Temperature estimation device, plasma processing system, temperature estimation method, and temperature estimation program

Abstract

This invention provides a temperature estimation device, a plasma processing system, a temperature estimation method, and a temperature estimation program for estimating the temperature corresponding to changes in the state within a plasma processing space. The temperature estimation device includes an estimation unit that sequentially estimates temperature data by inputting new time-series process data related to the state within the plasma processing space into a generated time-series model. The generated time-series model is a model that establishes a correlation between the time-series process data related to the state within the processing space and the temperature data of the time-series measured within the processing space at each time interval.

Description

Technical Field

[0001] This invention relates to a temperature estimation device, a plasma processing system, a temperature estimation method, and a temperature estimation procedure. Background Technology

[0002] In semiconductor manufacturing processes, maintaining the temperature within the plasma-processed wafer is crucial for ensuring its quality. However, in semiconductor manufacturing, it is difficult to permanently install temperature sensors within the processing space and directly measure its temperature. Therefore, the temperature within the processing space is determined indirectly through estimation or other methods.

[0003] As an example, Patent Document 1, etc., discloses a method for estimating the temperature within the processing space using process conditions during plasma processing. According to this method, by obtaining the process conditions before starting plasma processing, the temperature within the processing space during plasma processing can be estimated.

[0004] However, even under the same process conditions, the accuracy of temperature estimation decreases when the state within the processing space changes (e.g., when the temperature rises / falls at the start of plasma processing, or when the temperature of the components within the processing space rises during plasma processing).

[0005] Patent Document 1: Japanese Patent Application Publication No. 2019-212852 Summary of the Invention

[0006] This disclosure provides a temperature estimation device, a plasma processing system, a temperature estimation method, and a temperature estimation procedure for estimating the temperature corresponding to changes in the state within a processing space where plasma processing is performed.

[0007] According to one aspect of this disclosure, a temperature estimation apparatus has, for example, the following structure: It includes an estimation unit that sequentially estimates temperature data by inputting new time-series process data related to the state within the processing space where plasma processing is performed into a generated time-series model. The generated time-series model is a model that correlates data for each time range of the time-series process data related to the state within the processing space with data for each time period of the temperature data measured within the processing space.

[0008] According to this disclosure, a temperature estimation device, a plasma processing system, a temperature estimation method, and a temperature estimation procedure are provided to estimate the temperature corresponding to changes in the state within a processing space where plasma processing is performed. Attached Figure Description

[0009] Figure 1This is a diagram illustrating an example of the system structure of a plasma processing system using an applied temperature estimation device during the model generation phase.

[0010] Figure 2 This is a diagram illustrating an example of the system structure of a plasma processing system using an application temperature estimation device during the estimation phase.

[0011] Figure 3 This is a diagram illustrating an example of the hardware structure of a temperature estimation device.

[0012] Figure 4 This is a diagram representing an example of data used for model generation (process data sets used for model generation and temperature data used for model generation).

[0013] Figure 5 This is a graph representing an example of measured values ​​of process data and temperature data.

[0014] Figure 6 It is a flowchart representing the process of model generation and processing.

[0015] Figure 7 This is a diagram illustrating the functional structure and specific examples of the presumption section.

[0016] Figure 8 This is a diagram illustrating the functional structure and specific examples of the processing of the decision-making unit.

[0017] Figure 9 This is a diagram illustrating the functional structure and specific examples of the processing of the control unit.

[0018] Figure 10 This is a flowchart representing the presumed processing flow.

[0019] Explanation of reference numerals in the attached figures

[0020] 100: Plasma processing system; 110: Wafer before processing; 120: Chamber; 130: Wafer after processing; 140_1~140_n: Time series data acquisition device; 151: Temperature sensor; 152: Temperature sensor; 160: Temperature estimation device; 161: Model generation unit; 200: Estimation unit; 210: Judgment unit; 220: Control unit; 710: Completed time series model; 810: Detachment amount calculation unit; 820: Judgment reference setting unit; 830: Reporting unit; 910: Temperature adjustment unit; 920: Target temperature setting unit; 930: Process adjustment unit. Detailed Implementation

[0021] Hereinafter, various embodiments will be described with reference to the accompanying drawings. Furthermore, in this specification and the accompanying drawings, components having substantially the same functional configuration are omitted from repeated descriptions by using the same reference numerals.

[0022] [First Implementation Method]

[0023] <System Structure of Plasma Processing System>

[0024] First, the system structure of the plasma processing system using the temperature estimation device according to the first embodiment will be described. Furthermore, in the first embodiment, the temperature estimation device has functions performed during the model generation stage and functions performed during the estimation stage. Therefore, the plasma processing system using the temperature estimation device that performs each function will be described separately below.

[0025] (1) System structure of plasma processing system using temperature estimation device in model generation stage

[0026] Figure 1 This is a diagram illustrating an example of the system structure of a plasma processing system using an application temperature estimation device during the model generation phase. (See diagram for example.) Figure 1 As shown, in the model generation stage, the plasma processing system 100 includes a semiconductor manufacturing process, time series data acquisition devices 140_1 to 140_n, and a temperature estimation device 160.

