurality of towers TW (four towers TW in the present embodiment) that are arranged around the central robot CR in a plan view. Each of the towers TW includes a plurality of substrate processing apparatuses 100 (three substrate processing apparatuses 100 in the present embodiment) that are stacked vertically. The fluid cabinet 32 includes a plurality of fluid tanks 34 corresponding to the plurality of towers TW. The liquid in the fluid cabinet 32 is supplied to all of the substrate processing apparatuses 100 included in the tower TW corresponding to the fluid tank 34 via the fluid tank 34. In addition, the gas in the fluid cabinet 32 is supplied to all of the substrate processing apparatuses 100 included in the tower TW corresponding to the fluid tank 34 via the fluid tank 34. Figure 2 Figure 2 The control apparatus 20 controls various actions of the substrate processing system 10. The control apparatus 20 includes a control section 22 and a storage section 24. The control section 22 has a processor. The control section 22 has, for example, a central processing unit (CPU). Alternatively, the control section 22 can have a general-purpose computer.

[0079] The storage section 24 stores data and a computer program. The data includes recipe data. The recipe data includes information for displaying a plurality of recipes. Each of the plurality of recipes specifies a processing content and a processing procedure of the substrate W.

[0080] The storage section 24 includes a main storage device and an auxiliary storage device. The main storage device is, for example, a semiconductor memory. The auxiliary storage device is, for example, a semiconductor memory and / or a hard disk drive. The storage section 24 can also include a removable medium. The control section 22 executes the computer program stored in the storage section 24, thereby performing a substrate processing action.

[0081] The storage section 24 includes a main storage device and an auxiliary storage device. The main storage device is, for example, a semiconductor memory. The auxiliary storage device is, for example, a semiconductor memory and / or a hard disk drive. The storage section 24 can also include a removable medium. The control section 22 executes the computer program stored in the storage section 24, thereby performing a substrate processing action.

[0082] ​A computer program in which the processes are determined in advance is stored in the storage section 24. The substrate processing apparatus 100 operates in accordance with the processes determined by the computer program.

[0083] Further, although Figure 2 The substrate processing system 10 is provided with one control apparatus 20, but the control apparatus 20 can be provided for each of the substrate processing apparatuses 100. However, in this case, it is preferable that the substrate processing system 10 be provided with another control apparatus for controlling the plurality of substrate processing apparatuses 100 and apparatuses other than the substrate processing apparatuses 100.

[0084] Next, the substrate processing apparatus 100 according to the present embodiment will be described with reference to Figure 3 The substrate processing apparatus 100 according to the present embodiment will be described. Figure 3 A schematic view of the substrate processing apparatus 100 according to the present embodiment. Further, although the configuration of the substrate processing apparatus 100 will be described, the substrate processing apparatus 100L also has the same configuration as the substrate processing apparatus 100.

[0085] The substrate processing apparatus 100 processes the substrate W. The substrate processing apparatus 100 is provided with a chamber 110, a substrate holding section 120, a chemical liquid supply section 130, a cleaning liquid supply section 140, a replacement liquid supply section 150, a processing liquid supply section 160, a shielding member 170, and a fan filter unit (FFU) 180. The chamber 110 houses the substrate W. The substrate holding section 120 holds the substrate W. The substrate holding section 120 rotatably holds the substrate W.

[0086] The chamber 110 is a substantially box shape having an internal space. The chamber 110 houses the substrate W. Here, the substrate processing apparatus 100 is of a single-wafer type for processing the substrate W one by one, and the substrate W is housed in the chamber 110 one by one. The substrate W is housed in the chamber 110 and is processed in the chamber 110. In the chamber 110, the substrate holding section 120, the chemical liquid supply section 130, the cleaning liquid supply section 140, the replacement liquid supply section 150, the processing liquid supply section 160, and the shielding member 170 are each housed at least in part. On the other hand, the FFU 180 is installed outside the upper surface of the chamber 110.

[0087] The substrate holding section 120 holds the substrate W. The substrate holding section 120 holds the substrate W horizontally in such a manner that the upper surface Wa of the substrate W faces upward and the back surface (lower surface) Wb of the substrate W faces vertically downward. Further, the substrate holding section 120 rotates the substrate W while holding the substrate W.

[0088] For example, the substrate holding portion 120 can be a chuck type that holds the end portions of the substrate W. Alternatively, the substrate holding portion 120 can have any mechanism that holds the substrate W from the back surface Wb. For example, the substrate holding portion 120 can be a vacuum type. In this case, the substrate holding portion 120 adsorbs the central portion of the back surface Wb of the substrate W, which is a non-device formation surface, to the upper surface, thereby horizontally holding the substrate W. Alternatively, the substrate holding portion 120 can combine a chuck type that contacts a plurality of chuck pins to the peripheral end surface of the substrate W with a vacuum type.

[0089] For example, the substrate holding portion 120 includes a spin base 121, a chuck member 122, a shaft 123, and a motor 124. The chuck member 122 is provided to the spin base 121. The chuck member 122 holds the substrate W. Typically, a plurality of chuck members 122 are provided to the spin base 121.

[0090] The shaft 123 is a hollow shaft. The shaft 123 extends in the vertical direction along the rotation axis Ax. The spin base 121 is coupled to the upper end of the shaft 123. The back surface Wb of the substrate W faces the spin base 121, and the substrate W is placed above the spin base 121.

[0091] The spin base 121 is a circular plate that horizontally supports the substrate W. The shaft 123 extends downward from the central portion of the spin base 121. The motor 124 applies a rotational force to the shaft 123. The motor 124 rotates the shaft 123 in the rotational direction, thereby rotating the substrate W and the spin base 121 about the rotation axis Ax. Here, the rotational direction is the counterclockwise direction.

[0092] The chemical liquid supply portion 130 supplies a chemical liquid to the substrate W. Thereby, the substrate W is processed with the chemical liquid.

[0093] For example, the chemical liquid includes hydrofluoric acid (hydrogen fluoride water: HF). Alternatively, the chemical liquid can be a liquid including at least one of sulfuric acid, acetic acid, nitric acid, hydrochloric acid, citric acid, buffered hydrofluoric acid (BHF), diluted hydrofluoric acid (DHF), ammonia water, diluted ammonia water, hydrogen peroxide water, an organic base (for example, TMAH: tetramethylammonium hydroxide), and the like), a surfactant, and an anticorrosive agent. Alternatively, the chemical liquid can be a mixed liquid that mixes the above liquids. As an example of a chemical liquid that mixes these liquids, for example, SPM (sulfuric acid hydrogen peroxide water mixed liquid), SC1 (ammonia water hydrogen peroxide water mixed liquid), SC2 (hydrochloric acid hydrogen peroxide water mixed liquid), and the like can be given.

[0094] The chemical liquid supply section 130 includes a nozzle 132, a pipe 134, and a valve 136. The nozzle 132 faces the upper surface Wa of the substrate W and ejects a chemical liquid toward the upper surface Wa of the substrate W. The pipe 134 is coupled to the nozzle 132. The nozzle 132 is provided at a front end of the pipe 134. The pipe 134 is supplied with a chemical liquid from a supply source. The valve 136 is provided to the pipe 134. The valve 136 opens and closes a flow path in the pipe 134.

[0095] The chemical liquid supply section 130 further includes a nozzle moving section 138. The nozzle moving section 138 moves the nozzle 132 between an ejection position and a retreat position. In a case where the nozzle 132 is located at the ejection position, the nozzle 132 is located above the substrate W. In a case where the nozzle 132 is located at the ejection position, the nozzle 132 ejects a chemical liquid toward the upper surface Wa of the substrate W. In a case where the nozzle 132 is located at the retreat position, the nozzle 132 is located further outside in a diameter direction of the substrate W than the substrate W.

[0096] The nozzle moving section 138 includes an arm 138a, a rotation shaft 138b, and a moving mechanism 138c. The arm 138a extends in a substantially horizontal direction. The nozzle 132 is attached to a front end portion of the arm 138a. The arm 138a is coupled to the rotation shaft 138b. The rotation shaft 138b extends in a substantially vertical direction. The moving mechanism 138c rotates the rotation shaft 138b about a rotation axis in the substantially vertical direction and rotates the arm 138a about a substantially horizontal plane. As a result, the nozzle 132 moves about the substantially horizontal plane. For example, the moving mechanism 138c includes an arm swing motor that rotates the rotation shaft 138b about the rotation axis. The arm swing motor is, for example, a servo motor. Further, the moving mechanism 138c raises and lowers the rotation shaft 138b in the substantially vertical direction, thereby raising and lowering the arm 138a. As a result, the nozzle 132 moves in the substantially vertical direction. For example, the moving mechanism 138c includes a ball screw mechanism and an arm raising and lowering motor that imparts a driving force to the ball screw mechanism. The arm raising and lowering motor is, for example, a servo motor.

[0097] The cleaning liquid supply section 140 supplies a cleaning liquid to the substrate W. The cleaning liquid can include any one of deionized water (DIW), carbonated water, electrolytic ion water, ozone water, ammonia water, hydrochloric acid water of a dilute concentration (for example, about 10 ppm to 100 ppm), or reduced water (hydrogen water).

[0098] The cleaning liquid supply section 140 includes a nozzle 142, a pipe 144, and a valve 146. The nozzle 142 faces the upper surface Wa of the substrate W and ejects a cleaning liquid toward the upper surface Wa of the substrate W. The pipe 144 is coupled to the nozzle 142. The nozzle 142 is provided at a front end of the pipe 144. The pipe 144 is supplied with a cleaning liquid from a supply source. The valve 146 is provided to the pipe 144. The valve 146 opens and closes a flow path in the pipe 144.

[0099] The cleaning liquid supply section 140 further includes a nozzle moving section 148. The nozzle moving section 148 moves the nozzle 142 between an ejection position and a retreat position. In a case where the nozzle 142 is located at the ejection position, the nozzle 142 is located above the substrate W. In a case where the nozzle 142 is located at the ejection position, the nozzle 142 ejects the cleaning liquid toward the upper surface Wa of the substrate W. In a case where the nozzle 142 is located at the retreat position, the nozzle 142 is located further outside in the diameter direction of the substrate W than the substrate W.

[0100] The nozzle moving section 148 includes an arm 148a, a turning shaft 148b, and a moving mechanism 148c. For example, the nozzle moving section 148 has the same configuration as the nozzle moving section 138.

[0101] The displacement liquid supply section 150 supplies the displacement liquid to the substrate W. As described later, after the displacement liquid is supplied to the upper surface of the substrate W covered with the liquid film of the cleaning liquid, the treatment liquid is supplied to the upper surface of the substrate W covered with the liquid film of the displacement liquid. The displacement liquid is a liquid that is dissolved with both the cleaning liquid and the treatment liquid.

[0102] For example, the displacement liquid is IPA (isopropyl alcohol) or HFE (hydrofluoroether). The displacement liquid can also be a mixed liquid of IPA and HFE, or can include at least one of IPA and HFE and a component other than IPA and HFE. IPA and HFE are liquids that are dissolved with both water and a fluorohydrocarbon compound.

[0103] The displacement liquid supply section 150 includes a nozzle 152, a pipe 154, and a valve 156. The nozzle 152 faces the upper surface Wa of the substrate W and ejects the displacement liquid toward the upper surface Wa of the substrate W. The pipe 154 is coupled to the nozzle 152. The nozzle 152 is provided at the front end of the pipe 154. The pipe 154 is supplied with the displacement liquid from a supply source. The valve 156 is provided to the pipe 154. The valve 156 opens and closes a flow path in the pipe 154.

[0104] The displacement liquid supply section 150 further includes a nozzle moving section 158. The nozzle moving section 158 moves the nozzle 152 between an ejection position and a retreat position. In a case where the nozzle 152 is located at the ejection position, the nozzle 152 is located above the substrate W. In a case where the nozzle 152 is located at the ejection position, the nozzle 152 ejects the displacement liquid toward the upper surface Wa of the substrate W. In a case where the nozzle 152 is located at the retreat position, the nozzle 152 is located further outside in the diameter direction of the substrate W than the substrate W.

[0105] The nozzle moving section 158 includes an arm 158a, a turning shaft 158b, and a moving mechanism 158c. For example, the nozzle moving section 158 has the same configuration as the nozzle moving section 138.

[0106] The processing liquid supply section 160 supplies a processing liquid to the substrate W. The processing liquid is a solution containing a sublimation substance and a solvent. The sublimation substance corresponds to a solute, and the solvent and the sublimation substance are dissolved in each other. The sublimation substance can also be a substance that changes from a solid to a gas without passing through a liquid at normal temperature (synonymous with room temperature) or normal pressure (pressure in the substrate processing apparatus 100, for example, a value of about one atmosphere). The solvent can be such a substance, or a substance other than such a substance. That is, the processing liquid can also contain two or more kinds of substances that change from a solid to a gas without passing through a liquid at normal temperature or normal pressure.

[0107] The sublimation substance can also be, for example, any one of 2-methyl-2-propanol (also known as an alcohol such as tert-butyl alcohol (t-butyl alcohol or tertiary butyl alcohol) or cyclohexanol), a fluorinated hydrocarbon compound, 1,3,5-trioxane (also known as trioxane), camphor (also known as campho or camphora), naphthalene, iodine, cyclohexane, and cyclohexanone oxime, or a substance other than the above.

[0108] The solvent is, for example, at least one selected from the group consisting of pure water, IPA, HFE (hydrofluoroether), acetone, PGMEA (propylene glycol monomethyl ether acetate), PGEE (propylene glycol ethyl ether) (also known as 1-ethoxy-2-propanol), ethylene glycol, and hydrofluorocarbon.

[0109] For example, the sublimation substance is preferably cyclohexanone oxime, and the solvent is preferably IPA. In this case, the vapor pressure of IPA is higher than that of cyclohexanone oxime at room temperature (23°C or a value near thereto). The freezing point of cyclohexanone oxime is 90.5°C, and the boiling point is 210C; the freezing point of IPA is -90°C, and the boiling point is 83°C. Here, the substrate processing apparatus 100 is disposed in a clean room maintained at room temperature, and the processing liquid can maintain a state in which the sublimation substance is dissolved in the solvent as a liquid.

[0110] The processing fluid supply unit 160 includes a nozzle 162, a pipe 164, and a valve 166. The nozzle 162 faces the upper surface Wa of the substrate W and sprays processing fluid towards the upper surface Wa of the substrate W. The pipe 164 is connected to the nozzle 162. The nozzle 162 is located at the front end of the pipe 164. Processing fluid is supplied to the pipe 164 from a supply source. The valve 166 is located on the pipe 164. The valve 166 opens and closes the flow path within the pipe 164.

