Control method, control device, and computer program

Abstract

To provide a control method, control device, and computer program for accurately adjusting chamber pressure.SOLUTION: A control method for controlling a substrate processing apparatus including a tank for storing a predetermined gas, a chamber to which the gas is supplied from the tank and used for substrate processing, and an exhaust valve for exhausting the gas from the chamber includes: acquiring a chamber pressure that is the pressure inside the chamber; acquiring a gas discharge rate from the tank to the chamber; acquiring a target value for future chamber pressure; specifying a future exhaust valve opening that minimizes the difference between a predicted future chamber pressure according to the acquired chamber pressure, the gas discharge rate, and the exhaust valve opening and the target value; and adjusting the exhaust valve opening according to the specified future exhaust valve opening.SELECTED DRAWING: Figure 1

Description

[Technical Field]

[0001] The present invention relates to a control method, a control device, and a computer program. [Background technology]

[0002] In substrate processing, which involves etching, film formation, or the like, on substrates such as semiconductor wafers, glass substrates, or flat panel substrates, the processing is carried out in a chamber. For example, with a substrate placed in the chamber, a predetermined gas is flowed into the chamber and processing is carried out. Patent Document 1 discloses an example of a substrate processing apparatus that carries out processing by flowing a gas into the chamber. [Prior art documents] [Patent documents]

[0003] [Patent Document 1] Japanese Patent Application Publication No. 2019-165117 Summary of the Invention [Problem to be solved by the invention]

[0004] In order to perform appropriate substrate processing using a chamber, it is necessary to appropriately adjust the chamber pressure, which is the pressure inside the chamber. Conventionally, the chamber pressure has been adjusted using feedback control, but it is difficult to adjust the chamber pressure accurately. The present disclosure provides a control method, a control device, and a computer program for accurately adjusting chamber pressure. [Means for solving the problem]

[0005] A control method according to one aspect of the present disclosure is a control method for controlling a substrate processing apparatus including a tank for storing a predetermined gas, a chamber to which the gas is supplied from the tank and in which substrate processing is performed, and an exhaust valve for exhausting the gas from the chamber, the control method comprising the steps of: acquiring a chamber pressure, which is the pressure inside the chamber; acquiring a gas discharge amount from the tank to the chamber; acquiring a target value for a future chamber pressure; identifying a future opening degree of the exhaust valve that minimizes an error between a predicted value of the future chamber pressure based on the acquired chamber pressure, the gas discharge amount, and an opening degree of the exhaust valve and the target value; and adjusting the opening degree of the exhaust valve based on the identified future opening degree of the exhaust valve. [Effects of the Invention]

[0006] According to the present disclosure, it is possible to provide a control method, a control device, and a computer program for adjusting the chamber pressure with high precision. [Brief explanation of the drawings]

[0007] [Figure 1] FIG. 2 is a block diagram showing an example of a functional configuration of the substrate processing apparatus. [Figure 2] 2 is a block diagram showing an example of the internal functional configuration of the control device according to the first embodiment. FIG. [Figure 3] 4 is a flowchart showing an example of a procedure of a process executed by the control device according to the first embodiment. [Figure 4] 10 is a flowchart showing a first example of a subroutine procedure for processing gas discharge amount calculation. [Figure 5] 10 is a diagram showing an example of the contents of a pressure difference table. [Figure 6] FIG. 10 is a conceptual diagram illustrating an example of the function of a pressure difference estimation model. [Figure 7] 10 is a flowchart showing a second example of the procedure of a subroutine for processing gas discharge amount calculation. [Figure 8] 10 is a table showing an example of the contents of a chamber pressure estimation model. [Figure 9] FIG. 10 is a conceptual diagram illustrating an example of the function of a chamber pressure estimation model. [Figure 10] FIG. 10 is a block diagram showing an example of the internal functional configuration of a control device according to a second embodiment. [Figure 11] FIG. 4 is a conceptual diagram illustrating an example of the function of a valve opening estimation model. [Figure 12] 10 is a flowchart showing an example of a procedure of a process executed by a control device according to a second embodiment. DETAILED DESCRIPTION OF THE INVENTION

[0008] The present invention will now be described in detail with reference to the drawings showing embodiments thereof. <Embodiment 1> 1 is a block diagram showing an example of the functional configuration of a substrate processing apparatus 100. The substrate processing apparatus 100 includes a chamber 2 in which substrate processing such as etching or film formation is performed. A substrate is placed inside the chamber 2, and a predetermined gas such as an etching gas or a film formation gas is introduced into the chamber 2, where the substrate processing such as etching or film formation is performed inside the chamber 2. A tank 4 that stores the predetermined gas is connected to the chamber 2 via a pipe. A chamber valve 51 is provided in the pipe connecting the chamber 2 and the tank 4. The chamber valve 51 is a valve that opens and closes. A mass flow controller (MFC) 53 is connected to the tank 4 via a pipe.

[0009] The MFC 53 supplies a predetermined gas to the tank 4. For example, the MFC 53 supplies the gas to the tank 4 at a constant flow rate. When the chamber valve 51 is closed, the tank 4 does not discharge the gas, the supply of gas from the MFC 53 continues, and the tank pressure, which is the pressure inside the tank 4, increases. When the chamber valve 51 is open, the tank 4 discharges the gas, and the gas is supplied from the tank 4 to the chamber 2 through the piping and the chamber valve 51.