[0027] In a semiconductor manufacturing process, a workpiece (pre-processing wafer 110) is subjected to plasma treatment in a chamber 120, which serves as a processing space for plasma treatment, to generate a result (post-processing wafer 130). Furthermore, the term "pre-processing wafer 110" refers to the wafer (substrate) before plasma treatment in the chamber 120, and "post-processing wafer 130" refers to the wafer (substrate) after plasma treatment in the chamber 120.

[0028] During the model generation stage, a temperature sensor is installed inside the chamber 120. Furthermore, the temperature sensor measures time-series data (temperature data for model generation) before or during plasma processing of the pre-processed wafer 110 in the chamber 120. Figure 1 The example shows the position of temperature sensor 151 mounted on the upper electrode inside chamber 120, and temperature sensor 152 mounted on the side wall inside chamber 120. Furthermore, the mounting positions of temperature sensors 151 and 152 within chamber 120 are just one example; they can also be mounted in... Figure 1 Different locations. Furthermore, the number of temperature sensors installed within chamber 120 is not limited to two.

[0029] The time series data (temperature data used for model generation) measured by temperature sensor 151 and temperature sensor 152 are stored in the model generation data storage unit 163 of the temperature estimation device 160.

[0030] Time series data acquisition devices 140_1 to 140_n respectively measure time series data sets (process data sets for model generation) before plasma processing of the pre-processed wafer 110 begins in chamber 120, or during plasma processing. Time series data acquisition devices 140_1 to 140_n are, for example, composed of multiple sensors (e.g., n sensors), and each time series data acquisition device measures process data for different types of measurement items. Furthermore, the number of measurement items measured by each of the time series data acquisition devices 140_1 to 140_n can be either one or multiple.

[0031] During the model generation stage, the time series data set (process data set for model generation) measured by the time series data acquisition devices 140_1 to 140_n and the time series data (temperature data for model generation) are stored in a corresponding manner in the model generation data storage unit 163.

[0032] A temperature estimation program is installed in the temperature estimation device 160. During the model generation stage, the temperature estimation device 160 functions as the model generation unit 161 by executing this program.

[0033] The model generation unit 161 uses the model generation data (process data set for model generation, temperature data for model generation) stored in the model generation data storage unit 163 to perform model generation processing on the time series model and provides the generated time series model.

[0034] (2) System structure of plasma treatment system using temperature estimation device in estimation stage

[0035] Figure 2 This is a diagram illustrating an example of the system structure of a plasma processing system using an application temperature estimation device during the estimation phase. Figure 1 The difference is that no temperature sensor is installed in chamber 120, and the temperature estimation device 160 performs different functions than in the model generation stage.

[0036] By performing a temperature estimation procedure during the estimation phase, the temperature estimation device 160 functions as an estimation unit 200, a determination unit 210, and a control unit 220.

[0037] The estimation unit 200 has a completed time series model provided by the model generation unit 161 during the model generation stage. The estimation unit 200 sequentially inputs data from each time period of the time series data set (the process data set for estimation) measured and stored in the time series data set storage unit 230 by the time series data acquisition devices 140_1 to 140_n to the completed time series model. Thus, the estimation unit 200 acquires estimated temperature data sequentially output from the completed time series model. Furthermore, the estimation unit 200 notifies either or both of the acquired estimated temperature data to the judgment unit 210 or the control unit 220.

[0038] The determination unit 210 continuously monitors the current estimated temperature data within the chamber 120, as notified by the estimation unit 200, and issues an alarm in real time if the temperature exceeds the specified range. Thus, the user of the plasma processing system 100 can monitor in real time whether the temperature within the chamber 120 during plasma processing exceeds the specified range.

[0039] Alternatively, the estimation unit 200 may notify estimated temperature data based on, for example, the time series data of plasma processing of a specific processed wafer 130 stored in the past time series data group (estimated process data group) in the time series data group storage unit 230. In this case, the determination unit 210 visualizes the time range during which plasma processing was performed outside the predetermined temperature range. As a result, the user of the plasma processing system 100 can trace and understand whether the temperature during plasma processing of a specific processed wafer 130 exceeded the predetermined temperature range.

[0040] The control unit 220 controls the semiconductor manufacturing process in real time (e.g., appropriately setting the power level of the RF power supply or the processing time of the plasma treatment) so that the current estimated temperature data in the chamber 120 notified by the estimation unit 200 reaches the target temperature. Thus, for example, it is possible to adjust the initial temperature when plasma treatment of the wafer 110 before processing is started to the target temperature.