[0111] The processing liquid supply unit 160 further includes a nozzle moving unit 168. The nozzle moving unit 168 moves the nozzle 162 between a spraying position and a retracted position. When the nozzle 162 is in the spraying position, the nozzle 162 is located above the substrate W. When the nozzle 162 is in the spraying position, the nozzle 162 sprays the processing liquid toward the upper surface Wa of the substrate W. When the nozzle 162 is in the retracted position, the nozzle 162 is located radially outward of the substrate W.

[0112] The nozzle moving part 168 includes an arm 168a, a rotating shaft 168b, and a moving mechanism 168c. For example, the configuration of the nozzle moving part 168 is the same as that of the nozzle moving part 138.

[0113] The shielding member 170 includes a shielding plate 172, a support shaft 174, a lifting unit 176, and an inert gas supply unit 178. The shielding plate 172 is horizontally disposed above the substrate holding unit 120. The shielding plate 172 comprises a thin, circular plate. The shielding plate 172 is horizontally supported by a cylindrical support shaft 174, which extends upward from the center of the shielding plate 172. The centerline of the shielding plate 172 is located on the rotation axis Ax of the substrate W.

[0114] The lower surface 172L of the shielding plate 172 faces the upper surface of the substrate W. The lower surface 172L of the shielding plate 172 is parallel to the upper surface Wa of the substrate W and has an outer diameter greater than or equal to the diameter of the substrate W.

[0115] The lifting unit 176 raises and lowers the shielding plate 172 along the vertical direction. The lifting unit 176 moves the shielding plate 172 to any position from the upper position to the lower position. Figure 3 In the middle, the shielding plate 172 is located in the upper position. The upper position is when the shielding plate 172 is retracted to a height that allows the nozzles 132, 142, 152, and 162 to enter between the shielding member 170 and the substrate W. The lower position is when the lower surface 172L of the shielding plate 172 approaches the upper surface Wa of the substrate W, thereby preventing the nozzles 132, 142, 152, and 162 from entering between the substrate W and the shielding member 170.

[0116] The inert gas supply part 178 supplies inert gas to the substrate W. The inert gas is, for example, helium gas, argon gas, or nitrogen gas. Alternatively, the inert gas can be a gas other than helium gas, argon gas, or nitrogen gas.

[0117] The inert gas supply part 178 includes a nozzle 178a, a pipe 178b, a valve 178c, a flow rate adjustment valve 178d, and a temperature adjuster 178e. The nozzle 178a faces the upper surface Wa of the substrate W and ejects inert gas toward the upper surface Wa of the substrate W. The pipe 178b is coupled to the nozzle 178a. The nozzle 178a is provided at the front end of the pipe 178b. The pipe 178b is supplied with inert gas from a supply source.

[0118] The valve 178c, the flow rate adjustment valve 178d, and the temperature adjuster 178e are provided to the pipe 178b. The valve 178c opens and closes a flow path in the pipe 178b. The flow rate adjustment valve 178d adjusts the flow rate of inert gas passing through the flow path in the pipe 178b. The temperature adjuster 178e adjusts the temperature of the inert gas. For example, the temperature adjuster 178e heats or cools the inert gas.

[0119] The nozzle 178a ejects inert gas downward via an upper central opening that is opened in the central portion of the lower surface 172L of the shielding plate 172. The nozzle 178a extends up and down along the rotation axis Ax. The nozzle 178a is disposed in a through-hole that penetrates the central portions of the shielding plate 172 and the support shaft 174 up and down. The nozzle 178a is raised and lowered together with the shielding plate 172. The ejection port of the nozzle 178a is disposed above the upper central opening of the shielding plate 172.

[0120] The pipe 178b guides inert gas to the nozzle 178a. When the valve 178c is opened, inert gas is continuously ejected downward from the ejection port of the nozzle 178a at a flow rate corresponding to the degree of opening of the flow rate adjustment valve 178d. For example, the inert gas ejected from the nozzle 178a is nitrogen gas.

[0121] The FFU 180 is disposed outside the upper portion of the chamber 110. An air supply port is provided in the upper portion of the chamber 110, and the FFU 180 supplies air into the chamber 110 from the air supply port of the chamber 110.

[0122] The substrate processing apparatus 100 further includes a cup 190. The cup 190 recovers liquid scattered from the substrate W. The cup 190 can be raised to the side of the substrate W in the vertical upward direction. Alternatively, the cup 190 can be lowered from the side of the substrate W in the vertical downward direction.

[0123] The control device 20 controls various actions of the substrate processing device 100. The control section 22 controls the substrate holding section 120, the chemical liquid supply section 130, the cleaning liquid supply section 140, the displacement liquid supply section 150, the processing liquid supply section 160, the shielding member 170, the FFU 180, and / or the cover 190. In one example, the control section 22 controls the electric motor 124, the valves 136, 146, 156, 166, 178c, the flow rate adjustment valve 178d, the moving mechanisms 138c, 148c, 158c, 168c, the lifting unit 176, the FFU 180, and / or the cover 190.

[0124] The substrate processing device 100 of the present embodiment performs chemical liquid processing, cleaning processing, displacement processing, and sublimation drying processing on the substrate W. Further, although not illustrated in the drawings in order to avoid the drawings from being too complex, Figure 3 However, it is preferable to detect the flow rate and temperature of the chemical liquid, the cleaning liquid, the displacement liquid, the processing liquid, and the inert gas as needed by a sensor or the like.

[0125] Next, the substrate processing device 100 of the present embodiment will be described with reference to Figures 1-4 The substrate processing device 100 of the present embodiment will be described with reference to Figure 4 is a block diagram of a substrate processing system 10 provided with the substrate processing device 100.

[0126] As shown in Figure 4 , the control device 20 controls various actions of the substrate processing system 10. The control device 20 controls the index robot IR, the center robot CR, the substrate holding section 120, the chemical liquid supply section 130, the cleaning liquid supply section 140, the displacement liquid supply section 150, the processing liquid supply section 160, the shielding member 170, the FFU 180, and the cover 190. In detail, the control device 20 transmits a control signal to the index robot IR, the center robot CR, the substrate holding section 120, the chemical liquid supply section 130, the cleaning liquid supply section 140, the displacement liquid supply section 150, the processing liquid supply section 160, the shielding member 170, the FFU 180, and the cover 190, thereby controlling the index robot IR, the center robot CR, the substrate holding section 120, the chemical liquid supply section 130, the cleaning liquid supply section 140, the displacement liquid supply section 150, the processing liquid supply section 160, the shielding member 170, the FFU 180, and the cover 190.

[0127] In detail, the control section 22 controls the index robot IR, and the index robot IR picks up and delivers the substrate W.

[0128] The control section 22 controls the central robot CR to take and deliver the substrate W by the central robot CR. For example, the central robot CR takes the unprocessed substrate W and carries the substrate W into any one of the plurality of chambers 110. Further, the central robot CR takes the processed substrate W from the chamber 110 and carries the substrate W out.

[0129] The control section 22 controls the substrate holding section 120 to control the start of rotation of the substrate W, the change of the rotation speed, and the stop of rotation of the substrate W. For example, the control section 22 can control the substrate holding section 120 to change the number of rotations of the substrate holding section 120. Specifically, the control section 22 changes the number of rotations of the electric motor 124 of the substrate holding section 120, whereby the number of rotations of the substrate W can be changed.

[0130] The control section 22 controls the valve 136 of the chemical liquid supply section 130, the valve 146 of the cleaning liquid supply section 140, the valve 156 of the displacement liquid supply section 150, the valve 166 of the processing liquid supply section 160, and the valve 178c of the inert gas supply section 178, whereby the state of the valves 136, 146, 156, 166, and 178c can be switched to the open state and the closed state. Specifically, the control section 22 controls the valves 136, 146, 156, 166, and 178c and sets the valves 136, 146, 156, 166, and 178c to the open state, whereby the chemical liquid, the cleaning liquid, the displacement liquid, the processing liquid, and the inert gas flowing in the pipes 134, 144, 154, 164, and 178b toward the nozzles 132, 142, 152, 162, and 178a, respectively, can pass through. Further, the control section 22 controls the valves 136, 146, 156, 166, and 178c of the chemical liquid supply section 130, the cleaning liquid supply section 140, the displacement liquid supply section 150, the processing liquid supply section 160, and the inert gas supply section 178 and sets the valves 136, 146, 156, 166, and 178c to the closed state, whereby the chemical liquid, the cleaning liquid, the displacement liquid, the processing liquid, and the inert gas flowing in the pipes 134, 144, 154, 164, and 178b toward the nozzles 132, 142, 152, 162, and 178a, respectively, are stopped.

[0131] Further, the control section 22 performs control to move the arms 138a, 148a, 158a, and 168a in the horizontal direction and / or the vertical direction by the moving mechanisms 138c, 148c, 158c, and 168c. By this, the control section 22 moves the nozzles 132, 142, 152, and 162 installed at the front end of the arms 138a, 148a, 158a, and 168a on the upper surface Wa of the substrate W. Further, the control section 22 can move the nozzles 132, 142, 152, and 162 installed at the front end of the arms 138a, 148a, 158a, and 168a between the ejection position and the retreat position.

[0132] Further, the control section 22 controls the shielding member 170 to move the shielding plate 172 between the upper position and the lower position. For example, the control section 22 drives the lifting unit 176 to move the shielding plate 172 between the upper position and the lower position.

[0133] The control section 22 controls the FFU 180 to deliver air from the air supply port above the chamber 110 into the chamber. Further, the control section 22 controls the cover 190 to raise the cover 190 to the side of the substrate W.

[0134] The substrate processing apparatus 100 of the present embodiment is suitable for use in forming a semiconductor device. For example, the substrate processing apparatus 100 is suitable for use in manufacturing a memory of a NAND configuration. The substrate processing apparatus 100 can also be used in manufacturing a 3D NAND configuration.

[0135] In the substrate processing apparatus 100 of the present embodiment, the storage section 24 stores the learned model LM and the control program PG. The substrate processing apparatus 100 operates in accordance with the process determined by the control program PG.

[0136] Further, the control section 22 includes an input information acquisition section 22a and a sublimation drying processing condition information acquisition section 22b. The input information acquisition section 22a acquires at least one of substrate information and processing liquid information as input information.

[0137] The input information acquisition section 22a includes a substrate information acquisition section 22al and a processing liquid information acquisition section 22a2. The substrate information acquisition section 22al acquires substrate information of a processing target substrate Wp. The substrate information of the processing target substrate Wp includes substrate attribute information or pattern formation condition information. The substrate attribute information shows attributes of a trench and a configuration body provided to the processing target substrate Wp. The pattern formation condition information shows a formation condition of a pattern that forms the trench and the configuration body provided to the processing target substrate Wp. Further, the substrate information of the processing target substrate Wp can include both the substrate attribute information and the pattern formation condition information.

[0138] For example, the substrate attribute information displays an attribute of at least one of the trench and the structure. In one example, the substrate attribute information displays any one of a surface area of the substrate, a depth of the trench, a width of the trench, an aspect ratio of the trench, a layout pattern of the trench and the structure, a degree of concentration of the trench, a density of the trench or the structure, an opening ratio of the trench, and a composition of a surface of the structure. Further, the surface area of the substrate is preferably determined in consideration of the depth and the opening ratio of the trench with respect to the structure. However, the surface area of the substrate can also not be directly determined. As the surface area of the substrate, a length of a periphery of the substrate (a circumference length when the substrate is circular, and lengths of four sides when the substrate is quadrangular) that can be easily calculated can also be used. For example, the density of the trench can also be displayed by a ratio of an area of the trench with respect to an area of the surface region of the substrate. Further, the density of the structure can also be displayed by a ratio of an area of a surface of the structure including a portion exposed via the trench with respect to an area of the surface region of the substrate.

[0139] The pattern formation condition information displays a formation condition of the pattern in which the trench and the structure are formed. For example, in a case where the trench of the processing target substrate Wp is formed by etching, the pattern formation condition information displays an etching processing condition.

[0140] Alternatively, in a case where an attribute of at least one of the trench and the structure of the processing target substrate Wp changes due to processing within the substrate processing apparatus 100, the pattern formation condition information can also display a condition of the processing. For example, in a case where an attribute of at least one of the trench and the structure of the processing target substrate Wp changes due to a chemical liquid, the pattern formation condition information displays a chemical liquid processing condition. In one example, the pattern formation condition information displays a kind, a concentration, a temperature, and a supply amount of the chemical liquid. Further, the pattern formation condition information can also display a processing condition other than the chemical liquid processing condition. Further, the substrate information acquisition unit 22al can also acquire, from the storage unit 24, other information other than the substrate attribute information and the pattern formation condition information as the substrate information.

[0141] The processing liquid information acquisition unit 22a2 acquires processing liquid information for displaying an attribute of a processing liquid supplied to the processing target substrate Wp. That is, the processing liquid information acquisition unit 22a2 acquires processing liquid information for displaying an attribute of a processing liquid supplied to the processing target substrate Wp from the processing liquid supply unit 160. For example, the processing liquid information includes information for displaying a kind of a solute, a kind of a solvent, a concentration, or a temperature of the processing liquid. Further, the temperature of the processing liquid is preferably a temperature of the processing liquid supplied onto the processing target substrate Wp. However, the temperature of the processing liquid can also be a temperature of the processing liquid within the piping 164.

[0142] The learned model LM generates sublimation drying process condition information based on information of at least one of the substrate information and the processing liquid information. Typically, when the learned model LM is inputted with the substrate information and the processing liquid information, sublimation drying process condition information corresponding to the substrate information and the processing liquid information is outputted. In one example, when the learned model LM is inputted with information to show the depth of the trench of the processing target substrate Wp and the concentration of the processing liquid used for the processing target substrate Wp, sublimation drying process condition information corresponding to the inputted information is outputted from the learned model LM.

[0143] The sublimation drying process condition information acquisition section 22b acquires sublimation drying process condition information from the learned model LM. The sublimation drying process condition information acquisition section 22b acquires sublimation drying process condition information corresponding to the inputted information of the processing target substrate Wp from the learned model LM.

[0144] For example, the sublimation drying process condition information shows any one of the supply amount of the processing liquid, the ejection pattern of the processing liquid, and the rotation speed of the learning target substrate WL in the sublimation drying process. Further, the sublimation drying process condition information can also show any one of the temperature inside the chamber 110 (or the substrate W (especially the upper surface Wa)), the atmosphere concentration inside the chamber 110, the position of the shielding member 170, and the flow rate of the inert gas supplied from the inert gas supply section 178. In addition, the degree of evaporation of the solvent greatly varies depending on the temperature inside the chamber 110 and the atmosphere concentration of the solvent. In particular, the temporal variation of the temperature inside the chamber 110 or the atmosphere concentration of the solvent at the time of evaporation of the solvent greatly affects the pattern collapse, and thus the temperature inside the chamber 110 and the atmosphere concentration of the solvent become useful information.

