Diagnostic device, diagnostic method, and semiconductor device manufacturing system

Abstract

The purpose of the present invention is to provide a technique capable of grasping the surface state of a sample stage and diagnosing the occurrence of an abnormality.　In order to provide the technique, the diagnostic device according to the present invention which diagnoses the presence or absence of an abnormality of a sample stage on which a sample is placed is characterized in that: the presence or absence of the abnormality of the sample stage is determined on the basis of a time constant in a transient change in the current when static elimination processing is performed; the static elimination processing is processing for detaching, from the sample stage, the sample electrostatically adsorbed on the sample stage; and the current is a current flowing through the sample stage when a voltage for electrostatically adsorbing the sample onto the sample stage is applied to the sample stage.

Description

Diagnostic apparatus, diagnostic method, and semiconductor device manufacturing system

[0001] The present invention relates to a diagnostic apparatus (PHM: Prognostics and Health Management) that uses time-series signals (sensor waveform data) sequentially acquired from a plurality of sensors in a plasma processing apparatus, a diagnostic method, and a semiconductor device manufacturing system.

[0002] The sample stage installed in a plasma processing apparatus is equipped with an electrostatic chuck (ESC) on the surface of the sample stage, which generates an electrostatic attraction force between the sample stage and the wafer during plasma processing. The surface condition of the sample stage (hereinafter also referred to as the "ESC surface condition") deteriorates over time due to damage to the sample stage surface, deposit adhesion, and other factors. This can lead to abnormalities in wafer processing speed and wafer adhesion. Therefore, there is a strong demand for technology that can detect changes in the ESC surface condition and enable maintenance before an abnormality occurs. However, there are no practical sensors for directly detecting the ESC surface condition, making it practically difficult to monitor the ESC surface condition in real time during operation of a plasma processing apparatus.

[0003] Residual charge can be considered as one indicator of the ESC surface condition. Patent Document 1, for example, addresses the issue of dechucking a workpiece from an electrostatic chuck. It discloses the following technology related to residual charge: A dechucking control method and a control device for a plasma processing apparatus are disclosed: "A dechucking control method for dechucking a workpiece from an electrostatic chuck 40 having a chuck electrode 40a that electrostatically attracts the workpiece, the method comprising: acquiring a time integral of a current obtained by measuring a current flowing from the chuck electrode for a predetermined period of time after turning off a voltage applied to the chuck electrode after plasma processing; calculating a difference between the amount of charge charged in the chuck electrode during plasma processing and the acquired time integral of the current; calculating a counter voltage corresponding to the amount of residual charge on the electrostatic chuck from the difference based on a predetermined correlation between the time integral of the current and a torque applied to support pins that support the workpiece; and applying the counter voltage to the chuck electrode while introducing gas into a processing chamber to generate plasma."

[0004] JP 2013-161899 A

[0005] If the ESC surface condition changes, for example, the amount of residual charge after a static elimination process that removes charge from the sample stage surface and sample may change, potentially resulting in abnormalities in the residual attraction force. However, plasma processing equipment generally does not measure the residual charge itself, making it difficult to directly use the residual charge as an indicator of the ESC surface condition. Patent Document 1 also proposes a technology that calculates the amount of charge accumulated on the ESC (referred to as the "electrode" in Patent Document 1) using the time integral of current to determine whether residual charge remains during static elimination. However, depending on the ESC configuration, leakage current components may be present, which may make it difficult to accurately calculate the amount of charge using the current time integral. Therefore, the present invention aims to provide a technology that can grasp the surface condition of the sample stage and diagnose any abnormalities.

[0006] In order to solve the above-mentioned problems, one representative diagnostic device of the present invention is a diagnostic device that diagnoses whether or not there is an abnormality in a sample stage on which a sample is placed, and is characterized in that the presence or absence of an abnormality in the sample stage is determined based on the time constant in the transient change in current when a discharge process is performed, the discharge process is a process of detaching the sample electrostatically adsorbed to the sample stage from the sample stage, and the current is a current that flows through the sample stage when a voltage for electrostatically adsorbing the sample to the sample stage is applied to the sample stage.

[0007] According to the present disclosure, it is possible to grasp the surface condition of the sample stage and diagnose the occurrence of abnormalities. Problems, configurations, and effects other than those described above will become clear from the description of the following embodiments of the present invention.