[0041] Furthermore, the control unit 220 determines whether the process conditions are appropriate based on the estimated temperature data within the chamber 120 notified by the estimation unit 200 (e.g., the estimated temperature data during plasma processing of the previous pre-processing wafer 110). Moreover, the control unit 220 optimizes the process conditions for plasma processing of the next pre-processing wafer 110 (e.g., appropriately setting the RF power supply or the plasma processing time). Thus, for example, it is possible to keep the temperature during plasma processing of the next pre-processing wafer 110 within a predetermined temperature range.

[0042] <Hardware Structure of Temperature Estimation Device>

[0043] Next, the hardware structure of the temperature estimation device 160 will be described. Figure 3 This is a diagram illustrating an example of the hardware structure of a temperature estimation device. (See diagram for example.) Figure 3 As shown, the temperature estimation device 160 includes a CPU (Central Processing Unit) 301, a ROM (Read Only Memory) 302, and a RAM (Random Access Memory) 303. Additionally, the temperature estimation device 160 includes a GPU (Graphics Processing Unit) 304. Furthermore, the processors (processing circuits) such as the CPU 301 and GPU 304, and the memories such as the ROM 302 and RAM 303, form what is known as a computer.

[0044] Furthermore, the temperature estimation device 160 includes: an auxiliary storage device 305, a display device 306, an operation device 307, an I / F (Interface) device 308, and a driver device 309. In addition, the various hardware components of the temperature estimation device 160 are interconnected via a bus 310.

[0045] CPU 301 is a computing device that executes various programs (e.g., temperature estimation programs) installed in auxiliary storage device 305.

[0046] ROM 302 is a non-volatile memory and functions as the main storage device. ROM 302 stores various programs and data required by CPU 301 to execute various programs installed on auxiliary storage device 305. Specifically, ROM 302 stores boot programs such as BIOS (Basic Input / Output System) and EFI (Extensible Firmware Interface).

[0047] RAM303 is a volatile memory such as DRAM (Dynamic Random Access Memory) or SRAM (Static Random Access Memory), and functions as the main storage device. RAM303 provides a working area for various programs installed on the auxiliary storage device 305 to be executed by the CPU 301.

[0048] GPU 304 is a computing device for image processing. In this embodiment, when the temperature estimation program is executed by CPU 301, high-speed computation based on parallel processing is performed on the time series data set. In addition, GPU 304 is equipped with internal memory (GPU memory) to temporarily hold the information required for parallel processing of the time series data set.

[0049] The auxiliary storage device 305 stores various programs and various data used when executing these programs via the CPU 301. For example, the auxiliary storage device 305 implements a model generation data storage unit 163 and a time series data group storage unit 230.

[0050] Display device 306 is, for example, a display device that displays alarms to users of the plasma processing system 100. Operating device 307 is, for example, an input device used by users of the plasma processing system 100 or administrators of the temperature estimation device 160 when inputting various instructions to the temperature estimation device 160. I / F device 308 is a connection device for communicating with a network connection (not shown).

[0051] The drive device 309 is a device for setting the recording medium 320. The recording medium 320 here includes media that record information optically, electrically, or magnetically, such as CD-ROMs, floppy disks, and optical discs. Alternatively, the recording medium 320 may include semiconductor memories such as ROMs or flash memory that record information electrically.

[0052] Furthermore, for example, by placing the distributed recording medium 320 on the drive device 309, the drive device 309 can read the various programs recorded on the recording medium 320, thereby installing the various programs to be installed on the auxiliary storage device 305. Alternatively, the various programs to be installed on the auxiliary storage device 305 can also be downloaded via a network (not shown) for installation.

[0053] <Specific examples of data used for model generation>

[0054] Next, a specific example of the model generation data (process data set for model generation and temperature data for model generation) stored in the model generation data storage unit 163 will be explained. Figure 4 This is a diagram representing an example of the data used for model generation (the process data set used for model generation and the temperature data used for model generation). Furthermore, in Figure 4 In the example, for the sake of simplicity, it is assumed that the time series data acquisition devices 140_1 to 140_n measure one-dimensional data respectively, but a single time series data acquisition device can also measure two-dimensional data (a dataset of one-dimensional data of multiple types).

[0055] Figure 4 (a) schematically shows a time series data set (process data set for model generation) consisting of time series data measured within the same time range by time series data acquisition devices 140_1 to 140_n.

[0056] exist Figure 4 In (a), for example, time series data 1 schematically shows specified gas flow rate data (e.g., data measured by a sensor that measures the gas flow rate supplied to chamber 120). Furthermore, the term "gas" here includes all types of gases supplied to chamber 120 (process gases and inert gases). Additionally, the sensors for measuring gas flow rate include sensors installed in the gas lines for each type of gas (wherein, for convenience, only one time series data is shown here).

[0057] In addition, Figure 4 In (a), for example, time series data 2 schematically shows pressure data within chamber 120 (e.g., data measured by a sensor that measures the pressure within chamber 120). Additionally, for example, time series data 3 schematically shows RF power data (e.g., data measured by a sensor that measures the power of an RF power supply used for plasma processing).