[0145] The control section 22 controls the substrate holding section 120 and the processing liquid supply section 160 in accordance with the sublimation drying process condition shown by the sublimation drying process condition information. In addition, the control section 22 can also control the substrate holding section 120, the processing liquid supply section 160, and the shielding member 170 in accordance with the sublimation drying process condition shown by the sublimation drying process condition information.

[0146] It is preferable that the substrate processing system 10 further includes a display section 42, an input section 44, and a communication section 46.

[0147] The display section 42 displays an image. The display section 42 is, for example, a liquid crystal display or an organic electroluminescence display.

[0148] The input section 44 is an input machine to input various information to the control section 22. For example, the input section 44 is a keyboard, a pointing device, or a touch panel.

[0149] The communication section 46 is connected to a network and communicates with external devices. In the present embodiment, the network includes, for example, the Internet, a LAN (Local Area Network), a public telephone network, and a short-range wireless network. The communication section 46 is a communication machine, such as a network interface controller.

[0150] Further, it is preferable that the substrate processing system 10 further include the sensor 50. Typically, the plurality of sensors 50 detect the state of each portion of the substrate processing system 10. For example, at least a portion of the sensors 50 detects the state of each portion of the substrate processing apparatus 100.

[0151] The storage section 24 stores the output results from the sensors 50 and the control parameters of the control program as time series data TD. Typically, the time series data TD is stored separately for each substrate W.

[0152] The sensor 50 detects a physical quantity of an object used in the substrate processing apparatus 100 during the period from the start of the processing of each substrate W until the end of the processing, and outputs a detection signal that shows the physical quantity to the control section 22. Also, the control section 22 associates the physical quantity shown by the detection signal output from the sensor 50 during the period from the start of the processing of each substrate W until the end of the processing with time, and stores the time series data TD in the storage section 24 for each substrate W.

[0153] The control section 22 acquires the time series data TD from the sensor 50, and causes the storage section 24 to store the time series data TD. In this case, the control section 22 associates the time series data TD with lot identification information, substrate identification information, processing order information, and lot interval information, and stores them in the storage section 24. The lot identification information is information that identifies a lot (for example, a lot number). A lot shows the processing unit of the substrates W. One lot is composed of a predetermined number of substrates W. The substrate identification information is information that identifies the substrates W. The processing order information is information that shows the order of the processing for the predetermined number of substrates W that constitute one lot. The lot interval information is information that shows the time interval from the end of the processing for a lot until the start of the processing for the next lot. The substrate information and the processing liquid information can also be acquired from the time series data TD.

[0154] Next, referring to Figures 1-5 The substrate processing method of the substrate processing apparatus 100 of the present embodiment will be described. Figure 5 (a) in FIG. 1 is a flowchart of the substrate processing method in the substrate processing apparatus 100 of the present embodiment, Figure 5 (b) in FIG. 1 is a flowchart of the sublimation drying processing in the substrate processing method of the present embodiment.

[0155] As shown in (a) of FIG. 6, in step SA, the processing target substrate Wp is carried into the substrate processing apparatus 100. The carried-in processing target substrate Wp is loaded on the substrate holding portion 120. Typically, the processing target substrate Wp is carried into the substrate processing apparatus 100 by the central robot CR. Figure 5 Next, in step SRa, the substrate holding portion 120 starts rotating in a state where the processing target substrate Wp is loaded.

[0156]

[0157] In step S10, the processing target substrate Wp is processed with the chemical liquid. The chemical liquid supply portion 130 supplies the chemical liquid to the processing target substrate Wp. The chemical liquid is sprayed from the nozzle 132 of the chemical liquid supply portion 130 toward the upper surface Wa of the processing target substrate Wp. The chemical liquid covers the upper surface Wa of the processing target substrate Wp. Thus, the processing target substrate Wp is processed with the chemical liquid.

[0158] In step S20, the processing target substrate Wp is cleaned with the cleaning liquid. The cleaning liquid supply portion 140 supplies the cleaning liquid to the processing target substrate Wp. The cleaning liquid is sprayed from the nozzle 142 of the cleaning liquid supply portion 140 toward the upper surface Wa of the processing target substrate Wp. The cleaning liquid covers the upper surface Wa of the processing target substrate Wp. Thus, the processing target substrate Wp is processed with the cleaning liquid.

[0159] In step S30, the processing target substrate Wp is replaced with the replacement liquid. The replacement liquid supply portion 150 supplies the replacement liquid to the processing target substrate Wp. The replacement liquid is sprayed from the nozzle 152 of the replacement liquid supply portion 150 toward the upper surface Wa of the processing target substrate Wp. The replacement liquid covers the upper surface Wa of the processing target substrate Wp. Thus, the upper surface Wa of the processing target substrate Wp is replaced with the replacement liquid.

[0160] After that, in step S40, the processing target substrate Wp is subjected to sublimation drying processing. The sublimation drying processing includes supplying a processing liquid (step S41), forming a frozen body (step S42), and sublimation (step S43).

[0161] In step S41, the processing liquid is supplied to the processing target substrate Wp. The processing liquid supply portion 160 supplies the processing liquid to the processing target substrate Wp. The processing liquid is sprayed from the nozzle 162 of the processing liquid supply portion 160 toward the upper surface Wa of the processing target substrate Wp. The processing liquid covers the upper surface Wa of the processing target substrate Wp. Thus, the upper surface Wa of the processing target substrate Wp is replaced with the processing liquid.

[0162] Specifically, after the processing liquid is started to be supplied to the processing target substrate Wp, the rotation speed of the substrate holding portion 120 is maintained at a predetermined speed. Thus, the processing liquid of a predetermined thickness is formed on the upper surface Wa of the processing target substrate Wp.​

[0163] In step S42, a solidified substance is formed from the processing liquid. While the substrate holding unit 120 continues to rotate the substrate W while holding the processing liquid on the upper surface Wa of the substrate W, the solvent evaporates from the processing liquid. As the solvent evaporates, a solidified substance is formed, which is a sublimable substance that has solidified as a solute. Therefore, the grooves on the upper surface Wa of the substrate W are filled with the solidified substance. The thickness of the solidified substance corresponds to the thickness of the processing liquid formed on the upper surface Wa of the substrate Wp to be processed before the solidified substance is formed. Furthermore, during the formation of the solidified substance, the inert gas supply unit 178 of the shielding member 170 can also supply inert gas to the substrate W.

[0164] In step S43, the solidified material sublimates. As the substrate holding portion 120 continues to rotate the substrate W while holding the solidified material on the upper surface Wa of the substrate W, the solidified material continues to sublimate, and finally disappears from the grooves on the upper surface Wa of the substrate W. Furthermore, during the sublimation of the solidified material, the inert gas supply portion 178 of the shielding member 170 can also supply inert gas to the substrate W. Additionally, the substrate W can be heated during the sublimation of the solidified material.

[0165] Subsequently, in step SRb, the substrate holding part 120 stops rotating.

[0166] In step SB, the substrate Wp to be processed is detached from the substrate holding section 120 and removed. Typically, the substrate Wp to be processed is removed from the substrate processing apparatus 100 by a central robot CR. In this manner, the substrate Wp to be processed can be treated with a chemical solution, and the substrate Wp to be processed can be dried while suppressing structural damage.

[0167] Next, refer to Figures 1-5 (b) in this embodiment describes the sublimation drying process of the substrate Wp to be processed by the substrate processing apparatus 100.

[0168] In step S4a, substrate information of the target processing substrate Wp is obtained. The substrate information acquisition unit 22a1 acquires the substrate information of the target processing substrate Wp. The substrate information includes at least one of substrate attribute information and pattern forming condition information.

[0169] For example, the control unit 22 obtains substrate attribute information from the storage unit 24. The substrate attribute information displays attributes of at least one of the trenches and structures. For example, the substrate attribute information displays the depth of the trenches. Alternatively, the substrate attribute information may also display the width of the trenches or the aspect ratio (depth / width) of the trenches. Alternatively, the substrate attribute information may further display the configuration pattern of the trenches and structures, the density of the trenches, the density of the trenches or structures, and the opening ratio of the trenches.

[0170] Further, the substrate attribute information can also be information indicating a composition of the structure or a hydrophilicity of the structure with respect to water. For example, the information indicating the composition of the structure indicates that the composition of the structure is silicon oxide or silicon nitride. Further, the information indicating the hydrophilicity of the structure with respect to water indicates that the structure is hydrophilic or hydrophobic.

[0171] The substrate attribute information can also be measured within the substrate processing apparatus 100 or the substrate processing system 10. Alternatively, the substrate attribute information can also be measured outside the substrate processing system 10 or the substrate processing apparatus 100.

[0172] Alternatively, the control section 22 acquires the pattern formation condition information from the storage section 24. The pattern formation condition information indicates a formation condition of the structure and the trench. In a case where the trench of the substrate W is formed by etching, the control section 22 acquires etching condition information from the storage section 24. Alternatively, the pattern formation condition information indicates a condition of a process in which the attribute of the structure and the trench has been changed. For example, in a case where the depth or the width of the structure and the trench is changed by a chemical liquid, the pattern formation condition information indicates a concentration, a temperature, and a supply amount of the chemical liquid.

[0173] In step S4b, processing liquid information of the processing target substrate Wp is acquired. The processing liquid information acquisition section 22a2 acquires the processing liquid information of the processing target substrate Wp. The processing liquid information indicates any one of a kind of a sublimation substance and a solvent contained in a processing liquid used when the processing target substrate Wp is subjected to the sublimation drying process, a concentration, and a temperature of the processing liquid.

[0174] In step S4c, the substrate information and the processing liquid information of the processing target substrate Wp are input to the learned model LM. Although this will be described in detail later, the learned model LM is constructed from learning data including substrate information of a learning target substrate WL, processing liquid information indicating an attribute of a processing liquid used for the learning target substrate WL, sublimation drying process condition information indicating a processing condition of a sublimation drying process performed on the learning target substrate WL, and processing result information indicating a result of the sublimation drying process performed on the learning target substrate WL. The learned model LM outputs the sublimation drying process condition information Rp in correspondence with the substrate information and the processing liquid information of the processing target substrate Wp.

[0175] In step S4d, the sublimation drying process condition information is acquired from the learned model LM. The sublimation drying process condition information acquisition section 22b acquires the sublimation drying process condition information corresponding to the substrate information and the processing liquid information from the learned model LM.

[0176] In step S4e, the substrate holding portion 120 and the chemical liquid supply portion 130 perform sublimation drying treatment on the treatment target substrate Wp in compliance with the sublimation drying treatment condition information. In Figure 3 In the substrate processing apparatus 100 shown in FIG. 1, the treatment liquid supply portion 160 supplies a treatment liquid to the treatment target substrate Wp in compliance with the sublimation drying treatment condition information (step S41 in (a)). Thereafter, the treatment liquid forms a solidified body (step S42 in (a)), and the solidified body is sublimated (step S43 in (a)). Further, the shielding member 170 can also be controlled when performing sublimation drying treatment. In this way, the treatment target substrate Wp can be subjected to sublimation drying treatment. Figure 5 Figure 5 Figure 5

[0177] According to the present embodiment, the sublimation drying treatment condition information corresponding to the substrate information and the treatment liquid information of the treatment target substrate Wp is acquired from the learned model LM constructed through machine learning, and the sublimation drying treatment is performed in compliance with the sublimation drying treatment conditions indicated by the sublimation drying treatment condition information. According to the present embodiment, the sublimation drying treatment can be appropriately performed in accordance with the grooves and the pattern of the structure of the treatment target substrate Wp.

[0178] Further, in the above description with reference to Figure 5 (b), although the substrate information is acquired in step S4a and the treatment liquid information is acquired in step S4b, and the substrate information and the treatment liquid information are input as input information to the learned model LM, the present embodiment is not limited thereto. Either one of the substrate information and the treatment liquid information can be input as input information to the learned model LM. Even in this case, the sublimation drying treatment condition information corresponding to the input information can be acquired from the learned model LM, and the sublimation drying treatment can be performed in compliance with the sublimation drying treatment conditions.

[0179] Next, the substrate processing method of the present embodiment will be described with reference to Figure 6 and Figure 7 Figure 6 (a) to Figure 6 (e) and Figure 7 (a) to Figure 7 (c) are schematic diagrams showing the processing of the treatment target substrate Wp in the substrate processing method of the present embodiment.

[0180] As shown in (a) of Figure 6 , the treatment target substrate Wp is loaded in the substrate holding portion 120. Figure 6 (a) of Figure 5 corresponds to step SA of (a) of

[0181] As shown in (a) of​​​​Figure 6 (b) of FIG. 6, the processing target substrate Wp starts to be rotated. The substrate holding portion 120 starts to be rotated together with the processing target substrate Wp loaded. Figure 6 (b) of FIG. 6 corresponds to the step SRa of (a) of FIG. 6. Figure 5 (b) of FIG. 6 corresponds to the step SRa of (a) of FIG. 6.

[0182] As shown in (c) of FIG. 6, the processing target substrate Wp is supplied with a chemical liquid. The chemical liquid supplying portion 130 supplies the processing target substrate Wp with the chemical liquid. Figure 6 (c) of FIG. 6 corresponds to the step S10 of (a) of FIG. 6. Figure 6 (c) of FIG. 6 corresponds to the step S10 of (a) of FIG. 6. Figure 5 (c) of FIG. 6 corresponds to the step S10 of (a) of FIG. 6.

[0183] Further, in (c) of FIG. 6, a trench Wg and a pattern of a structure Ws in a certain region of the processing target substrate Wp are shown in an enlarged manner. The trench Wg and the structure Ws are formed in the processing target substrate Wp. Here, the trenches Wg are formed at intervals, thereby forming the structure Ws. Typically, the trenches of the processing target substrate Wp are formed in a dry etching apparatus different from the substrate processing apparatus 100. Figure 6 The depth of the trench Wg is Wd, and the width of the trench Wg is Ww. For example, the depth Wd of the trench Wg is 50 nm or more and 1000 nm or less, or can be 10 nm or more and 800 nm or less. The width Ww of the trench Wg is 5 nm or more and 100 nm or less, or can be 10 nm or more and 80 nm or less.

[0184] The chemical liquid is supplied to the upper surface Wa of the processing target substrate Wp. Therefore, from a microscopic point of view, the surface of the processing target substrate Wp and the trench Wg are immersed in the chemical liquid. Further, the trench Wg and the pattern of the structure Ws can be formed in the entirety of the upper surface Wa of the processing target substrate Wp, or can be formed in a region of a part of the upper surface Wa of the processing target substrate Wp.