[0010] A pump 54 is connected to the chamber 2 via piping. An exhaust valve 3 is provided in the middle of the piping connecting the chamber 2 and the pump 54. The exhaust valve 3 is an APC (Automatic Pressure Control) valve whose opening can be adjusted. The pump 54 operates to exhaust gas from the chamber 2. When the exhaust valve 3 is open, the gas is exhausted from the chamber 2 by the pump 54 through the piping and the exhaust valve 3. When the exhaust valve 3 is open, the gas is exhausted from the chamber 2, and by adjusting the opening of the exhaust valve 3, the amount of gas exhausted from the chamber 2 is adjusted, and the chamber pressure, which is the pressure inside the chamber 2, is adjusted.

[0011] A pressure gauge 21 is provided in the chamber 2. The chamber pressure is measured by the pressure gauge 21. A pressure gauge 41 is provided in the tank 4. The tank pressure is measured by the pressure gauge 41. A valve controller 52 is connected to the chamber valve 51. The valve controller 52 opens and closes the chamber valve 51. For example, the chamber valve 51 is a solenoid valve. The chamber valve 51 may be a valve whose opening degree is adjustable. An APC controller 31 is connected to the discharge valve 3. The APC controller 31 changes the opening degree of the discharge valve 3. The discharge valve 3 may be a valve other than an APC valve whose opening degree is adjustable.

[0012] The control device 1 is connected to the APC controller 31, the pressure gauge 21, the valve controller 52, the pressure gauge 41, and the MFC 53. The control device 1 acquires the chamber pressure measured by the pressure gauge 21 and the tank pressure measured by the pressure gauge 41. The control device 1 controls the APC controller 31, the valve controller 52, and the MFC 53. The control device 1 controls the valve controller 52 to control the opening and closing of the chamber valve 51. The control device 1 controls the APC controller 31 to adjust the opening of the discharge valve 3. The chamber pressure is adjusted by adjusting the opening of the discharge valve 3.

[0013] FIG. 2 is a block diagram showing an example of the internal functional configuration of the control device 1 according to the first embodiment. The control device 1 executes a control method for controlling the substrate processing apparatus 100. The control device 1 is configured using a computer such as a personal computer or a server device. The control device 1 includes an arithmetic unit 11, a memory 12, a storage unit 13, a reading unit 14, and an interface unit 15. The arithmetic unit 11 is configured using a processor, such as a CPU (Central Processing Unit), a GPU (Graphics Processing Unit), or a multi-core CPU. The arithmetic unit 11 may also be configured using a quantum computer. The memory 12 stores temporary data generated in conjunction with arithmetic operations. The memory 12 is, for example, a RAM (Random Access Memory). The storage unit 13 is nonvolatile, such as a hard disk or a nonvolatile semiconductor memory. The reading unit 14 reads information from a recording medium 10, such as an optical disc or a portable memory.

[0014] The interface unit 15 is connected to the APC controller 31, the pressure gauge 21, the valve controller 52, the pressure gauge 41, and the MFC 53. The calculation unit 11 acquires the chamber pressure measured by the pressure gauge 21 and the tank pressure measured by the pressure gauge 41 through the interface unit 15. The calculation unit 11 controls the APC controller 31, the valve controller 52, and the MFC 53 by sending and receiving control signals through the interface unit 15.

[0015] The calculation unit 11 causes the reading unit 14 to read the computer program (program product) 131 recorded on the recording medium 10, and stores the read computer program 131 in the storage unit 13. The calculation unit 11 executes processing to realize the functions of the control device 1 in accordance with the computer program 131. The computer program 131 may be stored in the storage unit 13 in advance, or may be downloaded from outside the control device 1. In this case, the control device 1 does not need to be equipped with the reading unit 14.

[0016] The computer program 131 can be deployed to run on a single computer or on multiple computers located at one site or distributed across multiple sites and interconnected by a communication network. That is, the control device 1 may be configured with multiple computers, and the computer program 131 may be executed on multiple computers connected via a communication network. The control device 1 may be configured using a cloud server.

[0017] The processing of each step described below for executing the control method can be executed by multiple computers. The processing of each step can also be executed by different computers. Data used during the processing can be stored in multiple computers. The processing of each step can also be executed using a virtual machine. The processing of each step can be executed by multiple calculation units. The processing of each step can also be executed by different calculation units.

[0018] The control device 1 adjusts the chamber pressure by adjusting the aperture of the exhaust valve 3. FIG. 3 is a flowchart showing an example of the procedure of the process executed by the control device 1 according to the first embodiment. Hereinafter, steps are abbreviated as "S." The control device 1 executes the following process by having the calculation unit 11 execute information processing in accordance with the computer program 131. The control device 1 acquires the latest chamber pressure (S101). In S101, the calculation unit 11 acquires the value of the chamber pressure measured by the pressure gauge 21 through the interface unit 15. The control device 1 then executes a process to calculate the gas discharge rate from the tank 4 to the chamber 2 (S102). The gas discharge rate is the amount of gas discharged from the tank 4 to the chamber 2 per cycle time in order to supply gas from the tank 4 to the chamber 2. Here, the cycle time is the time interval for adjusting the aperture of the exhaust valve 3, and is a predetermined time.

[0019] 4 is a flowchart showing a first example of the procedure of a subroutine for processing gas discharge rate calculation. The control device 1 acquires the latest tank pressure (S211). In S211, the calculation unit 11 acquires the value of the tank pressure measured by the pressure gauge 41 through the interface unit 15. The control device 1 acquires a tank pressure difference, which is the difference between the latest tank pressure and the tank pressure after the cycle time, according to the latest tank pressure and the latest chamber pressure (S212).