[0058] In addition, Figure 4 In (a), for example, the time series data n-1 schematically shows impedance data (e.g., data measured by a sensor used to measure the chamber impedance (matching position, current, voltage, etc.)). Furthermore, for example, the time series data n schematically shows heater power data (e.g., data measured by a sensor measuring the amount of power of the heater heating the chamber 120).

[0059] on the other hand, Figure 4 (b) schematically shows time-series data (temperature data used for model generation) measured within a corresponding time range by a temperature sensor 151 mounted at the position of the upper electrode. Additionally, Figure 4 (b) schematically shows time-series data (temperature data for model generation) measured by a temperature sensor 152 located on the inner wall of chamber 120 within a corresponding time range.

[0060] <Relationship between process data sets and temperature data>

[0061] Next, the relationship between the time series data set (process data set for model generation) measured by the time series data acquisition devices 140_1 to 140_n and the time series data (temperature data for model generation) measured by the temperature sensors 151 and 152 will be explained.

[0062] Figure 5 This is a graph showing an example of measured values ​​for process data and temperature data. Among them, Figure 5 (a) shows the measured values ​​of the RF power supply data, with the horizontal axis representing time and the vertical axis representing power level. Additionally, Figure 5 (b) is the measured value of the temperature data obtained by the temperature sensor 151 installed at the position of the upper electrode. The horizontal axis represents time and the vertical axis represents temperature.

[0063] like Figure 5 As shown in (a) and (b), the temperature data within chamber 120 changes with a predetermined time delay relative to the changes in process data. This is because the heat-conducting materials constituting the components within chamber 120 require a certain amount of time for heat conduction. Additionally, a certain amount of time is required from the time heat is conducted to the temperature of each component within chamber 120 until the temperature of each component stabilizes.

[0064] In view of such temperature characteristics within the chamber 120, in the temperature estimation apparatus 160 of this embodiment, a model generation process is performed on a time series model that is capable of representing a time delay, and the temperature data is estimated using the provided generated time series model.

[0065] <Overview of the Model Generation Department>

[0066] Next, the general outline of the model generation unit 161 of the temperature estimation device 160 will be described. The model generation unit 161 has, for example, a time series model such as an ARX model. Moreover, the model generation unit 161 performs batch model generation processing on the time series model using model generation data (process data set for model generation, temperature data for model generation) stored in the model generation data storage unit 163.

[0067] Specifically, the model generation unit 161 reads the process data set for model generation from the model generation data storage unit 163 and inputs it sequentially into the time series model. In addition, the model generation unit 161 compares the output results sequentially output from the time series model with the corresponding temperature data for model generation (e.g., temperature data measured by the temperature sensor 151 installed at the position of the upper electrode) and calculates the error for each.

[0068] Furthermore, the model generation unit 161 determines the model parameters of the time series model so as to minimize the sum of squares of each calculated error.

[0069] Furthermore, the completed time series model that determines the model parameters can be represented, for example, by the following formula.

[0070] Formula 1

[0071]

[0072] In the above formula, k represents the current time, and y(k) represents the current temperature data. Furthermore, y(k-1)~y(k-m) a () represents the past time range m compared to the temperature data of the current output object. a Temperature data. Additionally, u(k-1)~u(k-m) b ) represents the past time range m compared to the current process data set. b The process data set. Furthermore, a1~a ma b1~b mb This indicates the determined model parameters.

[0073] As can be seen from the above formula, in time series models, the determined model parameters are used to define the past time range m. b The process data set and the previously specified time range m a The temperature data are weighted and summed to calculate the current temperature.

[0074] Therefore, based on the temperature estimation device 160, even if the state within the chamber 120 changes (has changed), the temperature data corresponding to the changes in process data can be estimated using the generated time series model.

[0075] Furthermore, the case where the model parameters of the time series model are determined based on temperature data measured by temperature sensor 151 installed at the position of the upper electrode is described here. However, the model parameters of the time series model can also be determined by performing the same processing on temperature data measured by temperature sensor 152 installed at the position of the side wall inside the chamber 120.

[0076] However, since the temperature data measured by temperature sensor 151 and the temperature data measured by temperature sensor 152 are different model parameters, each is managed as a separately generated time series model in the model generation unit 161.

[0077] Furthermore, the above explanation did not mention the specified time range m. a m b The method for selecting the order is assumed to be appropriately adjusted and optimized. Similarly, the method for selecting the process data set input to the time series model in the process data set used for model generation is not mentioned in the above description, but it is assumed to be appropriately adjusted and optimized.

[0078] <Model Generation Process>

[0079] Next, the process of model generation and processing of the plasma processing system in the model generation stage will be explained. Figure 6 It is a flowchart representing the process of model generation and processing.

[0080] In step S601, the temperature estimation device 160 acquires data for model generation and data for verification. Specifically, the temperature estimation device 160 acquires process data sets and temperature data from time series data acquisition devices 140_1 to 140_n and temperature sensor 151 (or 152) as data for model generation. Similarly, the temperature estimation device 160 acquires process data sets and temperature data from time series data acquisition devices 140_1 to 140_n and temperature sensor 151 (or 152) as data for verification.