[0185] As shown in (d) of FIG. 6, the processing target substrate Wp is supplied with a cleaning liquid. The cleaning liquid supplying portion 140 supplies the processing target substrate Wp with the cleaning liquid.

[0186] (d) of FIG. 6 corresponds to the step S20 of (a) of FIG. 6. Figure 6 (d) of FIG. 6 corresponds to the step S20 of (a) of FIG. 6. Figure 6 Figure 5 As shown in (e) of FIG. 6, the processing target substrate Wp is supplied with a replacement liquid. The replacement liquid supplying portion 150 supplies the processing target substrate Wp with the replacement liquid. (e) of FIG. 6 corresponds to the step S30 of (a) of FIG. 6.

[0187] (e) of FIG. 6 corresponds to the step S30 of (a) of FIG. 6. Figure 6 Figure 6 As shown in (f) of FIG. 6, the processing target substrate Wp is supplied with a chemical liquid. The chemical liquid supplying portion 130 supplies the processing target substrate Wp with the chemical liquid. Figure 5 (f) of FIG. 6 corresponds to the step S40 of (a) of FIG. 6.

[0188] As shown in (g) of FIG. 6, the processing target substrate Wp is supplied with a cleaning liquid. The cleaning liquid supplying portion 140 supplies the processing target substrate Wp with the cleaning liquid. Figure 7 ​The processing liquid supply section 160 supplies the processing liquid to the processing target substrate Wp. Figure 7 The step S41 of (a) in FIG. 10 corresponds to Figure 5 The step S41 of (a) in FIG. 10 corresponds to Figure 7 As shown in (a) in FIG. 10, the surface of the processing target substrate Wp and the trench are immersed in the processing liquid.

[0189] Further, in a case where the pattern of the trench Wg and the structure Ws is locally formed on the processing target substrate Wp, the processing liquid supply section 160 can also scan the nozzle 162 in such a manner that the processing liquid is first ejected toward a region of the processing target substrate Wp where the pattern of the trench Wg and the structure Ws is locally formed, and then the target of the ejection of the processing liquid is directed toward the center of the processing target substrate Wp. The ejection pattern of the processing liquid is controlled in this manner, whereby the processing target substrate Wp can be efficiently processed with a smaller amount of the processing liquid.

[0190] As shown in (b) in FIG. 10, the solvent evaporates from the processing liquid in the processing target substrate Wp, thereby forming the solidified body Co. Figure 7 As shown in (b) in FIG. 10, the solvent evaporates from the processing liquid in the processing target substrate Wp, thereby forming the solidified body Co. Figure 7 The step S42 of (b) in FIG. 10 corresponds to the step S43 of (a) in FIG. 10. Here, the inert gas is supplied to the processing target substrate Wp from the inert gas supply section 178 of the shielding member 170. Thereby, the evaporation of the solvent from the processing liquid is promoted to maintain the gas-liquid balance, whereby the solidified body Co is formed in a short period. Figure 5 The step S42 of (b) in FIG. 10 corresponds to the step S43 of (a) in FIG. 10. Here, the inert gas is supplied to the processing target substrate Wp from the inert gas supply section 178 of the shielding member 170. Thereby, the evaporation of the solvent from the processing liquid is promoted to maintain the gas-liquid balance, whereby the solidified body Co is formed in a short period.

[0191] In (b) in FIG. 10, the thickness of the solidified body Co is shown as the thickness Ct. Here, the thickness Ct of the solidified body Co is smaller than the depth Wd of the trench Wg. However, the thickness Ct of the solidified body Co can also be larger than the depth of the trench Wg. In this case, the solidified bodies Co are not only connected to each other within the trench Wg but also connected to each other above the trench Wg. Figure 7 As shown in (c) in FIG. 10, the solidified body Co sublimates in the processing target substrate Wp.

[0192] The step S43 of (c) in FIG. 10 corresponds to the step S43 of (a) in FIG. 10. Here, the inert gas is supplied to the processing target substrate Wp from the inert gas supply section 178 of the shielding member 170. Thereby, the sublimation of the solidified body Co is promoted to maintain the gas-liquid balance, whereby the solidified body Co sublimates in a short period. Figure 7 The step S43 of (c) in FIG. 10 corresponds to the step S43 of (a) in FIG. 10. Here, the inert gas is supplied to the processing target substrate Wp from the inert gas supply section 178 of the shielding member 170. Thereby, the sublimation of the solidified body Co is promoted to maintain the gas-liquid balance, whereby the solidified body Co sublimates in a short period. Figure 7 Figure 5 As shown in (c) in FIG. 10, the solidified body Co sublimates in the processing target substrate Wp.

[0193] As shown in (c) in FIG. 10, the solidified body Co sublimates in the processing target substrate Wp. Figure 6 Figure 7 ​​As explained, according to the substrate processing method of the present embodiment, the liquid processing, the cleaning processing, the replacement processing, and the sublimation drying processing can be performed on the processing target substrate Wp. Thereby, the processing target substrate Wp can be processed with the liquid, and the collapse of the structure Ws with the drying of the liquid can be suppressed.

[0194] Further, as explained in (b) of Figure 7 When the solvent is evaporated from the processing liquid, the solidified body Co that fills the groove Wg of the processing target substrate Wp is formed. The collapse easiness of the structure Ws greatly varies depending on the ratio of the thickness Ct of the solidified body Co to the depth Wd of the groove Wg. For example, the smaller the ratio of the thickness Ct of the solidified body Co to the depth Wd of the groove Wg, the more easily the structure Ws collapses. On the other hand, the larger the ratio of the thickness Ct of the solidified body Co to the depth Wd of the groove Wg, the more difficultly the structure Ws collapses. Therefore, the thickness Ct of the solidified body Co becomes an important index of whether the structure Ws collapses or not.

[0195] In the above explanation, the Figure 6 corresponds to the process from the carrying-in of the processing target substrate Wp to the liquid processing, the cleaning processing, and the replacement processing, Figure 7 corresponds to the sublimation drying processing of the processing target substrate Wp. The sublimation drying processing condition of the processing target substrate Wp can also be determined depending on the processing target substrate Wp that is carried in. For example, in a case where a substrate on which a groove and a pattern of a structure are formed is carried in as the processing target substrate Wp, the sublimation drying processing condition can also be determined at a stage after the processing target substrate Wp is carried in.

[0196] Alternatively, the sublimation drying processing condition can also be determined depending on the processing procedure in the substrate processing apparatus 100 until the sublimation drying processing. For example, in a case where the attribute of the groove of the processing target substrate Wp changes due to the processing of the substrate processing apparatus 100, the sublimation drying processing condition can also be determined depending on the formation condition in the substrate processing apparatus 100 until the start of the sublimation drying processing.

[0197] As explained with reference to Figure 1 , the learned model LM is generated from the learning data LD, and the learning data LD is generated from the time series data TDL of the substrate processing apparatus 100L.

[0198] Next, the generation of the learning data LD is explained with reference to Figures 1-8 . Figure 8 is a block diagram of a substrate processing system 10L that is provided with a substrate processing apparatus 100L and a learning data generation apparatus 300. Here, the learning data generation apparatus 300 is communicably connected to the substrate processing apparatus 100L. Figure 8The substrate processing system 10L having the substrate processing apparatus 100L is the same as the block diagram of the substrate processing system 10 shown in FIG. 1 except that the control section 22L does not have the input information acquisition section 22a and the sublimation drying processing condition information acquisition section 22b and the storage section 24L does not store the learned model LM but stores the test recipe TR. Therefore, the same description as that of the substrate processing system 10 shown in FIG. 1 is omitted to avoid redundancy. Figure 4

[0199] The substrate processing system 10L has a plurality of substrate processing apparatuses 100L, an indexing robot IRL, a central robot CRL, a control apparatus 20L, a display section 42L, an input section 44L, a communication section 46L, and a sensor 50L. The substrate processing apparatus 100L, the indexing robot IRL, the central robot CRL, the control apparatus 20L, the display section 42L, the input section 44L, and the communication section 46L have the same configuration as those of the substrate processing apparatus 100, the indexing robot IR, the central robot CR, the control apparatus 20, the display section 42, the input section 44, and the communication section 46 of the substrate processing system 10 shown in FIG. 1. Figure 4

[0200] Further, the substrate processing apparatus 100L has a substrate holding section 120L, a chemical liquid supply section 130L, a cleaning liquid supply section 140L, a displacement liquid supply section 150L, a processing liquid supply section 160L, a shielding member 170L, an FFU 180L, and a cover 190L. It is preferable that the chamber 110, the substrate holding section 120L, the chemical liquid supply section 130L, the cleaning liquid supply section 140L, the displacement liquid supply section 150L, the processing liquid supply section 160L, the shielding member 170L, the FFU 180L, and the cover 190L have the same configuration as those of the chamber 110, the substrate holding section 120, the chemical liquid supply section 130, the cleaning liquid supply section 140, the displacement liquid supply section 150, the processing liquid supply section 160, the shielding member 170, the FFU 180, and the cover 190 shown in FIG. 1. Figure 3 Figure 4

[0201] The control apparatus 20L has a control section 22L and a storage section 24L. The storage section 24L stores a control program PGL. The substrate processing apparatus 100L operates in accordance with the process determined by the control program PGL.

[0202] Further, the storage section 24L stores a plurality of test recipes TR. The plurality of test recipes TR include recipes different in sublimation drying processing conditions. Therefore, in a case where the control section 22L processes the learning target substrate WL in accordance with the test recipe TR, different sublimation drying processing is performed on different learning target substrates WL.

[0203] ​​​​The storage section 24L stores time-series data TDL of the learning target substrate WL. The time-series data TDL is data for showing a time change of a physical quantity in the substrate processing apparatus 100L. The time-series data TDL shows a plurality of physical quantities detected by the sensor 50L. The time-series data TDL can also include data for showing a manufacturing process before the learning target substrate WL is processed by the substrate processing apparatus 100L. Further, the time-series data TDL includes substrate information including substrate attribute information or pattern formation condition information of the learning target substrate WL, processing liquid information, sublimation drying processing condition information showing a sublimation drying processing condition performed on the learning target substrate WL, and processing result information showing a result of the sublimation drying processing performed on the learning target substrate WL.

[0204] The learning data generation apparatus 300 is communicably connected to the substrate processing apparatus 100L. The learning data generation apparatus 300 communicates at least a part of the time-series data TDL of the substrate processing apparatus 100L.

[0205] The learning data generation apparatus 300 includes a control apparatus 320, a display section 342, an input section 344, and a communication section 346. The learning data generation apparatus 300 can communicate with the communication section 46L of the plurality of substrate processing apparatuses 100L via the communication section 346. The display section 342, the input section 344, and the communication section 346 have the same configuration as the display section 42, the input section 44, and the communication section 46.

[0206] The control apparatus 320 includes a control section 322 and a storage section 324. The storage section 324 stores a control program PG3. The learning data generation apparatus 300 operates in accordance with a procedure determined by the control program PG3.

[0207] The control section 322 receives at least a part of the time-series data TDL from the substrate processing apparatus 100L and causes the received time-series data TDL to be stored in the storage section 324. The storage section 324 stores at least a part of the time-series data TDL of the learning target substrate WL. The time-series data TDL is transmitted from the substrate processing apparatus 100L to the learning data generation apparatus 300 via the communication section 46L and the communication section 346. The control section 322 causes at least a part of the transmitted time-series data TDL to be stored in the storage section 324. The time-series data TDL stored in the storage section 324 includes the substrate information, the processing liquid information, the sublimation drying processing condition information, and the processing result information of the time-series data TDL.

[0208] The control section 322 acquires the substrate information, the processing liquid information, the sublimation drying process condition information, and the processing result information of the learning target substrate WL from the time series data TDL of the storage section 324. Further, the control section 322 aggregates the substrate information, the processing liquid information, the sublimation drying process condition information, and the processing result information of the learning target substrate WL and generates learning data LD, and the storage section 324 stores the learning data LD.

[0209] Next, the generation method of the learning data of the present embodiment will be described with reference to Figure 8 and Figure 9 The generation method of the learning data of the present embodiment will be described with reference to Figure 9 is a flowchart of the generation method of the learning data of the present embodiment. The generation of the learning data is performed in the learning data generation apparatus 300.

[0210] As shown in Figure 9 , in step S111, the time series data TDL of the learning target substrate WL is acquired. Typically, the learning data generation apparatus 300 receives at least a part of the time series data TDL of the learning target substrate WL from the substrate processing apparatus 100L. The storage section 324 stores the received time series data TDL.

[0211] In step S112, the substrate information is extracted from the time series data TDL of the learning target substrate WL stored in the storage section 324. The substrate information includes information of at least one of the substrate attribute information and the pattern formation condition information. The control section 322 acquires the substrate information of the learning target substrate WL from the time series data TDL of the storage section 324.

[0212] For example, the substrate attribute information shows the depth of the trench. Alternatively, the substrate attribute information can show the width of the trench or the aspect ratio (depth / width) of the trench. Further, the substrate attribute information can further show the arrangement pattern of the trench and the structure, the density of the trench, the density of the trench or the structure, or the opening ratio of the trench. Further, the substrate attribute information can show the surface area of the trench taking into account the predetermined depth, width, and opening ratio of the trench having the learning target substrate WL.

[0213] Further, the substrate attribute information can be information showing the composition of the structure or the affinity of the structure to water. For example, the information showing the composition of the structure shows that the composition of the structure is silicon oxide or silicon nitride. Further, the information showing the affinity of the structure to water shows that the structure is hydrophilic or hydrophobic.

[0214] In step S113, processing liquid information is extracted from the time series data TDL of the learning target substrate WL stored in the storage section 324. The processing liquid information includes processing liquid attribute information for displaying the attributes of the processing liquid. The control section 322 acquires the processing liquid information from the time series data TDL of the storage section 324.

[0215] In step S114, sublimation drying processing condition information of the learning target substrate WL is extracted from the time series data TDL of the learning target substrate WL stored in the storage section 324. The control section 322 acquires the sublimation drying processing condition information of the learning target substrate WL from the time series data TDL of the storage section 324.

[0216] In step S115, processing result information of the learning target substrate WL is extracted from the time series data TDL of the learning target substrate WL stored in the storage section 324. The control section 322 acquires the processing result information of the learning target substrate WL from the time series data TDL of the storage section 324.

[0217] In step S116, the substrate information, the processing liquid information, the sublimation drying processing condition information, and the processing result information of the learning target substrate WL are associated and generated as learning data LD, and the storage section 324 stores the learning data LD for each of the plurality of learning target substrates WL.