[0008] FIG. 1 is a diagram illustrating an example of the configuration of a diagnostic device. FIG. 2 is a diagram illustrating an example of the configuration of a chamber and an ESC in a plasma processing apparatus. FIG. 3 is an enlarged schematic diagram illustrating a contact portion between a first ESC unit 108 and a wafer 104. FIG. 4 is a diagram illustrating a data analysis model within the plasma processing apparatus. FIG. 5 is a diagram illustrating an example of sensor data. FIG. 6 is a diagram illustrating an example of a graph of sensor data. FIG. 7 is a flowchart illustrating a feature calculation process. FIG. 8 is a diagram illustrating a specific example of calculation of a feature value F1 during a charge removal process and an example of a change in the feature value F1 over time. FIG. 9 is a diagram illustrating a change in the feature value over time. FIG. 10 is a diagram illustrating an example of a GUI (Graphical User Interface) screen displaying a determination result. FIG. 11 is a flowchart illustrating a feature calculation process. FIG. 12 is a diagram illustrating a specific example of calculation of a feature value F3 during a suction process and an example of a change in the feature value F3 over time. FIG. 13 is a flowchart illustrating a feature calculation process.

[0009] Hereinafter, embodiments of the present invention will be described with reference to the drawings. Note that the present invention is not limited to these embodiments. In addition, in the drawings, identical parts are denoted by the same reference numerals. For example, while a diagnostic device for a plasma processing apparatus is described in this disclosure, the diagnostic device may be configured with general hardware including a processor and memory, and the diagnostic function may be performed by the processor executing a program stored in the memory, or the diagnostic device may be configured with dedicated hardware. Furthermore, the diagnostic device may be implemented by incorporating dedicated hardware into a computer and combining software and hardware implementation. Furthermore, the diagnostic device may be a standalone device or may be externally connected. Furthermore, the diagnostic device may be configured as a module that also functions for other data processing. Hereinafter, embodiments will be described with reference to the drawings.

[0010] (Configuration of Diagnostic Apparatus) The configuration of the diagnostic apparatus will be described with reference to FIG. 1. FIG. 1 is a diagram showing an example of the configuration of the diagnostic apparatus. As shown in FIG. 1, the diagnostic apparatus 20 includes a data collection unit 22, a data extraction processing unit 24, a feature calculation unit 26, and an abnormality detection unit 28. The diagnostic apparatus 20 is connected to the plasma processing apparatus 10, and determines abnormalities based on information acquired from the plasma processing apparatus. A system including the diagnostic apparatus 20 and the plasma processing apparatus 10 is also referred to as a semiconductor device manufacturing system (diagnostic system) 1. Note that the diagnostic apparatus 20 may be connected to a single plasma processing apparatus 10, or multiple plasma processing apparatuses may be connected. The multiple plasma processing apparatuses may be distinguished from each other by, for example, assigning apparatus IDs (identifications).

[0011] The diagnostic device 20 performs a process to diagnose whether or not there is an abnormality in a sample stage on which a sample is placed. The diagnostic device 20 determines whether or not there is an abnormality in the sample stage based on a time constant in a transient change in current when a static elimination process is performed. The static elimination process is a process of detaching the sample electrostatically attracted to the sample stage from the sample stage, and the current is a current that flows through the sample stage when a voltage for electrostatically attracting the sample to the sample stage is applied to the sample stage. The diagnostic device 20 also determines whether or not there is an abnormality in the sample stage based on a time constant in a transient change in current when the sample is electrostatically attracted to the sample stage. The current is a current that flows through the sample stage when a voltage for electrostatically attracting the sample to the sample stage is applied to the sample stage. The diagnostic device 20 is configured to include a GUI that displays the feature amounts, the changes in the feature amounts over time, and the results of the determination. The diagnostic device 20 also includes a data collection unit 22 that collects time-series data of a current flowing through the electrostatic chuck while a process including a discharge treatment is being performed, a data extraction processing unit 24 that extracts data for a first predetermined period corresponding to the discharge treatment from the time-series data of the current, a feature calculation unit 26 that calculates a time constant of a transient change in the data for the first predetermined period as a feature, and an abnormality detection unit 28 that determines that an abnormality exists in the sample stage when the feature exceeds a threshold value.

[0012] Specifically, the diagnostic device 20 receives time-series data (sensor data D) measured by a predetermined sensor during the process performed by the plasma processing device 10. ０ ) and outputs the analysis results. The sensor that performs the measurement may be a single sensor or multiple sensors. The plasma processing apparatus 10 processes a wafer in a chamber under set process conditions, and outputs the sensor data D ０ The data collection unit 22 transmits the sensor data D ０ Among these, sensor data D including ESC voltage and ESC current, which will be described later, １ and transmits it to the data extraction processing unit 24. The data extraction processing unit 24 １The part necessary for analysis is extracted from the sensor data D and transmitted to the feature calculation unit 26. As a method of extraction, when extracting a necessary item from a plurality of data, a period necessary for analysis may be extracted from time-series data. The feature calculation unit 26 calculates the extracted sensor data D ２ The time constant of the transient change in the ESC current is calculated as a feature value from the above. The abnormality detection unit 28 analyzes the change over time in the calculated feature value and outputs a determination result as to whether or not an abnormality has occurred.