[0081] In step S602, the temperature estimation device 160 selects the process data set required for estimating the temperature data from the process data set used for model generation.

[0082] In step S603, the temperature estimation device 160 selects the order of the time series model of the model generation unit 161.

[0083] In step S604, the temperature estimation device 160 performs model generation processing on the time series model and determines the model parameters of the time series model so as to minimize the sum of squares of errors when the selected process data set is input.

[0084] In step S605, the temperature estimation device 160 uses verification data to verify the time series model (the generated time series model) whose model parameters have been determined.

[0085] The verification result determines whether the target estimation accuracy has been achieved. If the target estimation accuracy has not been achieved (in step S605, if "No"), the process returns to step S602. In this case, in the temperature estimation device 160, in step S602, in addition to the existing selected process data set, process data that helps improve the estimation accuracy is added, or a process data set with a different combination than the previous combination is selected.

[0086] On the other hand, if it is determined in step S605 that the estimated accuracy of the target has been achieved (in step 605, if "yes"), the model generation process ends.

[0087] <The functional structure of the presumption section and specific examples of its processing>

[0088] Next, the functional structure of the estimation unit 200 and specific examples of the processing performed by the estimation unit 200 will be explained. Figure 7 This is a diagram illustrating the functional structure and specific examples of the presumption section.

[0089] like Figure 7 As shown, the estimation unit 200 has a completed time series model 710 provided by the model generation unit 161.

[0090] The estimation unit 200 sequentially inputs the time series data sets (process data sets for estimation) acquired by the time series data acquisition devices 140_1 to 140_n into the generated time series model 710. Furthermore, the time series data sets (process data sets for estimation) input at this time are the process data sets (process data sets related to the state of chamber 120) selected during the model generation process when they are deemed to have the prescribed estimation accuracy.

[0091] Furthermore, in the estimation unit 200, before inputting the selected process data set, a temperature initialization sequence is executed, for example, as described below.

[0092] • Used for processing that ensures sufficient temperature stability within chamber 120 for a sustained period of time, during which process data sets (selected process data sets) and temperature data are collected for a period of time after the model parameters are determined.

[0093] • Input the collected process data set and temperature data as initial values ​​into the generated time series model 710.

[0094] Figure 7 The example shows the current process data set as a time series data set of the time indicated by reference numeral 721, and what it looks like when the process data set is entered.

[0095] In addition, such as Figure 7 As shown, in the completed time series model 710, the calculation of the estimated temperature data uses a past time range m compared to the current process data set. b The process data set. Additionally, in the completed time series model 710, the calculation of the current estimated temperature data uses a past time range m compared to the current estimated temperature data (estimated temperature data at the time indicated by reference numeral 721). a Estimated temperature data.

[0096] That is, in the completed time series model 710, the estimated temperature data of the current (reference numeral 721) estimated object is calculated based on the following data:

[0097] • Compared to the current (see attached figure 721) process data set, the past time range m bThe process data set (process data sets within a specified time range compared to the current process data set); and

[0098] • Compared with the current (see attached figure 721) estimated temperature data, the past time range m a Estimated temperature data (estimated temperature data for a specified time range previously estimated compared to the estimated temperature data of the estimated object).

[0099] In addition, once the time series model 710 is generated, it will sequentially notify the determination unit 210 and the control unit 220 of the estimated temperature data of the current (reference numeral 721) estimated object.

[0100] In this way, the estimation unit 200 sequentially inputs the process data set of the new time series related to the state inside the chamber 120 into the generated time series model 710, and then outputs the estimated temperature data in sequence.

[0101] In addition, Figure 7 The process describes the case of estimating the position of the upper electrode using estimated temperature data. However, when using the time series model 710 for generating estimated temperature data for estimating the position of the sidewall inside the chamber 120, the estimated temperature data can be estimated sequentially.

[0102] <The functional structure of the decision-making unit and specific examples of its processing>

[0103] Next, the functional configuration of the determination unit 210 and specific examples of the processing performed by the determination unit 210 will be explained. Figure 8 This is a diagram illustrating the functional structure and specific examples of the processing of the decision-making unit.

[0104] like Figure 8 As shown, the determination unit 210 includes a separation amount calculation unit 810, a determination criterion setting unit 820, and a reporting unit 830.

[0105] The separation amount calculation unit 810 acquires, for example, real-time estimated temperature data notified by the estimation unit 200 at a predetermined period. In addition, whenever the separation amount calculation unit 810 acquires real-time estimated temperature data, it determines whether the estimated temperature data exceeds the temperature range preset by the determination reference setting unit 820.

[0106] Furthermore, if the detachment amount calculation unit 810 determines that the estimated temperature data exceeds a preset temperature range, it calculates the detachment amount and notifies the reporting unit 830. Figure 8In the figure, Figure 840 shows the real-time estimated temperature data notified by the estimation unit 200 to the separation amount calculation unit 810. In addition, Figure 840 shows the estimated temperature data when multiple pre-processing wafers are continuously subjected to plasma processing, with the horizontal axis representing the processing time of each pre-processing wafer and the vertical axis representing the estimated temperature data.