[0020] The memory unit 13 stores a pressure difference table 132 that records the relationship between tank pressure, chamber pressure, and tank pressure difference. FIG. 5 is a diagram showing an example of the contents of the pressure difference table 132. In the figure, the unit of pressure is Torr. The "***" included in the figure indicates the value of the tank pressure difference. A positive tank pressure difference value indicates that the tank pressure after the cycle time is lower than the most recent tank pressure. In the pressure difference table 132, the tank pressure difference value is recorded in association with each combination of tank pressure value and chamber pressure value.

[0021] The value of the tank pressure difference when the tank pressure and the chamber pressure are a certain combination of values ​​is theoretically or experimentally investigated and specified in advance. The pre-specified tank pressure difference values ​​are recorded in the pressure difference table 132. In S212, the calculation unit 11 obtains the tank pressure difference by reading, from the pressure difference table 132, the tank pressure difference value associated with the latest chamber pressure obtained in S101 and the tank pressure obtained in S211.

[0022] The control device 1 may be configured to acquire the tank pressure difference using a trained model instead of the pressure difference table 132. In this configuration, the control device 1 includes a pressure difference estimation model 133. The pressure difference estimation model 133 is a trained model. The pressure difference estimation model 133 is realized by the calculation unit 11 executing information processing in accordance with the computer program 131. The storage unit 13 stores data for realizing the pressure difference estimation model 133. The pressure difference estimation model 133 may be configured using hardware. The pressure difference estimation model 133 may be realized using a quantum computer. Alternatively, the pressure difference estimation model 133 may be provided external to the control device 1, and the control device 1 may execute processing using the external pressure difference estimation model 133. For example, the pressure difference estimation model 133 may be configured using the cloud.

[0023] 6 is a conceptual diagram showing an example of the function of the pressure difference estimation model 133. The pressure difference estimation model 133 is trained in advance so as to output a tank pressure difference when a tank pressure and a chamber pressure are input. The pressure difference estimation model 133 is configured using a neural network. The pressure difference estimation model 133 is trained in advance using data in which the tank pressure, chamber pressure, and tank pressure difference are associated, such as that recorded in the pressure difference table 132, as training data.

[0024] In S212, the calculation unit 11 inputs the latest chamber pressure acquired in S101 and the tank pressure acquired in S211 to the pressure difference estimation model 133, and causes the pressure difference estimation model 133 to execute processing. The pressure difference estimation model 133 performs calculations in response to the input tank pressure and chamber pressure, and outputs the tank pressure difference. The calculation unit 11 acquires the tank pressure difference output by the pressure difference estimation model 133.

[0025] The control device 1 calculates the gas discharge amount based on the acquired tank pressure difference (S213). The flow rate of gas discharged from the tank 4 is defined as the discharge flow rate. The discharge flow rate can be considered as the change in the amount of gas per unit time. The gas pressure is defined as P, the gas volume as V, the amount of gas substance as n, the gas constant as R, and the temperature as T. The current time point is defined as t, and the cycle time as Δt. The amount of gas substance at the current time point is defined as n. t , pressure P t The amount of gas after the cycle time is n t+Δt , pressure P t+Δt Let's say.

[0026] According to the gas equation of state PV=nRT, the amount of gas at the present time is n t =P t V / RT, and the amount of gas after the cycle time is n t+Δt =P t+Δt V / RT. The change in the amount of gas until the end of the cycle time is Δn, and the change in gas pressure is ΔP. ​​Δn=n t+Δt -n t =(P t+Δt -P t ) / RT = ΔPV / RT. The discharge flow rate is Δn / Δt = C × ΔPV / ΔtRT. Here, C is a conversion coefficient for converting the discharge flow rate unit from [mol / sec] to [sccm (standard cubic centimeter per minute)]. Specifically, C = 1.34484 × 10 6 The amount of gas discharged per cycle time (gas discharge amount) is the discharge flow rate multiplied by Δt. Therefore, the gas discharge amount is expressed by the following formula (1). Gas discharge rate = C × ΔPV / RT … (1)

[0027] The unit of the gas discharge rate calculated by equation (1) is [cc (cubic centimeter)]. A temperature sensor is provided in the tank 4, and the control device 1 acquires the temperature inside the tank 4 measured by the temperature sensor. The control device 1 stores the values ​​of C, the internal volume V of the tank 4, and the gas constant R in advance in the memory unit 13. Since ΔP is the tank pressure difference, in S213, the calculation unit 11 calculates the gas discharge rate by performing the calculation of equation (1) using the acquired tank pressure difference. The calculation unit 11 stores the calculated value of the gas discharge rate in the memory unit 13. After S213 is completed, the control device 1 ends the gas discharge rate calculation process of S102 and returns the process to the main.

[0028] 7 is a flowchart showing a second example of the procedure of the subroutine for processing gas discharge rate calculation. Gas continuously flows into the tank 4 from the MFC 53 at a constant flow rate. Furthermore, while gas flows in from the MFC 53, the tank 4 stops supplying gas to the chamber 2 for a predetermined first period, and then supplies the stored gas to the chamber 2 for a predetermined second period. The first period and the second period are repeated. That is, the tank 4 repeatedly starts and stops supplying gas to the chamber 2. The control device 1 acquires the flow rate of gas from the MFC 53 to the tank 4, the length of the first period, and the length of the second period (S221).