[0013] Although an example of the configuration of the semiconductor device manufacturing system 1 has been described, the present disclosure is not limited to this example. For example, the semiconductor device manufacturing system 1 may include a platform on which an application is implemented that diagnoses the presence or absence of an abnormality in a sample stage provided in a semiconductor manufacturing apparatus and on which a sample is placed. The application may execute a step of determining the presence or absence of an abnormality in the sample stage based on a time constant in a transient change in current when a static elimination process is performed, and a step of determining the presence or absence of an abnormality in the sample stage based on a time constant in a transient change in current when the sample is electrostatically attracted to the sample stage. Furthermore, the platform may be configured, for example, by a server.

[0014] (Configuration of Plasma Processing Apparatus) Next, a plasma processing apparatus 10 will be described with reference to FIG. 2 . FIG. 2 is a diagram illustrating an example of the configuration of the chamber 100 and the ESC of the plasma processing apparatus 10. The plasma processing apparatus 10 includes a chamber 100 into which a wafer (hereinafter also referred to as a “sample”) 104 is introduced and subjected to processes such as etching. In the chamber 100, the wafer 104 is placed on a sample stage 106. The sample stage 106 includes a first ESC unit (hereinafter also referred to as a “first electrostatic check unit”) 106 and a second ESC unit (hereinafter also referred to as a “second electrostatic check unit”) 110 disposed on a substrate 112. The first electrostatic chuck unit 108 and the second electrostatic chuck unit 110 are disposed on the surface of the sample stage 106 and generate an electrostatic adsorption force between the wafer 104 and the sample stage 106. In the following description, a separate type in which two ESC units, the first ESC unit 108 and the second ESC unit 110, are arranged on the surface of the sample stage 106 will be described, but the present disclosure is not limited to this case. The present disclosure can also be applied to ESCs with other configurations. Furthermore, when it is not necessary to specify either the first ESC unit 108 or the second ESC unit 110, they may simply be referred to as "ESC units."

[0015] As will be described later, the first ESC unit 108 is composed of a dielectric and an electrode unit disposed within the dielectric. A voltage V1 is applied to the electrode unit from a first power supply 116. A first ammeter 114 is disposed between the first power supply 116 and the first ESC unit 108, and measures the current flowing through the first electrostatic chuck unit 108.

[0016] The second ESC unit 110 is also configured with a dielectric and an electrode unit disposed within the dielectric, similar to the first electrostatic chuck unit 108. A voltage V2 is applied to the electrode unit from a second power supply 120. A second ammeter 118 is disposed between the second power supply 120 and the second electrostatic chuck unit 110, and measures the current flowing through the second ESC unit 110.

[0017] Next, the operation of the sample stage will be described. When processing a wafer 104 in the plasma processing apparatus 10, it is necessary to generate an electrostatic force between the sample stage 106 and the wafer 104 to fix the wafer 104 to the sample stage 106. Electrostatic forces are generated by applying a voltage V1 to the first ESC unit 108 and a voltage V2 to the second ESC unit 110, which charges the first ESC unit 108, the second ESC unit 110, and the wafer 104, thereby generating electrostatic forces between the first ESC unit 108 and the wafer 104 and between the second ESC unit 110 and the wafer 104.

[0018] Generally, the processes performed in the plasma processing apparatus 10 include a loading process in which the wafer 104 is transferred onto the sample stage 106, a chucking process in which the first power supply 116 and the second power supply 120 are turned on (voltage is applied) to chucking the wafer 104 onto the first ESC unit 108 and the second ESC unit 110, a processing process in which the plasma 102 is generated to etch the wafer 104, a discharge process in which the first power supply 116 and the second power supply 120 are turned off (voltage is not applied) to discharge the wafer 104 and the sample stage 106 after processing is completed, and a unloading process in which the wafer 104 is transferred after discharge. Sensor data acquired during this series of processes includes time series data of the voltage V1 applied to the first ESC unit 108, time series data of the current measured by the first ammeter 114, time series data of the voltage V2 applied to the second ESC unit 110, and time series data of the current measured by the second ammeter 118. The current measured by the first ammeter 114 is also referred to as the first ESC current, and the current measured by the second ammeter 118 is also referred to as the second ESC current. When no distinction is made between these currents, they may simply be referred to as the ESC current. Furthermore, the voltage applied to the first ESC unit 108 is also referred to as the first ESC voltage, and the voltage applied to the second ESC unit 110 is also referred to as the second ESC voltage. When no distinction is made between these voltages, they may simply be referred to as the ESC voltage. These voltages are acquired by predetermined sensors provided in the plasma processing apparatus 10.