[0107] In addition, Figure 8 In Figure 860, the time range 841 of Figure 840 is enlarged and shown. In Figure 860, the dashed line 862 shows the upper limit of the temperature range preset by the determination reference setting unit 820, and the dashed line 863 shows the lower limit of the temperature range preset by the determination reference setting unit 820.

[0108] According to Figure 860, since the estimated temperature data 861 exceeds the specified temperature range (in... Figure 8 In the example, since the temperature is below the lower limit of the dotted line 863, the detachment amount calculation unit 810 determines that the estimated temperature data 861 exceeds the preset temperature range and calculates the detachment amount. In this case, the reporting unit 830 reports an alarm indicating that the temperature exceeds the specified range to the user of the plasma processing system 100.

[0109] On the other hand, Figure 8 In the figure, graph 850 enlarges and shows the time range 842 of graph 840. In graph 850, dashed line 852 shows the upper limit of the temperature range preset by the determination reference setting unit 820, and dashed line 853 shows the lower limit of the temperature range preset by the determination reference setting unit 820.

[0110] According to Figure 850, since the estimated temperature data 851 converges within the specified temperature range, the separation amount calculation unit 810 determines that the estimated temperature data 851 does not exceed the preset temperature range. In this case, the reporting unit 830 does not issue an alarm to the user of the plasma processing system 100.

[0111] Furthermore, the above description describes the case where the determination unit 210 performs real-time determination processing on the estimated temperature data notified by the estimation unit 200. However, the same applies to the case where the determination processing is performed on the estimated temperature data notified by the estimation unit 200 based on the past time series data stored in the model generation data storage unit 163.

[0112] However, in this case, the reporting unit 830 reports to the user of the plasma processing system 100 by visualizing the time range of plasma processing outside the specified temperature range.

[0113] <Functional structure of the control unit and specific examples of its processing>

[0114] Next, the functional configuration of the control unit 220 and specific examples of the processing performed by the control unit 220 will be explained. Figure 9 This is a diagram illustrating the functional structure and specific examples of the processing of the control unit.

[0115] like Figure 9 As shown, the control unit 220 includes a temperature adjustment unit 910, a target temperature setting unit 920, a process adjustment unit 930, etc.

[0116] The temperature adjustment unit 910 generates control data, for example, based on real-time estimated temperature data notified by the estimation unit 200 during the aging process, so that the current estimated temperature data in the chamber 120 reaches the target temperature. Furthermore, the target temperature is set, for example, by the target temperature setting unit 920.

[0117] In addition, the temperature adjustment unit 910 adjusts the temperature inside the chamber 120 by sending the generated control data to the semiconductor manufacturing process.

[0118] Therefore, conventionally, even if the temperature at the start of plasma processing drops to the temperature shown by reference numeral 941, it is not possible to accurately estimate the estimated temperature data at that moment, and thus plasma processing is started as is. In contrast, since the current estimated temperature data is notified in real time by the estimation unit 200, adjustments can be made in the temperature adjustment unit 910, for example, during the aging process, so that the temperature inside the chamber 120 reaches the target temperature shown by reference numeral 942. As a result, for the first pre-processed wafer, the temperature change during plasma processing becomes the same as the temperature change for the second and subsequent pre-processed wafers, and the quality of the first post-processed wafer after plasma processing can also be maintained.

[0119] The process adjustment unit 930 determines whether the process conditions are appropriate, for example, based on the estimated temperature data output during plasma processing (e.g., the estimated temperature data output during plasma processing of the previous pre-processed wafer 110). Furthermore, based on the determination of whether the process conditions are appropriate, the process adjustment unit 930 optimizes the process conditions for plasma processing of the next pre-processed wafer 110 and adjusts the recipe. Moreover, the process adjustment unit 930 performs plasma processing using the adjusted recipe by sending it to the semiconductor manufacturing process.

[0120] Therefore, for example, even if the temperature inside the chamber 120 changes due to continuous plasma processing, the same process can be used in the past. However, according to the process adjustment unit 930, a process corresponding to the temperature change can be used (refer to reference numerals 951 and 952). As a result, the temperature when plasma processing is performed on the wafer 110 before the next processing can be kept within a specified temperature range.

[0121] In addition, Figure 9 In the example, the functions of adjusting the temperature in the chamber 120 during aging and adjusting the process during plasma processing are illustrated, but the functions of the control unit 220 in adjusting based on the estimated temperature data are not limited to these.

[0122] <Presumed processing flow>

[0123] Next, the process of the estimation process in the estimation stage of the plasma processing system will be explained. Figure 10 This is a flowchart representing the presumed processing flow.

[0124] In step S1001, the time series data acquisition devices 140_1 to 140_n begin to measure the time series data set (estimated process data set) and store the measured time series data set in the time series data set storage unit 230.