[0218] In the present embodiment, the generated learning data includes the substrate information, the processing liquid information, the sublimation drying processing condition information, and the processing result information of each of the learning target substrates WL, which are associated with each other. Such learning data is suitable for use in learning processing.

[0219] Further, in the explanation of Figure 8 and Figure 9 , although the learning data LD of the learning target substrate WL is generated by associating the substrate information, the processing liquid information, the sublimation drying processing condition information, and the processing result information, the present embodiment is not limited thereto. The learning data LD can be generated by associating at least one of the substrate information and the processing liquid information, the sublimation drying processing condition information, and the processing result information.

[0220] For example, in a case where the learning data LD is generated by associating the substrate information, the sublimation drying processing condition information, and the processing result information, the plurality of learning target substrates WL are subjected to sublimation drying processing using processing liquids having the same attributes. Further, in a case where the learning data LD is generated by associating the processing liquid information, the sublimation drying processing condition information, and the processing result information, the plurality of learning target substrates WL have the same substrate information.

[0221] InFigure 8 In the present embodiment, although the learning data generation device 300 is communicably connected to one substrate processing device 100L, the present embodiment is not limited thereto. The learning data generation device 300 can be communicably connected to a plurality of substrate processing devices 100L.

[0222] Further, in the description of Figure 8 and Figure 9 , although the time series data TDL generated by the substrate processing device 100L is transmitted to the learning data generation device 300 via the communication section 46L and the communication section 346, the present embodiment is not limited thereto. The control device 320 of the learning data generation device 300 is incorporated into the control device 20L of the substrate processing system 10L including the substrate processing device 100L, and the learning data LD is generated from the time series data TDL within the substrate processing system 10L, without transferring the time series data TDL via a network.

[0223] Next, the generation of the learned model LM of the present embodiment will be described with reference to Figure 10 Figure 10 is a schematic view of the learning data generation device 300 and the learning device 400 of the present embodiment. The learning data generation device 300 and the learning device 400 are communicable with each other.

[0224] The learning device 400 is communicably connected to the learning data generation device 300. The learning device 400 receives the learning data LD from the learning data generation device 300. The learning device 400 performs machine learning based on the learning data LD and generates the learned model LM.

[0225] The learning device 400 includes a control device 420, a display section 442, an input section 444, and a communication section 446. The display section 442, the input section 444, and the communication section 446 have the same configuration as the display section 42, the input section 44, and the communication section 46 of the substrate processing system 10 shown in Figure 4

[0226] The control device 420 includes a control section 422 and a storage section 424. The storage section 424 stores a control program PG4. The learning device 400 operates in accordance with the procedures determined by the control program PG4.

[0227] ​​Storage unit 424 stores learning data LD. The learning data LD is transmitted from learning data generation device 300 to learning device 400 via communication unit 346 and communication unit 446. Control unit 422 stores the transmitted learning data LD in storage unit 424. In the learning data LD stored in storage unit 424, substrate information, processing liquid information, sublimation drying processing condition information, and processing result information of time series data TDL are associated with each other.

[0228] Storage unit 424 stores the learning program LPG. The learning program LPG is a program used to execute a machine learning algorithm. The machine learning algorithm finds certain rules from a plurality of learning data LDs and generates a learned model LM to represent the found rules. Control unit 422 executes the learning program LPG of storage unit 424, thereby performing machine learning on the learning data LDs, adjusting the parameters of the inference program, and generating the learned model LM.

[0229] Machine learning algorithms are not particularly limited as long as they are supervised learning methods; for example, they can be decision trees, nearest neighbor methods, simple Bayesian classifiers, support vector machines, or neural networks. Therefore, a learned model (LM) can include decision trees, nearest neighbor methods, simple Bayesian classifiers, support vector machines, or neural networks. Backpropagation can also be used in the machine learning process that generates the learned model (LM).

[0230] For example, a neural network consists of an input layer, an output layer, and one or more intermediate layers. Specifically, the neural network is a deep neural network (DNN), a recurrent neural network (RNN), or a convolutional neural network (CNN), and performs deep learning. For example, a deep neural network consists of an input layer, an output layer, and one or more intermediate layers.

[0231] The control unit 422 includes an acquisition unit 422a and a learning unit 422b. The acquisition unit 422a acquires learning data LD from the storage unit 424. The learning unit 422b executes the learning program LPG of the storage unit 424, thereby performing machine learning on the learning data LD and generating a learned model LM from the learning data LD.

[0232] Learning unit 422b performs machine learning on a plurality of learning data LDs based on the learning program LPG. As a result, certain rules are identified from the plurality of learning data LDs, and a learned model LM is generated. That is, the learned model LM is constructed by performing machine learning LDs on the learning data. Storage unit 424 stores the learned model LM.

[0233] After that, typically, the learned model LM is transferred to Figure 4 The substrate processing system 10 shown in FIG. 1, the storage section 24 stores the learned model LM. In this case, as described with reference to Figure 4 The storage section 24 of the control device 20 in the substrate processing system 10 stores the learned model LM, and the sublimation drying processing condition information acquisition section 22b acquires the sublimation drying processing condition information from the learned model LM of the storage section 24.

[0234] However, the present embodiment is not limited thereto. It can also be that the storage section 24 does not store the learned model LM, and the sublimation drying processing condition information acquisition section 22b acquires the sublimation drying processing condition information from outside of the substrate processing system 10. For example, it can also be that the sublimation drying processing condition information acquisition section 22b transmits information of at least one of the substrate information and the processing liquid information of the processing target substrate Wp to the learned model LM of the learning device 400 via the communication section 46 and the communication section 446, and receives the sublimation drying processing condition information output in the learned model LM from the learning device 400 via the communication section 446 and the communication section 46.

[0235] Next, the learning method in the learning device 400 of the present embodiment will be described with reference to Figures 1-11 Figure 11 is a flowchart of the learning method of the present embodiment. The learning of the learning data LD and the generation of the learned model LM are performed in the learning device 400.

[0236] As shown in FIG. 12, in step S122, the acquisition section 422a of the learning device 400 acquires a plurality of learning data LD from the storage section 424. In the learning data LD, the substrate information, the processing liquid information, the sublimation drying processing condition information, and the processing result information of the learning target substrate WL are associated with each other. Figure 11 Next, in step S124, the learning section 422b performs machine learning on the plurality of learning data LD based on the learning program LPG.

[0237] Next, in step S126, the learning section 422b determines whether or not the machine learning of the learning data LD is ended. Whether or not the machine learning is ended is determined in accordance with a condition prepared in advance. For example, when the machine learning is performed on a predetermined number or more of learning data LD, the machine learning is ended.

[0238] In the case where the machine learning is not ended (NO in step S126), the processing returns to step S122. In this case, the machine learning is repeated. On the other hand, in the case where the machine learning is ended (YES in step S126), the processing proceeds to step S128.

[0239] In the case where the machine learning is not ended (NO in step S126), the processing returns to step S122. In this case, the machine learning is repeated. On the other hand, in the case where the machine learning is ended (YES in step S126), the processing proceeds to step S128.​

[0240] In step S128, the learning unit 422b outputs the model (one or more functions) to which the latest plurality of parameters (coefficients), i.e., the plurality of learned parameters (coefficients), are applied, as a learned model LM. The storage unit 424 stores the learned model LM.

[0241] In this way, the learning method ends, and the learned model LM is generated. According to the present embodiment, the learned model LM can be generated by performing machine learning on the learning data LD.

[0242] Further, in the present embodiment, although the learning device 400 is communicably connected to one learning data generation device 300, the present embodiment is not limited thereto. The learning device 400 can be communicably connected to a plurality of learning data generation devices 300. Figure 10 Further, in the description with reference to

[0243] and Figure 10 , although the learning data LD generated by the learning data generation device 300 is transmitted to the learning device 400 via the communication unit 346 and the communication unit 446, the present embodiment is not limited thereto. It can be that the control device 420 of the learning device 400 is incorporated into the control device 320 of the learning data generation device 300, and the learned model LM is generated from the learning data LD within the learning data generation device 300, without transferring the learning data LD via the network. Figure 11 Further, in the description with reference to

[0244] , although the time series data TDL generated by the substrate processing device 100L is transmitted to the learning data generation device 300 via the communication unit 46L and the communication unit 346, and the learning data LD generated by the learning data generation device 300 is transmitted to the learning device 400 via the communication unit 346 and the communication unit 446, the present embodiment is not limited thereto. It can be that the control device 320 of the learning data generation device 300 and the control device 420 of the learning device 400 are incorporated into the control device 20L of the substrate processing system 100L, and the learned model LM is generated from the time series data TDL via the learning data LD within the substrate processing system 10L, without transferring the time series data TDL and the learning data LD via the network. Figures 8-11 Further, in the description with reference to

[0245] , although the time series data TDL generated by the substrate processing device 100L is transmitted to the learning data generation device 300 via the communication unit 46L and the communication unit 346, and the learning data LD generated by the learning data generation device 300 is transmitted to the learning device 400 via the communication unit 346 and the communication unit 446, the present embodiment is not limited thereto. It can be that the control device 320 of the learning data generation device 300 and the control device 420 of the learning device 400 are incorporated into the control device 20L of the substrate processing system 100L, and the learned model LM is generated from the time series data TDL via the learning data LD within the substrate processing system 10L, without transferring the time series data TDL and the learning data LD via the network. Figures 8-11In the above description, although the substrate information of the learning target substrate WL is acquired and the processing liquid information to show the attribute of the processing liquid used for the sublimation drying process of the learning target substrate WL is acquired, and the substrate information and the processing liquid information are included in the learning data, the present embodiment is not limited thereto. Only one of the substrate information and the processing liquid information can be included in the learning data. Even in this case, the learning data in which only one of the substrate information and the processing liquid information, the sublimation drying process condition information, and the processing result information are associated and stored can be learned, and thus the learning completed model LM can be generated.

[0246] Next, referring to Figure 12 An example of the learning data LD will be described. Figure 12 is a view showing an example of the learning data LD. In Figure 12 , the learning data LD includes the learning data LD1 to LD1000.

[0247] Figure 12 The learning data LD1 to LD1000 of the learning data LD show the substrate information, the processing liquid information, the sublimation drying process condition, and the processing result of the learning target substrate WL. For example, the substrate information includes at least one of the substrate attribute information and the pattern formation condition information of the learning target substrate WL. In an example, the substrate information shows the depth, the width, or the aspect ratio of the trench of the learning target substrate WL. Further, the depth, the width, or the aspect ratio of the trench of the learning target substrate WL can be acquired by measuring the trench of the learning target substrate WL.

[0248] The processing liquid information shows the attribute of the processing liquid used for the sublimation drying process of the learning target substrate WL. For example, the processing liquid information shows the concentration or the temperature of the processing liquid.

[0249] The sublimation drying process condition information shows the condition of the sublimation drying process performed on the learning target substrate WL. The sublimation drying process condition includes, for example, the supply amount of the processing liquid, the ejection pattern of the processing liquid, and the rotation speed of the learning target substrate WL in the sublimation drying process.

[0250] The processing result information shows the processing result of the sublimation drying process performed on the learning target substrate WL. The processing result can also be determined by whether or not the collapse of the structure has been found in the learning target substrate WL. For example, in a case where the learning target substrate WL after the sublimation drying process is found to be good without the collapse of the structure, O is shown. On the other hand, in a case where the collapse of the structure is found and thus the result of the learning target substrate is found to be bad, X is shown.

[0251] The learning data LD1 shows the substrate information of the learning substrate WL1, the processing liquid information, the sublimation drying processing conditions, and the processing result. In the learning data LD1, La1 shows the substrate information of the learning substrate WL1, Lb1 shows the processing liquid information of the processing liquid used for the learning substrate WL1, and Ls1 shows the sublimation drying processing conditions for the learning substrate WL1. In addition, in the learning data LD1, since the learning substrate WL1 subjected to the sublimation drying processing is not found to have a structure collapse and is good, the processing result is shown as O.

[0252] The learning data LD2 to LD1000 are generated in correspondence with the learning substrates WL2 to WL1000. The sublimation drying processing conditions for the learning substrates WL2 to WL1000 can be the same or different. Even if the sublimation drying processing conditions are the same, the processing result greatly varies depending on the substrate information and the processing liquid information of the learning substrate WL. Conversely, even if the substrate information and the processing liquid information of the learning substrate WL are the same, the processing result greatly varies depending on the sublimation drying processing conditions of the learning substrate WL.

[0253] In addition, in the learning data LD shown in FIG. 10, Figure 12 In the learning data LD shown in FIG. 10, although the number of data is 1000, the present embodiment is not limited thereto. The number of data can be less than 1000 or more than 1000. However, the more the number of data, the better.

[0254] In addition, although Figure 12 In the learning data LD shown in FIG. 10, although the number of data is 1000, the present embodiment is not limited thereto. The number of data can be less than 1000 or more than 1000. However, the more the number of data, the better.

[0255] In the learning data LD, the substrate information preferably includes a plurality of items. For example, the substrate information can include information showing the depth of the groove, the width of the groove, the aspect ratio of the groove, the composition of the structure (particularly, the composition of the structure surface), the pattern arrangement of the groove and the structure.

[0256] Next, with reference to Figure 13 An example of the learning data LD will be described. Figure 13 is a view showing an example of the learning data LD. Figure 13 The learning data LD of FIG. 11 is the same as the learning data LD described with reference to FIG. 10 except that the substrate information of the learning substrate WL shows the depth of the groove, the width of the groove, the aspect ratio of the groove, the composition of the structure, and the pattern arrangement of the groove and the structure. Therefore, the overlapping description is omitted in order to avoid redundancy. Figure 12

[0257] In​Figure 13 In this case, the learning data LD includes learning data LD1 to LD1000. In this case, the substrate information includes trench depth information, trench width information, trench aspect ratio information, structure body composition information, and pattern arrangement information. The trench depth information shows the depth of the trench of the learning substrate WL. The trench width information shows the width of the trench of the learning substrate WL. The trench aspect ratio information shows the aspect ratio (depth / width) of the trench of the learning substrate WL. Further, the depth, width, and aspect ratio of the trench can also be measured before the learning substrate WL is carried to the substrate processing apparatus 100L. Alternatively, the depth, width, and aspect ratio of the trench can also be measured before the substrate processing apparatus 100L performs the sublimation drying process on the learning substrate WL.

[0258] The structure body composition information shows the composition of the surface portion of the structure body of the learning substrate WL. The pattern arrangement information shows the arrangement pattern of the trench and the structure body in the learning substrate WL.