[0029] The flow rate of the gas from the MFC 53 to the tank 4, the length of the first period, and the length of the second period are predetermined and stored in the memory unit 13. In S221, the calculation unit 11 acquires the gas flow rate, the length of the first period, and the length of the second period by reading the values ​​of the gas flow rate, the length of the first period, and the length of the second period from the memory unit 13. The calculation unit 11 may acquire the flow rate of the gas supplied by the MFC 53 to the tank 4 from the MFC 53. Alternatively, the substrate processing apparatus 100 may include a flow meter that measures the flow rate of the gas from the MFC 53 to the tank 4, and the control device 1 may acquire the flow rate measured by the flow meter.

[0030] The control device 1 calculates the discharge flow rate (S222). It has been experimentally demonstrated that the average discharge flow rate obtained by averaging the flow rate of gas discharged from the tank 4 over the second period can be approximated by the following equation (2). Average discharge flow rate = (flow rate of gas from MFC 53 to tank 4) × (length of first period + length of second period) / (length of second period) ... (2) In S222, the calculation unit 11 calculates the average discharge flow rate as the discharge flow rate using equation (2).

[0031] The control device 1 calculates the gas discharge rate (S223). In S223, the calculation unit 11 calculates the gas discharge rate by multiplying the calculated discharge flow rate by the cycle time. The calculation unit 11 stores the calculated value of the gas discharge rate in the storage unit 13. After S2123 is completed, the control device 1 ends the gas discharge rate calculation process of S102 and returns the process to the main. The control device 1 performs either the process of S211 to S213 or S221 to S223.

[0032] After completing the gas discharge rate calculation process in S102, the control device 1 determines the predicted value of the chamber pressure after the cycle time (S103). The memory unit 13 stores a chamber pressure table 134 that records the relationship between the chamber pressure, gas discharge rate, exhaust valve 3 opening, and the predicted value of the chamber pressure after the cycle time. FIG. 8 is a diagram showing an example of the contents of the chamber pressure estimation model 135. In the diagram, the opening of the exhaust valve 3 is taken as the valve opening. The unit of pressure is Torr, the unit of gas discharge rate is cc, and the unit of valve opening is deg (degree). The "****" included in the diagram indicates the predicted value of the chamber pressure after the cycle time.

[0033] In the chamber pressure table 134, a table in which predicted chamber pressure values ​​after a cycle time associated with each combination of chamber pressure value and gas discharge rate value is recorded is associated with each valve opening value. The opening of the exhaust valve 3 can take multiple values. Therefore, each valve opening value is associated with a combination of chamber pressure, gas discharge rate, and chamber pressure after a cycle time.

[0034] The predicted chamber pressure value after the cycle time when the chamber pressure, gas discharge rate, and exhaust valve 3 aperture are a certain combination is theoretically or experimentally investigated and specified in advance. The chamber pressure table 134 stores the predicted chamber pressure value after the pre-specified cycle time. In S103, the calculation unit 11 reads out the predicted chamber pressure value after the cycle time associated with the latest chamber pressure acquired in S101, the gas discharge rate calculated in S102, and the latest aperture of the exhaust valve 3 controlled by the control device 1, thereby specifying the predicted chamber pressure value after the cycle time. The calculation unit 11 stores the specified predicted chamber pressure value in the memory unit 13.

[0035] The control device 1 may be configured to determine a predicted value of the chamber pressure after the cycle time using a trained model instead of the chamber pressure table 134. In this configuration, the control device 1 includes a chamber pressure estimation model 135. The chamber pressure estimation model 135 is a trained model. The chamber pressure estimation model 135 is realized by the calculation unit 11 executing information processing in accordance with the computer program 131. The storage unit 13 stores data for realizing the chamber pressure estimation model 135. The chamber pressure estimation model 135 may be configured using hardware. The chamber pressure estimation model 135 may be realized using a quantum computer. Alternatively, the chamber pressure estimation model 135 may be provided external to the control device 1, and the control device 1 may execute processing using the external chamber pressure estimation model 135. For example, the chamber pressure estimation model 135 may be configured using the cloud.

[0036] 9 is a conceptual diagram showing an example of the function of the chamber pressure estimation model 135. The chamber pressure estimation model 135 is trained in advance so as to output a predicted value of the chamber pressure after the cycle time when the chamber pressure, the gas discharge rate, and the aperture of the exhaust valve 3 are input. The chamber pressure estimation model 135 is configured using a neural network. The chamber pressure estimation model 135 is trained in advance using, as training data, data in which the chamber pressure, the gas discharge rate, the aperture of the exhaust valve 3, and the predicted value of the chamber pressure after the cycle time are associated, as recorded in the chamber pressure table 134.

[0037] In S103, the calculation unit 11 inputs the latest chamber pressure acquired in S101, the gas discharge rate calculated in S102, and the latest opening of the exhaust valve 3 controlled by the control device 1 into the chamber pressure estimation model 135, and causes the chamber pressure estimation model 135 to execute processing. The chamber pressure estimation model 135 performs calculations in response to the input of the chamber pressure, the gas discharge rate, and the opening of the exhaust valve 3, and outputs a predicted value of the chamber pressure after the cycle time. The calculation unit 11 acquires the predicted value of the chamber pressure after the cycle time output by the chamber pressure estimation model 135. In this way, the calculation unit 11 determines the predicted value of the chamber pressure after the cycle time.