[0125] In step S1002, the temperature estimation device 160 determines whether to use the time series data set to perform real-time processing. Furthermore, real-time processing refers to the process of inputting the time series data set (the estimation process data set) before or during plasma processing into the generated time series model, thereby outputting the estimated temperature data in real time.

[0126] If it is determined in step S1002 that real-time processing is to be performed (if "yes" is in step S1002), proceed to step S1011.

[0127] In step S1011, the temperature estimation device 160 inputs a selected set of process data (a set of process data related to the state of the chamber) from the measured time series data set (the process data set for estimation) into the generated time series model. Furthermore, the temperature estimation device 160 inputs initial values ​​of the temperature data into the generated time series model at the start of the estimation process. Thus, the temperature estimation device 160 can output real-time estimated temperature data.

[0128] In step S1012, the temperature estimation device 160 determines whether to execute the determination mode or the control mode. If it is determined in step S1012 that the determination mode is to be executed, the process proceeds to step S1013.

[0129] In step S1013, the temperature estimation device 160 performs real-time judgment processing and issues an alarm if the current estimated temperature data exceeds the preset temperature range.

[0130] On the other hand, if the execution control mode is determined in step S1012, proceed to step S1014.

[0131] In step S1014, the temperature estimation device 160 performs a temperature adjustment process to adjust the temperature within the chamber so that the current estimated temperature data reaches the target temperature. Alternatively, the temperature estimation device 160 determines whether the process conditions for plasma treatment of the previous pre-processing wafer are appropriate. Furthermore, the temperature estimation device 160 optimizes the process conditions for plasma treatment of the next pre-processing wafer and adjusts the process accordingly.

[0132] On the other hand, if it is determined in step S1002 that no real-time processing has been performed (in step S1002, if "No"), proceed to step S1021.

[0133] In step S1021, the temperature estimation device 160 reads the following data from the time series data group (process data group for estimation) stored in the time series data group storage unit 230:

[0134] For example, time-series data sets during plasma processing of a specific processed wafer 130.

[0135] • Selected process data set (a time series of process data related to the state of the chamber). Additionally, the temperature estimation device 160 inputs the read process data set into the generated time series model and outputs estimated temperature data.

[0136] In step S1022, the temperature estimation device 160 performs a determination process, visualizes the time range of plasma processing that exceeds the specified temperature range from the estimated temperature data, and reports it to the user of the plasma processing system 100.

[0137] In step S1030, the temperature estimation device 160 determines whether to end the estimation process. If it is determined in step S1030 that the estimation process should continue (if "No" is determined in step S1030), the process returns to step S1002.

[0138] On the other hand, if it is determined in step S1030 that the estimation process is to end (in step S1030, if "yes"), the estimation process is to end.

[0139] <Summary>

[0140] As can be seen from the above description, for the temperature estimation device 160 involved in the first embodiment,

[0141] • A model generation process is performed on the time series model, which establishes a correlation between the time series process data set related to the state within the plasma processing chamber and the temperature data measured within the chamber at each time point. This provides the generated time series model.

[0142] • Temperature data can be estimated by sequentially inputting new time series process data sets related to the state within the chamber into the provided generated time series model.

[0143] Thus, in the first embodiment, considering the semiconductor manufacturing process's unique characteristic of making it difficult to directly measure temperature by constantly installing temperature sensors within the chamber, the temperature within the chamber is indirectly estimated based on a time-series process data set. In this first embodiment, the time-series temperature data is estimated by using data from each time range of a time-series process data set related to the state of the chamber.

[0144] Therefore, according to the first embodiment, even if the state inside the chamber changes (has changed), it is possible to estimate the temperature data corresponding to the change in process data.

[0145] That is, according to the first embodiment, a temperature estimation device, a plasma processing system, a temperature estimation method, and a temperature estimation procedure are provided to estimate the temperature corresponding to the change in state within the processing space where plasma processing is performed.

[0146] [Second Implementation]

[0147] In the first embodiment described above, the model generation unit 161 is described as performing model generation processing on an ARX model. However, the time series model that the model generation unit 161 performs model generation processing on is not limited to an ARX model. For example, any model capable of calculating the correlation of time series data can be used, such as other models (e.g., RNN (Recurrent Neural Network) models, etc.).

[0148] Furthermore, in the case of an RNN model, the model generation unit 161 determines the model parameters through learning processing and provides a learned RNN model. Specifically, in the case of an RNN model, the model generation unit 161 uses a dataset that takes process data for model generation as input data and temperature data for model generation as correct data for learning processing to determine the model parameters.

[0149] That is, the model generation unit 161 functions as a learning unit, performing learning processing through error backpropagation algorithms, etc., so that the output data when the input data is input into the RNN model is close to the correct data, thereby providing a learned RNN model.