[0259] The learning data LD1 shows the substrate information, the processing liquid information, the sublimation drying process conditions, and the processing result of a certain learning substrate WL1. In this case, in the learning data LD1, Lad1 shows the depth of the trench of the learning substrate WL1, Law1 shows the width of the trench of the learning substrate WL1, and Laa1 shows the aspect ratio of the trench of the learning substrate WL1. Further, Lam1 shows the composition of the structure body of the learning substrate WL1, and Lap1 shows the pattern arrangement of the structure body and the trench of the learning substrate WL1.

[0260] The learning data LD2 to LD1000 are generated in correspondence with the learning substrates WL2 to WL1000. The processing result greatly varies depending on the substrate information of the learning substrate WL. Therefore, even if the processing liquid and the sublimation drying process are the same, the processing result greatly varies depending on the substrate information of the learning substrate WL.

[0261] The learning data LD preferably has items that greatly cause variation in the processing result of the sublimation drying process of the learning substrate WL. In particular, since the processing result greatly varies depending on the depth, width, and aspect ratio of the trench of the learning substrate WL, in the learning data LD, the substrate information preferably includes information showing the depth, width, and aspect ratio of the trench of the learning substrate WL.

[0262] Further, in Figure 13In the illustrated learning data LD, although the substrate information is any one of the substrate attribute information showing the attribute of at least one of the trench and the structure formed in the learning target substrate WL, the present embodiment is not limited thereto. The substrate information can also include pattern formation condition information showing the formation condition of the trench and the structure of the learning target substrate WL.

[0263] For example, in a case where the trench and the structure of the substrate W are formed by dry etching, the substrate information can also be information showing the dry etching condition. Alternatively, in a case where the attribute of the trench and the structure of the substrate W is changed by the processing in the substrate processing apparatus 100, the substrate information can also be information showing the condition of the processing performed in the substrate processing apparatus 100.

[0264] Next, referring to Figure 14 The substrate processing method of the present embodiment will be described. Figure 14 (a) to Figure 14 (e) in FIG. 10 show a schematic view of the processing of the processing target substrate Wp in the substrate processing method of the present embodiment. Figure 14 (a) to Figure 14 (e) in FIG. 10 show the same view as Figure 6 (a) to Figure 6 (e) in FIG. 10 except that the depth of the trench of the processing target substrate Wp is changed by the chemical liquid in the substrate processing apparatus 100. Repetitive description is omitted to avoid redundancy.

[0265] As shown in (a) in FIG. 11, the processing target substrate Wp is loaded in the substrate holding portion 120. Figure 14 (a) in FIG. 11 corresponds to the step SA of (a) in FIG. 10. Figure 14 Figure 5 (a) in FIG. 11 corresponds to the step SA of (a) in FIG. 10.

[0266] As shown in (b) in FIG. 11, the processing target substrate Wp starts to rotate. The substrate holding portion 120 starts to rotate together with the loaded processing target substrate Wp. Figure 14 (b) in FIG. 11 shows the pattern of the trench Wg and the structure Ws of a certain region of the processing target substrate Wp in an enlarged manner. Here, the trenches Wg are formed at equal intervals in the processing target substrate Wp carried into the substrate processing apparatus 100, thereby forming the structure Ws. Here, the depth of the trench Wg is Wd0, and the width of the trench Wg is Ww. Figure 14

[0267] As shown in (c) in FIG. 11, the processing target substrate Wp is supplied with a chemical liquid. The chemical liquid supply portion 130 supplies the processing target substrate Wp with the chemical liquid. By the supply of the chemical liquid, the trench is formed in the processing target substrate Wp. Figure 14 ​​​

[0268] As in (c) of FIG. 8, Figure 6 As in (c) of FIG. 8, Figure 14 (c) of FIG. 8 enlarges and displays a pattern of the trench Wg of a certain region of the processing target substrate Wp and the structure Ws. The depth of the trench Wg is changed from WdO to Wd by the chemical liquid processing.

[0269] As in (d) of FIG. 9, the cleaning liquid is supplied to the processing target substrate Wp. The cleaning liquid supply section 140 supplies the cleaning liquid to the processing target substrate Wp. Figure 14 (d) of FIG. 9 corresponds to the step S20 of (a) of FIG. 8. Figure 6 Figure 5 As in (e) of FIG. 10, the displacement liquid is supplied to the processing target substrate Wp. The displacement liquid supply section 150 supplies the displacement liquid to the processing target substrate Wp.

[0270] As in (e) of FIG. 10, the displacement liquid is supplied to the processing target substrate Wp. The displacement liquid supply section 150 supplies the displacement liquid to the processing target substrate Wp. Figure 14 (e) of FIG. 10 corresponds to the step S30 of (a) of FIG. 8. Figure 14 Figure 5 The sublimation drying processing thereafter is the same as the above description of the sublimation drying processing described with reference to

[0271] As described above, the trench Wg and the structure Ws of the processing target substrate Wp can also be changed by the processing of the substrate processing apparatus 100. In this case, the attribute of the pattern of the trench Wg and the structure Ws is changed according to the pattern formation condition in the substrate processing apparatus 100. Therefore, instead of the substrate attribute information, the sublimation drying processing condition information can also be acquired based on the pattern formation condition information. Figure 7 Next, an example of the learning data LD in the case where the substrate information is used to display the pattern formation condition will be described with reference to

[0272] is a view showing an example of the learning data LD. In the learning data LD of Figure 15 The substrate information of the learning target substrate WL in the learning data LD of FIG. 8 is the same as the above-described learning data LD described with reference to Figure 15 FIG. 8 except that the pattern formation condition that changes the attribute of the trench and the structure is displayed. Repetitive description is omitted in order to avoid redundancy. Figure 15 Figure 12 In the learning data LD of FIG. 8, the learning data LD includes the learning data LD1 to LD1000. Here, the substrate information displays the concentration, the temperature, and the supply amount of the chemical liquid that changes the attribute of the trench and the structure of the learning target substrate WL. The substrate information includes: chemical liquid concentration information that displays the concentration of the chemical liquid; chemical liquid temperature information that displays the temperature of the chemical liquid; and chemical liquid supply amount information that displays the supply amount of the chemical liquid.

[0273] In the learning data LD of FIG. 8, the learning data LD includes the learning data LD1 to LD1000. Here, the substrate information displays the concentration, the temperature, and the supply amount of the chemical liquid that changes the attribute of the trench and the structure of the learning target substrate WL. The substrate information includes: chemical liquid concentration information that displays the concentration of the chemical liquid; chemical liquid temperature information that displays the temperature of the chemical liquid; and chemical liquid supply amount information that displays the supply amount of the chemical liquid. Figure 15

[0274] ​​​​The learning data LD1 displays substrate information, processing solution information, sublimation drying processing conditions, and processing results for a specific learning substrate WL1. Within the learning data LD1, Lac1 displays the concentration of the solution used on the learning substrate WL1, Lat1 displays the temperature of the solution used on the learning substrate WL1, and Lav1 displays the supply amount of the solution used on the learning substrate WL1.

[0275] Learning data LD2 to LD1000 are generated corresponding to learning target substrates WL2 to WL1000. The processing results vary significantly depending on the formation conditions that cause changes in the properties of the trenches in the learning target substrate WL. Preferably, the learning data LD has items that significantly cause changes in the processing results of the sublimation drying process of the learning target substrate WL.

[0276] In addition, in reference Figure 14 In the above description, although the depth of the trench Wg of the substrate W may vary depending on the processing liquid, this embodiment is not limited to this. The shape of the trench Wg of the substrate W may also vary depending on the processing liquid.

[0277] Next, refer to Figure 16 The substrate processing method of this embodiment is explained. Figure 16 (a) to Figure 16 (c) shows a schematic diagram of the processing of the substrate Wp to be processed in the substrate processing method of this embodiment. Figure 16 (a) to Figure 16 (c) except that the depth of the trenches on the substrate Wp to be processed in the substrate processing apparatus 100 varies according to the chemical solution, and... Figure 14 (a) to Figure 14 The same applies to (c) in the text. To avoid being too verbose, repeated entries are omitted.

[0278] like Figure 16 As shown in (a), the substrate Wp to be processed is loaded in the substrate holding section 120. Figure 16 (a) and Figure 5 Step SA corresponds to (a) in the text.

[0279] like Figure 16 As shown in (b), the substrate Wp to be processed begins to rotate. The substrate holding unit 120 and the loaded substrate Wp to be processed begin to rotate together. Figure 16 (b) shows a magnified view of the pattern of the trench Wg and the structure Ws in a certain area of ​​the substrate Wp being processed.

[0280] Here, the structure Ws includes an insulating layer Wi and a conductive layer Wc. The insulating layer Wi is made of silicon oxide or silicon nitride. The conductive layer Wc is made of polysilicon. In the structure Ws, the conductive layer Wc is disposed separately within the insulating layer Wi.

[0281] like Figure 16 As shown in (c), a chemical solution is supplied to the substrate Wp to be processed. The chemical solution supply unit 130 supplies the chemical solution to the substrate Wp to be processed. Through the supply of the chemical solution, trenches are formed on the substrate Wp to be processed.

[0282] and Figure 14 Similarly, in (c), Figure 16 Image (c) shows a magnified view of the pattern of the trench Wg and the structure Ws in a certain area of ​​the substrate Wp being processed. The conductive layer Wc is etched away by the chemical treatment, but the insulating layer Wi is not etched.

[0283] Subsequent cleaning, replacement fluid treatment, and sublimation drying processes are consistent with reference. Figure 6 , Figure 7 as well as Figure 14 The above description is the same, so it is omitted. As mentioned above, the trenches Wg and structures Ws of the substrate Wp to be processed can also vary depending on the processing of the substrate processing apparatus 100. This substrate processing apparatus 100 is suitable for use in manufacturing substrates W with 3D NAND structures.

[0284] In reference Figure 16 In the description, the conductive layer Wn is selectively etched by chemical treatment, and the trenches Wg extend laterally. Thus, the properties of the patterns of the trenches Wg and the structures Ws vary depending on the pattern formation conditions in the substrate processing apparatus 100. Therefore, sublimation drying process condition information can be obtained based on pattern formation condition information, instead of substrate property information.

[0285] Although it has been referred to Figure 13 as well as Figure 15 The study data LD includes substrate information with multiple items; however, the liquid information in the study data LD may also include multiple items.

[0286] Next, refer to Figure 17 This illustrates an example of learning data (LD) containing multiple items in the processing fluid information. Figure 17 This is a graph showing an example of learning data LD. Figure 17 In addition to displaying the concentration and temperature of the processing solution used in the sublimation drying process of the WL substrate for the learning object, the learning data LD also displays information about the processing solution compared to the reference solution. Figure 12 The learning data LD described above is the same, and repeated descriptions are omitted to avoid being too lengthy.

[0287] In Figure 13 , the learning data LD includes learning data LD1 to LD1000. Here, the processing liquid information includes processing liquid concentration information and processing liquid temperature information. The processing liquid concentration information shows the concentration of the processing liquid, and the processing liquid temperature information shows the temperature of the processing liquid.

[0288] The learning data LD1 shows the substrate information, the processing liquid information, the sublimation drying processing condition, and the processing result of a certain learning target substrate WL1. Here, in the learning data LD1, Lbc1 shows the concentration of the processing liquid used for the sublimation drying processing of the learning target substrate WL1, and Lbt1 shows the temperature of the processing liquid used for the sublimation drying processing of the learning target substrate WL1.

[0289] The learning data LD2 to LD1000 are generated in correspondence with the learning target substrates WL2 to WL1000. Even if the substrate information and the sublimation drying processing are equal, the processing result greatly varies depending on the processing liquid information of the learning target substrate WL. In particular, since the processing result greatly varies depending on the processing liquid concentration, it is preferable that the processing liquid information in the learning data LD includes information showing the concentration of the processing liquid.

[0290] In the above description, although the learning data LD in which the substrate information or the processing liquid information has a plurality of items has been described with reference to Figure 15 , Figure 17 and Figure 18 , the sublimation drying processing condition information in the learning data LD can also include a plurality of items.

[0291] Next, an example of the learning data LD in which the sublimation drying processing condition information includes a plurality of items will be described with reference to Figure 18 . Figure 18 is a diagram showing an example of the learning data LD. Figure 12 The learning data LD of Figures 1-19 is the same as the above-described learning data LD described with reference to , except that, in addition to the sublimation drying processing condition information of the sublimation drying processing that has been performed on the learning target substrate WL, the processing liquid supply amount, the processing liquid ejection pattern, the rotation speed of the learning target substrate WL, and the flow rate of the inert gas are shown, in order to avoid redundancy, the repeated description is omitted.

[0292] Figure 19The learning data LD includes learning data LD1 to LD1000. Here, the sublimation drying process condition information includes processing liquid supply information, processing liquid ejection pattern information, rotation speed information, and inert gas flow rate information. The processing liquid supply information displays the amount of processing liquid supplied, and the processing liquid ejection pattern information displays the time change in the position of the nozzle of the processing liquid supply unit 160L when the processing liquid is ejected. Furthermore, the rotation speed information displays the rotation speed of the learning target substrate WL during the sublimation drying process, and the inert gas flow rate information displays the flow rate of the inert gas supplied during the sublimation drying process.

[0293] The learning data LD1 displays substrate information, processing solution information, sublimation drying processing conditions, and processing results for a specific learning substrate WL1. Specifically, in the learning data LD1, Lsv1 displays the supply volume of the processing solution used for the sublimation drying process of the learning substrate WL1, and Lsi1 displays the time-varying position of the nozzle of the processing solution supply unit 160L relative to the learning substrate WL1. Furthermore, Lsr1 displays the rotational speed of the learning substrate WL1 during the sublimation drying process, and Lsn1 displays the flow rate of the inert gas supplied to the learning substrate WL1 during the sublimation drying process.

[0294] Learning data LD2 to LD1000 are generated corresponding to learning target substrates WL2 to WL1000. The processing results vary significantly depending on the sublimation drying process of the learning target substrate WL. Even assuming that the substrate information and processing liquid information are equal, the processing results will vary significantly due to the sublimation drying process of the learning target substrate WL. Preferably, the learning data LD has items that significantly cause changes in the processing results of the sublimation drying process of the learning target substrate WL.

[0295] Next, refer to Figure 19 This describes the substrate processing in the substrate processing apparatus 100 of this embodiment. Figure 19 (a) shows a schematic diagram of the substrate Wp to be processed. Figure 19 (b) The processing liquid supply unit 160 in the display substrate processing apparatus 100, Figure 19 (c) shows the sublimation drying process conditions Rp output from the learned model LM.

[0296] like Figure 19 As shown in (a), a pattern of trenches Wg and structures Ws is formed on the substrate Wp to be processed. Here, the depth of trench Wg is Wd, the width of trench Wg is Ww, and the aspect ratio of trench Wg is Wd / Ww.