[0038] The control device 1 acquires the target value of the chamber pressure after the cycle time (S104). The target value of the chamber pressure after the cycle time is determined in advance. For example, if control is performed to maintain a constant chamber pressure, the target value of the chamber pressure is a constant value. For example, if the chamber pressure is changed over time, such as by changing the chamber pressure linearly over time, the target value of the chamber pressure will be a value that varies over time. The target value of the chamber pressure is stored in advance in the memory unit 13, and in S104, the calculation unit 11 acquires the target value of the chamber pressure after the cycle time by reading the target value of the chamber pressure from the memory unit 13. Alternatively, the chamber pressure may be expressed by a predetermined function, the data of the function may be stored in the memory unit 13, and the calculation unit 11 may calculate the target value of the chamber pressure after the cycle time based on the function.

[0039] The control device 1 calculates the error between the predicted value and the target value of the chamber pressure after the cycle time (S105). In S105, the calculation unit 11 calculates the error between the predicted value and the target value of the chamber pressure after the cycle time by subtracting the target value of the chamber pressure acquired in S104 from the predicted value of the chamber pressure calculated in S103. The calculation unit 11 may calculate the absolute value or square of the difference between the predicted value and the target value as the error between the predicted value and the target value of the chamber pressure after the cycle time. The calculation unit 11 stores the calculated error in the memory unit 13.

[0040] The control device 1 determines a predicted chamber pressure value after another cycle time from the time point associated with the calculated predicted chamber pressure value (S106). For example, after S105 is completed, the control device 1 determines a predicted chamber pressure value after twice the cycle time has elapsed. In S106, the calculation unit 11 determines a predicted chamber pressure value after another cycle time from the time point associated with the calculated predicted chamber pressure value based on multiple possible values ​​for the exhaust valve 3 aperture at the time point associated with the calculated predicted chamber pressure value, the calculated predicted chamber pressure value, and the gas discharge rate. For example, the aperture of the exhaust valve 3 can be adjusted in 5-degree increments, such as 0 degrees, 5 degrees, 10 degrees, etc., and the calculation unit 11 determines a predicted chamber pressure value after another cycle time for all apertures.

[0041] In S106, the calculation unit 11 performs the same process as S103, using the calculated predicted value of the chamber pressure as the "chamber pressure" and a possible value for the aperture of the exhaust valve 3 at the time associated with the calculated predicted value of the chamber pressure as the "aperture of the exhaust valve 3." Specifically, the calculation unit 11 reads out the predicted value of the chamber pressure associated with each value of the aperture of the exhaust valve 3, the "chamber pressure" that is the calculated predicted value of the chamber pressure, and the gas discharge rate from the chamber pressure table 134. Alternatively, the calculation unit 11 inputs each value of the aperture of the exhaust valve 3, the calculated predicted value of the chamber pressure, and the gas discharge rate to the chamber pressure estimation model 135, and obtains the predicted value of the chamber pressure output by the chamber pressure estimation model 135.

[0042] In S106, the calculation unit 11 uses the gas discharge rate calculated in S102 as the gas discharge rate. Alternatively, the calculation unit 11 may calculate a predicted value of the gas discharge rate at a time point related to the calculated predicted value of the chamber pressure, and use the calculated predicted value of the gas discharge rate as the "gas discharge rate" to perform the process of S106. For example, the calculation unit 11 may calculate a predicted value of the tank pressure at that time point from the tank pressure difference, and calculate a predicted value of the gas discharge rate from the predicted value of the tank pressure.

[0043] By the process of S106, a predicted value of the chamber pressure after a further cycle time from a certain point in time is determined for each value of the exhaust valve 3 opening at that point in time. That is, the calculation unit 11 determines multiple predicted values ​​of the chamber pressure after a further cycle time from that point in time, according to multiple possible values ​​of the exhaust valve 3 opening at that point in time. The calculation unit 11 stores multiple predicted values ​​of the chamber pressure corresponding to multiple exhaust valve 3 openings in the memory unit 13.

[0044] The control device 1 then acquires the target value of the chamber pressure after another cycle time (S107). For example, if the chamber pressure is controlled to be constant, the target value of the chamber pressure is the same as the value acquired in S104. For example, if the chamber pressure is changed over time, the target value of the chamber pressure will be a value that corresponds to the time. In S107, the calculation unit 11 acquires the target value of the chamber pressure by reading it from the storage unit 13. Alternatively, the control device 1 may calculate the target value of the chamber pressure based on a predetermined function.

[0045] The control device 1 calculates the error between the predicted value and the target value of the chamber pressure (S108). In S108, the calculation unit 11 calculates the error by the same process as in S105. The calculation unit 11 calculates the error for each possible value of the opening of the exhaust valve 3. That is, the calculation unit 11 calculates multiple errors according to the multiple possible values ​​of the opening of the exhaust valve 3. The calculation unit 11 stores the calculated errors in the memory unit 13.

[0046] The control device 1 determines whether the processing of S106 to S108 has been repeated a predetermined number of times (S109). If the processing of S106 to S108 has not yet been repeated the predetermined number of times (S109: NO), the control device 1 returns to S106 and repeats the processing of S106 to S108. At this time, the calculation unit 11 performs the processing of S106 to S108 for each possible value of the aperture of the exhaust valve 3. By repeating the processing of S106 to S108, the error between the predicted value and the target value of the chamber pressure at each time point is obtained for each of the apertures of the exhaust valve 3 at multiple future time points. The storage unit 13 stores multiple combinations of the aperture of the exhaust valve 3 at multiple future time points and the error between the predicted value and the target value of the chamber pressure at each time point.