[0150] Furthermore, in the first embodiment described above, the process adjustment unit 930 was described as generating the adjusted process based on estimated temperature data. However, the process adjustment unit 930 may also, for example, analyze dynamic events within the chamber 120 based on estimated temperature data and calculate characteristic values ​​for adjusting the process. These characteristic values ​​may include, for example, C / D (Critical Dimension) values, E / R (Etching Rate) values, etc.

[0151] Furthermore, in the first embodiment described above, the temperature adjustment unit 910 adjusts the temperature within the chamber 120 during the aging process; however, the temperature adjustment unit 910's adjustment of the temperature within the chamber 120 is not limited to the aging process. For example, it can also be performed during plasma treatment of the wafer before processing. Plasma treatment here includes etching, film formation, ashing, and the like.

[0152] In addition, in the first embodiment described above, the semiconductor manufacturing process and the temperature estimation device 160 are separately configured, but at least a portion of the functions of the temperature estimation device 160 may also be implemented in the semiconductor manufacturing process.

[0153] Furthermore, in the first embodiment described above, it was explained that the model generation stage and the estimation stage are performed by the same temperature estimation device 160, but the temperature estimation device for performing the model generation stage and the temperature estimation device for performing the estimation stage may also be configured separately.

[0154] Furthermore, in the first embodiment described above, the temperature estimation device 160 for performing the estimation phase was described as functioning as an estimation unit 200, a determination unit 210, and a control unit 220. However, the temperature estimation device 160 for performing the estimation phase may also be configured separately as an estimation unit 200, a determination unit 210, and a control unit 220. That is, it may also be configured separately as a temperature estimation device with an estimation unit 200, an alarm output device with a determination unit 210, and a temperature control device with a control unit 220.

[0155] Furthermore, the present invention is not limited to combinations of other elements and structures listed in the above embodiments, etc., as shown herein. These aspects can be modified without departing from the spirit of the invention, and can be appropriately determined according to its application.

Claims

1. A temperature estimation device, The estimation unit has a process data estimation section that sequentially estimates temperature data by inputting new time series process data related to the state within the processing space into the generated time series model. The aforementioned generated time series model is based on the error between the temperature data output sequentially from the time series model by inputting process data related to the state within the aforementioned processing space and the temperature data of each time period measured within the aforementioned processing space. The parameters of the time series model are determined by this error.

2. The temperature estimation device according to claim 1, wherein, The new time-series process data related to the state within the aforementioned processing space are multiple time-series process data newly measured in each of the multiple sensors.

3. The temperature estimation device according to claim 2, wherein, The aforementioned multiple sensors include at least one of the following sensors: a sensor for measuring the pressure within the processing space, a sensor for measuring the gas flow rate supplied to the processing space, a sensor for measuring the power of the RF power supply used for plasma processing, a sensor for measuring the chamber impedance, and a sensor for measuring the power of the heater used to heat the processing space.

4. The temperature estimation device according to claim 2, wherein, The new time-series process data related to the state within the aforementioned processing space are multiple time-series process data newly measured before the start of plasma processing or during the plasma processing process.

5. The temperature estimation device according to claim 2, wherein, The aforementioned completed time series model uses the parameters of the completed time series model to perform a weighted summation of the process data within a specified time range compared to the current process data and the temperature data within a specified time range estimated compared to the temperature data of the estimated object, thereby outputting the temperature data of the estimated object.

6. The temperature estimation device according to claim 1, wherein, The aforementioned generated time series model is generated by determining the parameters of the time series model based on the error between the temperature data of the time series measured at a predetermined location within the aforementioned processing space and the measured location.

7. A plasma processing system, comprising: The temperature estimation device according to any one of claims 1 to 6; and The alarm output device outputs an alarm when the temperature data estimated sequentially by the above-mentioned estimation unit exceeds a predetermined temperature range.

8. The plasma processing system according to claim 7, wherein, The plasma processing system described above also includes a control device that sets the processing time of the plasma processing so that the temperature data sequentially estimated by the estimation unit before or during the plasma processing reaches the target temperature.

9. The plasma processing system according to claim 7, wherein, The plasma processing system described above also includes a control device that sets the power of the RF power supply so that the temperature data sequentially estimated by the estimation unit before or during the plasma processing reaches the target temperature.

10. A method for estimating temperature, It includes a process of sequentially estimating temperature data by inputting new time-series process data related to the state within the processing space into a generated time-series model, wherein... The aforementioned generated time series model is based on the error between the temperature data output sequentially from the time series model by inputting process data related to the state within the aforementioned processing space and the temperature data of each time period measured within the aforementioned processing space. The parameters of the time series model are determined by this error.

11. A temperature estimation program product, This is used to enable a computer to execute the following process: estimating temperature data by sequentially inputting new time-series process data related to the state within the processing space into a generated time-series model, wherein... The aforementioned generated time series model is based on the error between the temperature data output sequentially from the time series model by inputting process data related to the state within the aforementioned processing space and the temperature data of each time period measured within the aforementioned processing space. The parameters of the time series model are determined by this error.

Images