[0297] Furthermore, a substrate Wp with a pattern of trenches Wg and structures Ws can also be fed into the substrate processing apparatus 100. In this case, the properties of the trenches Wg and structures Ws can be measured before or after the substrate Wp is fed into the substrate processing apparatus 100. Moreover, the properties of the trenches and structures of the substrate Wp can be varied within the substrate processing apparatus 100.

[0298] like Figure 18 As shown in (b), the processing fluid supply unit 160 further includes a sensor 160s. The sensor 160s measures the properties of the processing fluid flowing within the piping 164. For example, the sensor 160s measures the concentration of the processing fluid. Furthermore, the concentration of the processing fluid supply unit 160 can be preset according to the processing fluid generation conditions.

[0299] In this case, the trench depth information, trench width information, and treatment liquid concentration information are input into the learned model LM, thereby obtaining the sublimation drying treatment condition information Rp.

[0300] Figure 19 (c) in the figure shows the sublimation drying process condition information Rp. The sublimation drying process condition information Rp includes the supply amount of processing liquid for the substrate Wp to be processed, the spray pattern of the processing liquid, the rotation speed of the substrate Wp to be processed during the sublimation drying process, and the inert gas flow rate.

[0301] In the sublimation drying process condition information Rp, Rsv displays the supply amount of the processing liquid used on the substrate Wp to be processed, and Rsi displays the ejection pattern of the processing liquid used on the substrate Wp to be processed. In addition, Rsr displays the rotational speed of the substrate Wp to be processed during the sublimation drying process, and Rsn displays the flow rate of the inert gas supplied to the substrate Wp during the sublimation drying process.

[0302] In this case, the control unit 22 controls the substrate holding unit 120, the processing liquid supply unit 160, and the shielding member 170 to perform sublimation drying treatment on the substrate Wp to be processed, according to the sublimation drying treatment conditions shown in the sublimation drying treatment condition information Rp. As a result, the substrate Wp to be processed can be dried properly while suppressing the collapse of the structure Ws.

[0303] In addition, in reference Figure 12 as well as Figure 13In the above-described explanation, although the sublimation drying processing conditions have four items of the supply amount of the processing liquid, the ejection pattern of the processing liquid, the rotation speed of the processing target substrate, and the inert gas flow rate, the present embodiment is not limited thereto. The sublimation drying processing conditions can have one or more of any of the four items. Alternatively, the sublimation drying processing conditions can have a combination of one or more of any of the four items and other items. Alternatively, the sublimation drying processing conditions can have one or more items different from the four items.

[0304] Further, there can be a case where the same substrate processing apparatus 100 or the same type of substrate processing apparatus 100 processes the processing target substrates Wp of other uses having different properties of the trenches (for example, the depth, the width, or the aspect ratio of the trenches). Typically, although the aspect ratio of the trenches necessary in the processing target substrates Wp of the logic use is different from the aspect ratio of the trenches necessary in the processing target substrates Wp of the memory use, there can be a case where the same substrate processing apparatus 100 or the same type of substrate processing apparatus 100 is used for the manufacture of the processing target substrates Wp of the logic use and the processing target substrates Wp of the memory use.

[0305] In this case, the substrate processing apparatus 100 can perform the sublimation drying processing on the processing target substrates Wp of the logic use with the sublimation drying processing conditions suitable for the trenches of the processing target substrates Wp of the logic use, and perform the sublimation drying processing on the processing target substrates Wp of the memory use with the sublimation drying processing conditions suitable for the trenches of the processing target substrates Wp of the memory use. Further, the substrate processing apparatus 100 can perform the sublimation drying processing with the sublimation drying processing conditions suitable for the trenches of the processing target substrates Wp even if the uses of the processing target substrates Wp are the same, depending on different processes.

[0306] Further, in the learning data LD shown in FIG. 6, the processing result is shown as O when the processing result is good, and shown as X when the processing result is not good, and the processing result of the learning data LD is binarized, but the present embodiment is not limited thereto. The processing result can be classified into 3 or more plural values. The processing result can be shown by the number of counted collapsed structures. Alternatively, the processing result can be shown by the collapse rate of the number of collapsed structures with respect to the number of structures that should be formed on the substrate. Figure 15 、 Figures 17-18 、 Figure 20 、 Figure 20 In the learning data LD shown in FIG. 6, the processing result is shown as O when the processing result is good, and shown as X when the processing result is not good, and the processing result of the learning data LD is binarized, but the present embodiment is not limited thereto. The processing result can be classified into 3 or more plural values. The processing result can be shown by the number of counted collapsed structures. Alternatively, the processing result can be shown by the collapse rate of the number of collapsed structures with respect to the number of structures that should be formed on the substrate.

[0307] For example, the processing result can be classified into any value between the minimum value and the maximum value. For example, the processing result can be numerically valued by considering the characteristics of the learning target substrate WL and further considering the usage amount (supply amount) of the processing liquid or the time required for the sublimation drying processing, and the like.

[0308] Next, referring to Figure 20 An example of the learning data LD that shows the processing conditions in proportion is described. Figure 12 is a view showing an example of the learning data LD. Figures 13-20 The learning data LD of Figure 21 described above is the same as the learning data LD described above, and the repeated description is omitted to avoid redundancy.

[0309] In Figure 21 , the learning data LD includes learning data LD1 to LD1000. Here, the processing result is represented by a collapse rate of the number of collapsed structures with respect to the number of structures to be formed on the learning target substrate WL.

[0310] The learning data LD1 shows the substrate information, the processing liquid information, the sublimation drying processing conditions, and the processing result of a certain learning target substrate WL1. Here, in the learning data LD1, the collapse rate of the structure is 32%. The collapse rate of the structure can be obtained, for example, by analyzing the learning target substrate WL1 that has undergone the sublimation drying processing.

[0311] The learning data LD2 to LD1000 are generated in correspondence with the learning target substrates WL2 to WL1000. By inputting information of at least one of the substrate information and the processing liquid information of the processing target substrate Wp to the learning completed model LM generated from the learning data LD in which the processing result has been numerically processed, the sublimation drying processing conditions suitable for the processing target substrate Wp can be obtained with higher accuracy.

[0312] Further, although the learning data LD has been described with reference to Figure 21 each of the substrate information, the processing liquid information, the sublimation drying processing conditions, and the processing result of the learning data LD, it goes without saying that the learning data LD preferably has an arbitrary number of items for each combination of the substrate information, the processing liquid information, the sublimation drying processing conditions, and the processing result.

[0313] Further, for the learning data LD, a certain item can also have a profile that changes with time. For example, the concentration and the temperature of the processing liquid can also be adjusted with time.

[0314] Next, referring to Figure 3 the substrate processing apparatus 100 of the present embodiment is described. Figure 21 is a schematic view of the substrate processing apparatus 100 of the present embodiment. Further, Figure 22 the substrate processing apparatus 100 of Figure 22The substrate processing apparatus 100 explained above is the same, and the repeated description is omitted to avoid redundancy.

[0315] In Figure 22 The substrate processing apparatus 100 of the present embodiment shown in FIG. 1 includes a processing liquid supply section 160. The processing liquid supply section 160 supplies a processing liquid to the substrate W. The processing liquid is, for example, a cyclohexanone oxime solution. The cyclohexanone oxime solution is generated by mixing cyclohexanone oxime and a solvent. The solvent is, for example, IPA. The processing liquid supply section 160 can change the concentration of the processing liquid and supply the processing liquid to the substrate W.

[0316] The processing liquid supply section 160 includes a nozzle 162, a pipe 164, a valve 166, a nozzle moving section 168, and further includes individual pipes 164a, 164b, valves 166a, 166b, flow rate adjustment valves 167a, 167b, circulation pipes 169s, 169t, tanks 169a, 169b, and pumps 169p, 169q.

[0317] The tank 169a stores the processing liquid. The circulation pipe 169s is connected to the tank 169a. A first end portion of the circulation pipe 169s is communicated to a portion of the tank 169a, and a second end portion of the circulation pipe 169s is communicated to another portion of the tank 169a. The pump 169p is disposed in the circulation pipe 169s. By driving the pump 169p, the processing liquid in the tank 169a is circulated through the circulation pipe 169s.

[0318] The individual pipe 164a connects the circulation pipe 169s and the pipe 164. The individual pipe 164a connects a portion of the circulation pipe 169s and one end of the pipe 164. By the pump 169p, the processing liquid in the tank 169a is delivered to the individual pipe 164a. The valve 166a and the flow rate adjustment valve 167a are disposed in the individual pipe 164a. The valve 166a opens and closes a flow path in the individual pipe 164a. The flow rate adjustment valve 167a adjusts the amount of the processing liquid passing through the individual pipe 164a.

[0319] The tank 169b stores the processing liquid. The circulation pipe 169t is connected to the tank 169b. A first end portion of the circulation pipe 169t is communicated to a portion of the tank 169b, and a second end portion of the circulation pipe 169t is communicated to another portion of the tank 169b. The pump 169q is disposed in the circulation pipe 169t. By driving the pump 169q, the processing liquid in the tank 169b is circulated through the circulation pipe 169t.

[0320] Individual piping 164b connects to circulating piping 169t and piping 164. Individual piping 164b connects a portion of circulating piping 169t to one end of piping 164. Processing fluid in storage tank 169b is delivered to individual piping 164b via pump 169q. Valve 166b and flow control valve 167b are located on individual piping 164b. Valve 166b opens and closes the flow path within individual piping 164b. Flow control valve 167b adjusts the amount of processing fluid passing through individual piping 164b.

[0321] The concentration of the treatment solution in storage tank 169a (the concentration of sublimable substances contained in the treatment solution) differs from the concentration of the treatment solution in storage tank 169b. Therefore, when valves 166a and 166b are opened, the treatment solutions with different concentrations mix with each other in piping 164, and the uniformly mixed treatment solution is sprayed out from nozzle 162. Furthermore, by changing the opening degree of at least one of flow regulating valves 167a and 167b, the concentration of the treatment solution sprayed from nozzle 162 can be changed.

[0322] The control device 20 sets the opening degrees of valves 166a, 166b, flow regulating valve 167a, and flow regulating valve 167b based on the concentration of the specified treatment fluid. Therefore, by adjusting the flow regulating valves 167a and 167b, the flow rate of the treatment fluid flowing in the individual pipes 164a and 164b can be changed.

[0323] Next, refer to Figure 21 This describes the learning data LD used in the learning method of this embodiment. Figure 22 This is a graph showing an example of learning data LD. Figure 12 The learning data LD is suitable for use in generating Figures 1-22 The learning completion module LM of the substrate processing apparatus 100 shown. Furthermore... Figure 23 In addition to displaying the value of at least one item in the learning data and showing the distribution of the time variation of physical property values ​​at points, the learning data LD also displays the reference data. Figure 23 The learning data LD used in the description is the same, and repeated descriptions are omitted to avoid being too lengthy.

[0324] like Figure 23 As shown, the learning data LD includes learning data LD1 to LD1000. The processing fluid information includes the processing fluid concentration distribution and the processing fluid temperature distribution. The processing fluid concentration distribution shows the time-varying concentration of the processing fluid used on the learning target substrate WL. The processing fluid temperature distribution shows the time-varying temperature of the processing fluid used on the learning target substrate WL.

[0325] In the learning data LD1, the Lbp1 shows the concentration distribution of the processing liquid supplied to the learning target substrate WL1, and the Lbq1 shows the temperature distribution of the processing liquid supplied to the learning target substrate WL1.

[0326] The same applies to the learning data LD2 to LD1000. For the learning target substrate WL, the result of the sublimation drying process of the learning target substrate WL greatly varies depending on the concentration change and the temperature change of the processing liquid. It is preferable that the learning data LD have items that greatly cause variation in the result of the sublimation drying process of the learning target substrate WL.

[0327] In addition, in the above description with reference to Figure 1 , the learning completed model LM is input with information of at least one of the substrate information and the processing liquid information as the input information, and outputs the sublimation drying process condition information from the learning completed model LM, but the present embodiment is not limited thereto. It is also possible to input the learning completed model LM with information of at least one of the substrate information and the processing liquid information and further input a part of the sublimation drying process condition information as the input information, and output other sublimation drying process condition information from the learning completed model LM.

[0328] Next, a substrate processing learning system 200 provided with the substrate processing apparatus 100 of the present embodiment will be described with reference to Figure 23 . Figure 5 is a schematic view of the substrate processing learning system 200. Figure 24 The substrate processing learning system 200 of Figure 24 has the same configuration as the substrate processing learning system 200 of , except that the input information includes information of at least one of the substrate information and the processing liquid information from the time series data TD of the substrate processing apparatus 100 and further includes a part of the sublimation drying process condition information, and the description of the overlapping parts will be omitted.

[0329] Figure 24 As shown in , the substrate processing apparatus 100L outputs the time series data TDL. The time series data TDL is data for showing the time change of a physical quantity in the substrate processing apparatus 100L.

[0330] The learning data generation apparatus 300 generates the learning data LD based on the time series data TDL or at least a part of the time series data TDL. The learning data generation apparatus 300 outputs the learning data LD.

[0331] The learning apparatus 400 performs machine learning on the learning data LD, thereby generating the learning completed model LM. The learning apparatus 400 outputs the learning completed model LM.

[0332] The substrate processing apparatus 100 outputs time-series data TD. The time-series data TD is data to show a time change of a physical quantity in the substrate processing apparatus 100.

[0333] Input information Cp for the processing target substrate Wp is generated from the time-series data TD. The input information Cp for the processing target substrate Wp includes information of at least one of substrate information of the processing target substrate Wp and processing liquid information and information to show a part of the sublimation drying processing conditions (part of sublimation drying processing condition information) that is a part of the sublimation drying processing conditions.

[0334] The substrate information of the processing target substrate Wp shows properties or formation conditions of a trench and a structure of the processing target substrate Wp. The processing liquid information shows properties of a processing liquid used for the sublimation drying processing for the processing target substrate Wp. The part of sublimation drying processing condition information shows conditions of a part of the sublimation drying processing that is a part of the sublimation drying processing. It is preferable that the part of the sublimation drying processing is at least a part of the first half of the entire sublimation drying processing. For example, as shown in (a) of FIG. 10, it is preferable that, in a case where the sublimation drying processing includes processing liquid supply, solid formation, and sublimation, the part of the sublimation drying processing contained in the input information Cp is a part of either one of the processing liquid supply and the solid formation. Figure 3

[0335] Based on the input information Cp for the processing target substrate Wp, sublimation drying processing condition information Rp is output from the learned model LM, the sublimation drying processing condition information Rp shows sublimation drying processing conditions suitable for the processing target substrate Wp in the substrate processing apparatus 100. Further, in this case, the sublimation drying processing condition information Rp shows processing conditions other than a part of the sublimation drying processing conditions in the sublimation drying processing conditions. After that, the sublimation drying processing is performed in accordance with the sublimation drying processing conditions shown by the sublimation drying processing condition information Rp. In this way, it is also possible that the input information input to the learned model LM includes information of at least one of substrate information and processing liquid information and part of sublimation drying processing condition information, and sublimation drying processing condition information to show conditions of the sublimation drying processing thereafter is output from the learned model LM.