[0047] If the processes of S106 to S108 have been repeated a predetermined number of times (S109: YES), the control device 1 calculates the sum of errors between the predicted and target values ​​of the chamber pressure over multiple future time points (S110). Through the processes of S103 to S109, the errors between the predicted and target values ​​of the chamber pressure at each time point from the time after the cycle time to the time after {(predetermined number + 1) × cycle time} for all possible values ​​of the aperture of the exhaust valve 3 at each time point from the time after the cycle time to the time after (predetermined number × cycle time) have been calculated. In S110, the calculation unit 11 sums up the errors between the predicted and target values ​​of the chamber pressure at multiple time points for each of multiple combinations of the aperture of the exhaust valve 3 at multiple future time points. If the number of possible values ​​of the aperture of the exhaust valve 3 is a and the predetermined number is b, the combinations of values ​​that the aperture of the exhaust valve 3 can take at multiple future time points are a b The calculation unit 11 calculates (b+1) errors for each combination. b For each combination of street openings, sum the (b+1) errors.

[0048] The control device 1 determines the future opening degree of the exhaust valve 3 so as to minimize the error between the predicted value and the target value of the chamber pressure (S111). In S111, the calculation unit 11 determines the future opening degree of the exhaust valve 3 by selecting, from among multiple possible values ​​for the future opening degree of the exhaust valve 3, a value that minimizes the sum of the errors between the predicted value and the target value of the chamber pressure calculated in S110. Here, "the sum of the errors between the predicted value and the target value of the chamber pressure is minimized" may also be "the absolute value of the sum of the errors between the predicted value and the target value of the chamber pressure is minimized."

[0049] In S111, the calculation unit 11 specifies only the opening degree of the discharge valve 3 after the cycle time from the present time as the future opening degree of the discharge valve 3. More specifically, the calculation unit 11 performs a b From the combinations of the opening degrees, one combination that minimizes the total error between the predicted value and the target value of the chamber pressure is selected, and the opening degree of the exhaust valve 3 at the end of the cycle time from the current time point, which is included in the selected combination of opening degrees, is identified as the future opening degree of the exhaust valve 3. The calculation unit 11 may identify the opening degrees of the exhaust valve 3 at multiple points in time as the future opening degree of the exhaust valve 3. The calculation unit 11 stores the identified value of the future opening degree of the exhaust valve 3 in the memory unit 13.

[0050] The control device 1 then adjusts the opening degree of the discharge valve 3 according to the identified future opening degree of the discharge valve 3 (S112). In S112, the calculation unit 11 transmits a control signal from the interface unit 15 to the APC controller 31 to set the opening degree of the discharge valve 3 after the cycle time to the opening degree identified in S111. The APC controller 31 controls the discharge valve 3 in accordance with the control signal so that the opening degree after the cycle time becomes the opening degree identified in S111. In this way, the opening degree of the discharge valve 3 is adjusted.

[0051] After S112 is completed, the control device 1 ends the process. The processes of S101 to S112 are repeatedly executed. The control device 1 executes the processes of S101 to S112 for each cycle time.

[0052] As described above in detail, the control device 1 specifies the future aperture of the exhaust valve 3 so as to minimize the error between the predicted value of the future chamber pressure, which is determined based on the chamber pressure, the gas discharge rate, and the aperture of the exhaust valve 3, and the target value of the future chamber pressure. The control device 1 adjusts the aperture of the exhaust valve 3 based on the specified aperture, thereby making it possible to bring the aperture of the exhaust valve 3 as close as possible to the target value. In this manner, in this embodiment, feedforward control is performed using the predicted value of the chamber pressure. Because feedforward control is performed, the chamber pressure can be adjusted more accurately than a method of adjusting the chamber pressure using only feedback control.

[0053] The substrate processing apparatus 100 can accurately adjust the chamber pressure, thereby stabilizing the chamber pressure during substrate processing and enabling stable substrate processing. Since the time during which the chamber pressure becomes unstable and the substrate processing becomes unstable is reduced, the length of time during which stable substrate processing can be performed within a given period of time is increased, thereby improving the efficiency of substrate processing or shortening the time required for substrate processing. Furthermore, by appropriately setting a target value for the chamber pressure, the control device 1 can adjust the waveform of the chamber pressure change over time to any waveform, such as changing the chamber pressure linearly over time. By adjusting the waveform of the chamber pressure change over time, the substrate processing apparatus 100 can perform substrate processing at any chamber pressure.

[0054] <Embodiment 2> In the first embodiment, the required amount of calculation is enormous, which may make processing difficult. In the second embodiment, a configuration in which the amount of calculation is reduced is shown. FIG. 10 is a block diagram showing an example of the internal functional configuration of a control device 1 according to the second embodiment. The control device 1 includes a valve opening estimation model 136. The valve opening estimation model 136 is a trained model. The valve opening estimation model 136 is realized by the calculation unit 11 executing information processing in accordance with a computer program 131. The storage unit 13 stores data for realizing the valve opening estimation model 136. The valve opening estimation model 136 may be configured using hardware. The valve opening estimation model 136 may be realized using a quantum computer. Alternatively, the valve opening estimation model 136 may be provided external to the control device 1, and the control device 1 may execute processing using the external valve opening estimation model 136. For example, the valve opening estimation model 136 may be configured using the cloud. The other configurations of the control device 1 are the same as those in the first embodiment.

[0055] 11 is a conceptual diagram showing an example of the function of the valve position estimation model 136. The valve position estimation model 136 is trained in advance so as to output the future position of the exhaust valve 3 when the chamber pressure, gas discharge rate, exhaust valve 3 position, and target values ​​for future chamber pressure are input. The valve position estimation model 136 is configured using a neural network.

[0056] The valve position estimation model 136 is trained in advance using training data that associates the chamber pressure, the gas discharge rate, the position of the exhaust valve 3, and the target value of the future chamber pressure with the future position of the exhaust valve 3 that minimizes the error between the predicted value and the target value of the future chamber pressure. The training data is created theoretically or experimentally. For example, the training data may be created using the future position of the exhaust valve 3 determined by calculations similar to those in S101 to S111 in the first embodiment.

[0057] 12 is a flowchart showing an example of the procedure of the process executed by the control device 1 according to the second embodiment. The control device 1 acquires the latest chamber pressure (S31), and executes a process of calculating the gas discharge rate (S32) similar to the process of S102. The control device 1 then acquires a target value of the chamber pressure after the cycle time (S33).

[0058] The control device 1 then inputs the latest chamber pressure, gas discharge rate, latest exhaust valve 3 aperture, and future chamber pressure target value to the valve aperture estimation model 136 (S34). In S34, the calculation unit 11 inputs the latest chamber pressure acquired in S31, the gas discharge rate calculated in S32, and the latest exhaust valve 3 aperture controlled by the control device 1 to the valve aperture estimation model 136. The calculation unit 11 also inputs the chamber pressure target value after the cycle time acquired in S33 to the valve aperture estimation model 136 as the future chamber pressure target value.

[0059] The calculation unit 11 causes the valve opening estimation model 136 to execute processing. The valve opening estimation model 136 performs calculations in response to inputs of the latest chamber pressure, the gas discharge rate, the latest opening of the exhaust valve 3, and a target value for the future chamber pressure, and outputs the opening of the exhaust valve 3 after the cycle time as the future opening of the exhaust valve 3. The valve opening estimation model 136 may be configured to output the opening of the exhaust valve 3 at multiple future points in time as the future opening of the exhaust valve 3.

[0060] The calculation unit 11 then identifies the future opening degree of the discharge valve 3 (S35). In S35, the calculation unit 11 identifies the future opening degree of the discharge valve 3 by acquiring the opening degree of the discharge valve 3 after the cycle time output by the valve opening degree estimation model 136. The calculation unit 11 identifies the opening degree of the discharge valve 3 after the cycle time as the future opening degree of the discharge valve 3.

[0061] The control device 1 then adjusts the opening degree of the discharge valve 3 according to the identified future opening degree of the discharge valve 3 (S36). In S36, the calculation unit 11 sends a control signal from the interface unit 15 to the APC controller 31 to set the opening degree of the discharge valve 3 after the cycle time to the opening degree identified in S35. The APC controller 31 controls the discharge valve 3 in accordance with the control signal so that the opening degree after the cycle time will be the opening degree identified in S35. In this way, the opening degree of the discharge valve 3 is adjusted.

[0062] After S36 is completed, the control device 1 ends the process. The processes of S31 to S36 are repeatedly executed. The control device 1 executes the processes of S31 to S36 for each cycle time.

[0063] In the second embodiment, the control device 1 can also specify the future aperture of the exhaust valve 3 so that the error between the predicted value of the future chamber pressure, which is based on the chamber pressure, the gas discharge rate, and the aperture of the exhaust valve 3, and the target value of the future chamber pressure is minimized. The control device 1 can adjust the chamber pressure with high accuracy by adjusting the aperture of the exhaust valve 3 according to the specified aperture. In the second embodiment, the use of a trained model can reduce the amount of calculation required for processing to specify the future aperture of the exhaust valve 3. This allows the control device 1 to achieve high-precision adjustment of the chamber pressure in real time.

[0064] In the second embodiment, the substrate processing apparatus 100 can also adjust the chamber pressure with high precision, thereby enabling stable substrate processing. As in the first embodiment, the second embodiment also makes it possible to improve the efficiency of substrate processing or shorten the time required for substrate processing. Furthermore, as in the first embodiment, the control apparatus 1 can adjust the waveform of the time change in chamber pressure to an arbitrary waveform.

[0065] The present invention is not limited to the above-described embodiments, and various modifications are possible within the scope of the claims. In other words, embodiments obtained by combining technical means modified appropriately within the scope of the claims are also included in the technical scope of the present invention.

[0066] The matters described in each embodiment can be combined with each other. In addition, the independent claims and dependent claims described in the claims can be combined with each other in any combination, regardless of the reference format. Furthermore, although the claims use a format in which a claim references two or more other claims (multiple claim format), this is not limited to this format. A multiple claim (multi-multi claim) that references at least one other multiple claim may also be used. [Explanation of symbols]

[0067] 100 Substrate processing apparatus 1. Control device 10 Recording media 11 Arithmetic section 13 Storage section 131 Computer Programs 2 chambers 3 Exhaust valve 4 Tank

Claims

1. 1. A control method for controlling a substrate processing apparatus including: a tank for storing a predetermined gas; a chamber to which the gas is supplied from the tank and in which substrate processing is performed; and an exhaust valve for exhausting the gas from the chamber, the method comprising: obtaining a chamber pressure, which is the pressure inside the chamber; obtaining a gas discharge amount from the tank to the chamber; obtaining a target value for the future chamber pressure; Identifying a future opening degree of the exhaust valve that minimizes an error between a predicted value of the future chamber pressure according to the acquired chamber pressure, the gas discharge rate, and the opening degree of the exhaust valve and the target value; The opening degree of the exhaust valve is adjusted according to the identified future opening degree of the exhaust valve. Control method.