[0336] Further, the temperature of the substrate W varies in the sublimation drying processing. For example, after the processing liquid is supplied to the substrate W, the temperature of the substrate W decreases in a process in which the solvent evaporates from the processing liquid. After that, the temperature of the substrate W increases in a process in which the sublimation substance solidifies. Therefore, by measuring the temperature of the substrate W, it is possible to more correctly grasp the state of the substrate W in the sublimation drying processing.

[0337] ​Therefore, it is preferable to measure the temperature of the substrate W in the sublimation drying process in the substrate processing apparatus 100. For example, the temperature of the substrate W can be measured in a contact manner or in a non-contact manner.

[0338] Next, referring to Figure 24 The substrate processing apparatus 100 of the present embodiment will be described. Figure 25 is a schematic view of the substrate processing apparatus 100. Figure 25 The substrate processing apparatus 100 of Figure 25 The substrate processing apparatus 100 of

[0339] As shown in Figure 24 The substrate processing apparatus 100 further includes a temperature measuring unit 128. The temperature measuring unit 128 measures the temperature of the substrate W. The temperature measuring unit 128 measures the temperature of the substrate W in a contact manner or in a non-contact manner.

[0340] For example, the temperature measuring unit 128 measures the temperature of the substrate W in the sublimation drying process. The temperature of the substrate W varies during the sublimation drying process. By measuring the temperature of the substrate W, the state in the substrate W in the sublimation drying process can be grasped more correctly.

[0341] Next, referring to Figure 25 The learning data LD used in the learning method of the present embodiment will be described. Figure 18 is a view showing an example of the learning data LD. Figure 25 The learning data LD of Figure 5 is suitable for use in generating the learned model LM of the substrate processing apparatus 100 shown in Figures 1-25 The learning data LD of Figure 26 The learning data LD of

[0342] As shown in Figure 26 The learning data LD includes learning data LD1 to LD1000. The sublimation drying process conditions show the process liquid supply amount, the process liquid ejection pattern, the substrate temperature, the rotation speed distribution of the learning target substrate WL, and the flow rate distribution of the inert gas.

[0343] The learning data LD1 shows the substrate information, the processing liquid information, the sublimation drying processing conditions, and the processing result of the learning target substrate WL1. In the learning data LD1, Lsv1 shows the supply amount of the processing liquid used for the sublimation drying processing of the learning target substrate WL1, Lsi1 shows the time change of the position of the nozzle of the processing liquid supply part 160L with respect to the learning target substrate WL1, Lst1 shows the time change of the temperature of the learning target substrate WL1 during the sublimation drying processing. Further, Lsr1 shows the distribution to show the time change of the rotation speed of the learning target substrate WL1 during the sublimation drying processing, and Lsn1 shows the distribution to show the time change of the flow rate of the inert gas supplied to the learning target substrate WL1 during the sublimation drying processing. The same applies to the learning data LD2 to LD1000.

[0344] Further, in the case where the learning completed model LM is generated using such learning data LD, in addition to the substrate information and the processing liquid information, information to show the processing liquid supply amount, the processing liquid ejection pattern, and the substrate temperature during the period until the solid is formed in the sublimation drying processing can be further included as the input information for the processing target substrate Wp. The processing liquid supply amount, the processing liquid ejection pattern, and the substrate temperature are also part of the sublimation drying processing conditions. In this case, the sublimation drying processing condition information corresponding to the input information can also be output from the learning completed model LM. At this time, the processing can also be performed following the sublimation drying processing conditions shown by the sublimation drying processing condition information in the remaining period in the sublimation drying processing. For example, Figure 4 The rotation of the processing target substrate Wp and the supply of the inert gas in the sublimation step shown in step S43 of (a) can also be performed following the sublimation drying processing conditions.

[0345] Further, in the above description of Figure 26 , although the storage part 24 of the substrate processing apparatus 100 or the storage part 424 of the learning apparatus 400 stores the learning completed model LM constructed by machine learning, the present embodiment is not limited thereto. The storage part 24 of the substrate processing apparatus 100 or the storage part 424 of the learning apparatus 400 can store the conversion table CT instead of the learning completed model LM.

[0346] Next, the substrate processing apparatus 100 of the present embodiment will be described with reference to Figure 27 The substrate processing apparatus 100 of the present embodiment will be described with reference to Figure 27 The substrate processing apparatus 100 of the present embodiment has the same configuration as the substrate processing apparatus 100 described with reference to ​ except that the storage part 24 stores the conversion table CT instead of the learning completed model LM, and the repeated description is omitted in order to avoid redundancy.

[0347] As described above, the substrate processing apparatus 100 of the present embodiment can output the sublimation drying processing condition information corresponding to the input information from the learning completed model LM, and can perform the processing following the sublimation drying processing conditions shown by the sublimation drying processing condition information. ​As shown, in the substrate processing apparatus 100, the storage section 24 stores the conversion table CT. The conversion table CT associates the substrate information, the processing liquid information, and the sublimation drying processing condition information of the processing target substrate Wp.

[0348] The substrate information of the processing target substrate Wp can be, for example, substrate attribute information of the processing target substrate Wp, or pattern formation condition information. The processing liquid information of the processing target substrate Wp includes processing liquid attribute information that shows the attribute of the processing liquid. Furthermore, the conversion table CT is created based on the substrate information, the processing liquid information, the sublimation drying processing condition information, and the processing result information of the learning target substrate WL.

[0349] The substrate information acquisition section 22al acquires the substrate information from the storage section 24. For example, the substrate information acquisition section 22al acquires the substrate attribute information or the pattern formation condition information from the storage section 24.

[0350] The processing liquid information acquisition section 22a2 acquires the processing liquid information from the storage section 24. For example, the processing liquid information acquisition section 22a2 acquires the processing liquid attribute information that shows the attribute of the processing liquid from the storage section 24.

[0351] The sublimation drying processing condition information acquisition section 22b acquires the sublimation drying processing condition information from the substrate information and the processing liquid information based on the conversion table CT. Typically, the sublimation drying processing condition information acquisition section 22b extracts the values corresponding to the substrate information and the processing liquid information from the conversion table CT, and acquires the sublimation drying processing condition information based on the relationship among the substrate information, the processing liquid information, and the sublimation drying processing condition information that are associated in the conversion table CT. In this way, the sublimation drying processing condition information acquisition section 22b acquires the sublimation drying processing condition information corresponding to the substrate information and the processing liquid information using the conversion table CT.

[0352] After that, the control section 22 controls the substrate holding section 120 and the processing liquid supply section 160 in accordance with the sublimation drying processing condition shown by the sublimation drying processing condition information. Furthermore, the control section 22 can also control the substrate holding section 120, the processing liquid supply section 160, and the shielding member 170 in accordance with the sublimation drying processing condition.

[0353] Furthermore, here, the conversion table CT can also be generated by associating at least one of the substrate information and the processing liquid information of the processing target substrate Wp with the sublimation drying processing condition information.

[0354] ​ is a diagram showing an example of the conversion table CT. As shown in FIG. 8, the conversion table CT is a table in which the substrate information, the processing liquid information, and the sublimation drying processing condition information are associated with each other. In the conversion table CT, the substrate information, the processing liquid information, and the sublimation drying processing condition information are associated with each other in a one-to-one manner. ​As shown, the conversion table CT shows the substrate information of the processing target substrate Wp, the processing liquid information, and the sublimation drying processing conditions. In the conversion table CT, the substrate information includes at least one of the substrate attribute information and the pattern formation condition information. Here, the substrate information of the processing target substrate Wp includes the trench depth information.

[0355] Further, in the conversion table CT, the processing liquid information includes the processing liquid attribute information. Here, the processing liquid information includes the processing liquid concentration information.

[0356] The conversion table CT1 shows the sublimation drying processing conditions corresponding to certain substrate information and processing liquid information. Here, in the conversion table CT1, Ad1 shows the depth of the trench of a certain processing target substrate Wp. Bc1 shows the concentration of the processing liquid used for the processing target substrate Wp. Rp1 shows the sublimation drying processing conditions that should be performed on the processing target substrate Wp. Thus, in the case where the depth of the trench of the processing target substrate Wp is Ad1 and the concentration of the processing liquid is Bc1, the substrate processing apparatus 100 performs the sublimation drying processing with the sublimation drying processing conditions shown by Rp1.

[0357] The conversion tables CT2 to CT1000 are also the same. Typically, for the conversion tables CT1 to CT1000, at least one of the substrate information and the processing liquid information is different.

[0358] Further, in the case where the values of the substrate information and the processing liquid information of the processing target substrate Wp do not match the values shown by the conversion table CT, the sublimation drying processing conditions of the processing target substrate Wp can also be determined by linear interpolation of the values of the sublimation drying processing conditions shown by the conversion table CT. Alternatively, the sublimation drying processing conditions of the processing target substrate Wp can also be determined by polynomial interpolation of the values of the sublimation drying processing conditions shown by the conversion table.

[0359] Embodiments of the present application have been described above with reference to the accompanying drawings. However, the present application is not limited to the above-described embodiments, but can be implemented in various ways without departing from the spirit of the present application. Further, various applications can be formed by appropriately combining a plurality of constituent elements disclosed in the above-described embodiments. For example, it is also possible to delete some of all the constituent elements shown in the embodiments. Further, the constituent elements in different embodiments can be appropriately combined. In order to easily understand the present application, the accompanying drawings show each constituent element substantially and schematically, and the thickness, length, number, interval, and the like of each constituent element shown in the drawings can be different from the actual ones due to the relationship of drawing. Further, the material, shape, size, and the like of each constituent element shown in the above-described embodiments are examples, and are not particularly limited, and various changes can be made within a range not substantially departing from the effects of the present application.

[0360] Industrial applicability

[0361] The present application is suitable for use in a substrate processing apparatus, a substrate processing method, a learning data generation method, a learning method, a learning apparatus, a learned model generation method, and a learned model.

[0362] Explanation of symbols

[0363] 10 substrate processing system

[0364] 20 control apparatus

[0365] 22 control section

[0366] 22a input information acquisition section

[0367] 22b sublimation drying processing condition information acquisition section

[0368] 24 storage section

[0369] LM learned model

[0370] 100 substrate processing apparatus

[0371] 130 chemical liquid supply section

[0372] 140 cleaning liquid supply section

[0373] 150 displacement liquid supply section

[0374] 160 processing liquid supply section

[0375] 200 substrate processing learning system

[0376] 300 learning data generation apparatus

[0377] 400 learning apparatus

Claims

1. A substrate processing apparatus comprising: a substrate holding section configured to rotatably hold a processing target substrate provided with a pattern of grooves and structures; a processing liquid supply section configured to supply a processing liquid containing a sublimation substance and a solvent to the processing target substrate; an input information acquisition section configured to acquire input information for the processing target substrate, the input information including information of at least one of substrate information and processing liquid information, the substrate information showing properties or formation conditions of the grooves and structures, the processing liquid information showing properties of the processing liquid; a sublimation drying processing condition information acquisition section configured to acquire, based on the input information, sublimation drying processing condition information for the processing target substrate from a learned model, the sublimation drying processing condition information showing sublimation drying processing conditions; and a control section configured to control the substrate holding section and the processing liquid supply section based on the sublimation drying processing condition information acquired in the sublimation drying processing condition information acquisition section to perform sublimation drying processing on the processing target substrate, wherein the learned model is constructed by machine learning of learning data, wherein the learning data is learning data for a learning target substrate provided with a pattern of grooves and structures, and is given information associated with: information of at least one of substrate information and processing liquid information, the substrate information showing properties or formation conditions of the grooves and structures, the processing liquid information showing properties of a processing liquid containing a sublimation substance and a solvent used when the learning target substrate is subjected to sublimation drying processing; sublimation drying processing condition information showing conditions for sublimation drying processing of the learning target substrate; and processing result information showing a result of sublimation drying processing of the learning target substrate with the processing liquid, and further comprising a storage section configured to store the learned model. Further comprising a storage section configured to store the learned model. For each of the processing target substrate and the learning target substrate, the substrate information includes information showing any one of a surface area of the substrate, a depth of the grooves, a width of the grooves, an aspect ratio of the grooves, and a density of grooves or structures. For each of the processing target substrate and the learning target substrate, the processing liquid information includes information showing any one of a concentration of the sublimation substance with respect to the processing liquid and a temperature. For each of the processing target substrate and the learning target substrate, the sublimation drying processing condition information includes information showing any one of a supply amount of the processing liquid, a spray pattern of the processing liquid, and a rotation speed of the substrate. The input information includes temperature information showing a temperature of the processing target substrate in the sublimation drying processing.

7. A substrate processing method comprising: rotatably holding a processing target substrate provided with a pattern of grooves and structures; ​ ​ ​ ​ ​ ​ 2. The substrate processing apparatus of claim 1, wherein ​ 3. The substrate processing apparatus according to claim 1 or 2, wherein ​ 4. The substrate processing apparatus according to any one of claims 1 and 2, wherein ​ 5. The substrate processing apparatus according to any one of claims 1 and 2, wherein ​ 6. The substrate processing apparatus according to any one of claims 1 and 2, wherein ​ ​ ​ acquiring input information for the processing target substrate, the input information including information of at least one of substrate information showing properties or formation conditions of the trench and the structure, and processing liquid information showing properties of a processing liquid including a sublimable substance and a solvent; acquiring sublimation drying processing condition information showing sublimation drying processing conditions for the processing target substrate from a learned model based on the input information; and subjecting the processing target substrate to sublimation drying processing in accordance with the sublimation drying processing conditions of the sublimation drying processing condition information; in the step of acquiring the sublimation drying processing condition information, the learned model is constructed by machine learning of learning data for a learning target substrate provided with a pattern of a trench and a structure, the learning data being given information associated with: information of at least one of substrate information showing properties or formation conditions of the trench and the structure, and processing liquid information showing properties of a processing liquid including a sublimable substance and a solvent used when the learning target substrate is subjected to sublimation drying processing; sublimation drying processing condition information showing conditions of sublimation drying processing performed on the learning target substrate; and processing result information showing results of sublimation drying processing performed on the learning target substrate. ​

Images