2. determining a predicted value of the chamber pressure after a predetermined time based on the latest chamber pressure, the gas discharge amount, and the latest opening degree of the exhaust valve; Calculating the error between the predicted and target chamber pressure values ​​after a predetermined time; repeatedly determining a plurality of predicted chamber pressure values ​​a predetermined time after a future time point in accordance with a plurality of possible values ​​of the predicted chamber pressure at the future time point, the gas discharge rate, and the opening degree of the exhaust valve at the future time point, and calculating a plurality of errors between the predicted chamber pressure value and the target chamber pressure value at the future time point in accordance with the plurality of possible values ​​of the predicted chamber pressure value, the gas discharge rate, and the opening degree of the exhaust valve at the future time point in accordance with the plurality of possible values ​​of the predicted chamber pressure value, the gas discharge rate, and the opening degree of the exhaust valve at the future time point in accordance with the plurality of possible values ​​of the predicted chamber pressure value, the target chamber pressure value, and the target chamber pressure value at the future time point in accordance with the plurality of possible values ​​of the predicted chamber pressure value, the target chamber pressure value and the target chamber pressure value at the future time point in accordance with the plurality of possible values ​​of the predicted chamber pressure value, the target chamber pressure value and the target chamber pressure value at the future time point in accordance with the plurality of possible values ​​of the predicted chamber pressure value, the target chamber pressure value and the target chamber pressure value at the future time point in accordance with the plurality of possible values ​​of the predicted chamber pressure value, the target chamber pressure value and the target chamber pressure value at the future time point in accordance with the plurality of possible values ​​of the predicted chamber pressure value, the target chamber pressure value and the target chamber pressure value at the future time point in accordance with the plurality of possible values ​​of the target ... The future opening degree of the exhaust valve is determined by selecting, from among a plurality of possible values ​​for the future opening degree of the exhaust valve, a value that minimizes the sum of errors between the predicted value and the target value of the chamber pressure over a plurality of future time points. The control method according to claim 1 .

3. The predicted value of the chamber pressure after a predetermined time is read from a table in which the relationship between the chamber pressure, the gas discharge amount, the exhaust valve opening, and the predicted value of the chamber pressure after a predetermined time is recorded, thereby identifying the predicted value of the chamber pressure. The control method according to claim 1 or 2.

4. A predicted value of the chamber pressure is identified using a trained model that outputs a predicted value of the chamber pressure after a predetermined time when the chamber pressure, the gas discharge amount, and the opening of the exhaust valve are input. The control method according to claim 2 .

5. The future opening degree of the exhaust valve is identified using a trained model that outputs the future opening degree of the exhaust valve that minimizes the error between the predicted value of the future chamber pressure and the target value when the latest chamber pressure, the gas discharge amount, the target value, and the latest opening degree of the exhaust valve are input. The control method according to claim 1 .

6. A tank pressure difference is obtained, which is the difference between the tank pressure, which is the pressure inside the tank, and the tank pressure after a predetermined time. Calculate the gas discharge amount based on the acquired tank pressure difference. The control method according to claim 1 .

7. Get the latest tank pressure, The tank pressure difference is obtained by reading the latest tank pressure and the tank pressure difference corresponding to the latest chamber pressure from a table that records the relationship between the tank pressure, chamber pressure, and tank pressure difference. The control method according to claim 6.

8. Get the latest tank pressure, The tank pressure difference is obtained using a trained model that outputs the tank pressure difference when the tank pressure and chamber pressure are input. The control method according to claim 6.

9. The tank is continuously filled with the gas, a first period during which the gas flows into the tank while the supply of the gas from the tank to the chamber is stopped, and a second period during which the gas is supplied from the tank to the chamber are repeated; The gas discharge amount is calculated based on the flow rate of the gas flowing into the tank, the length of the first period, and the length of the second period. The control method according to claim 1 .

10. A control device for controlling a tank that stores a predetermined gas, a chamber that receives the gas from the tank and performs substrate processing, and an exhaust valve that exhausts the gas from the chamber, A calculation unit is provided, The calculation unit obtaining a chamber pressure, which is the pressure inside the chamber; obtaining a gas discharge amount from the tank to the chamber; obtaining a target value for the future chamber pressure; Identifying a future opening degree of the exhaust valve that minimizes an error between a predicted value of the future chamber pressure according to the acquired chamber pressure, the gas discharge rate, and the opening degree of the exhaust valve and the target value; The opening degree of the exhaust valve is adjusted according to the identified future opening degree of the exhaust valve. Control device.

11. A computer program that causes a computer to execute a process for controlling a tank that stores a predetermined gas, a chamber that receives the gas from the tank and performs substrate processing, and an exhaust valve that exhausts the gas from the chamber, obtaining a chamber pressure, which is the pressure inside the chamber; obtaining a gas discharge amount from the tank to the chamber; obtaining a target value for the future chamber pressure; Identifying a future opening degree of the exhaust valve that minimizes an error between a predicted value of the future chamber pressure according to the acquired chamber pressure, the gas discharge rate, and the opening degree of the exhaust valve and the target value; The opening degree of the exhaust valve is adjusted according to the identified future opening degree of the exhaust valve. A computer program that causes a computer to perform a process.

Images