[0019] (ESC Surface State) Next, the ESC surface state will be described with reference to FIGS. 3 and 4. FIG. 3 is an enlarged schematic diagram showing the contact portion between the first ESC unit 108 and the wafer 104. The surface of the first ESC unit 108 (the surface facing the positive z-axis direction, hereinafter simply referred to as the "ESC surface") and the back surface of the wafer 104 (the surface facing the negative z-axis direction) are not ideally flat and have roughness. Therefore, during adsorption processing, it is assumed that there is a first contact portion 1080 between the first ESC unit 108 and the wafer 104 where they come into contact, and a first gap portion 1086 where they do not come into contact and form a space. When processing is being performed in the plasma processing apparatus 10, He gas is supplied to the first gap portion 1086 to cool the wafer 104.

[0020] Although the relationship between the first ESC unit 108 and the wafer 104 has been described, a similar relationship also exists between the second ESC unit 110 and the wafer 104 .

[0021] 4A and 4B are diagrams showing a model for analyzing data within the plasma processing apparatus 10. Fig. 4A shows an analytical model that schematically represents the structure of the plasma processing apparatus 10, and Fig. 4B shows the first ESC unit 108 portion of the analytical model as an equivalent circuit.

[0022] 4A, from the positive z-axis direction, the plasma 102, the wafer 104, the first ESC unit 108 or the second ESC unit 110, and the first power supply 116 or the second power supply 120 are arranged. The plasma 102 is set to the ground potential, and the first power supply 116 and the second power supply 120 are also set to the ground potential.

[0023] The first ESC unit 108 includes a first contact portion 1080, a first bulk portion 1082, a first electrode portion 1084, and a first gap portion 1086. The first contact portion 1080 is the portion of the first ESC unit 108 that contacts the wafer 104. The first bulk portion 1082 is a portion that constitutes the first ESC unit 108 and is made of, for example, a dielectric material. The first electrode portion 1084 is an electrode to which a voltage is supplied from the first power supply 116. The first gap portion 1086 is a space portion between the first ESC unit 108 and the wafer 104.

[0024] In addition, the second ESC unit 110 also includes a second contact portion 1100, a second bulk portion 1102, a second electrode portion 1104, and a second gap portion 1106, corresponding to the first contact portion 1080, the first bulk portion 1082, the first electrode portion 1084, and the first gap portion 1086 of the first ESC unit 108, respectively.

[0025] 4B shows an equivalent circuit of the first ESC unit 108. The capacitor unit 1086C corresponds to the capacitance component of the first contact portion 1080, and the resistor unit 1080R corresponds to the resistance component of the first contact portion 1080. The capacitor unit 1082C corresponds to the capacitance of the first bulk portion 1082, and the resistor unit 1082R corresponds to the resistance component of the first bulk portion 1082. The resistor unit 1084R corresponds to the resistance component of the first electrode unit 1084. The second ESC unit 110 can also be represented as an equivalent circuit similar to that of the first ESC unit 108.

[0026] According to the above analytical model and equivalent circuit, it is inferred that the areas of the first gap portion 1086 and the second gap portion 1106 affect the capacitance, and that the areas of the first contact portion 1080 and the second contact portion 1100 affect the contact resistance between the wafer 104 and the ESC unit. For example, if the ESC surface deteriorates and the roughness of the ESC surface increases, the area of ​​the gap portion increases, increasing the amount of accumulated charge, and the area of ​​the contact portion decreases, increasing the contact resistance. In this case, during the static elimination process when the first power supply 116 and the second power supply 120 are turned off, it is assumed that the ESC current becomes larger than before deterioration and the rate of change of the transient change of the ESC current slows. Furthermore, if the rate of change of the transient change of the ESC current slows, there is a possibility that charge remains when the wafer 104 begins to be transported, which may result in an error in which the residual attraction force is not generated properly.

[0027] (Sensor Data) Next, sensor data will be described with reference to Figs. 5 and 6. Fig. 5 is a diagram showing an example of sensor data. The sensor data in Fig. 5 is data processed in the feature calculation unit 26. The item "Timestamp" indicates the time when the sensor data was acquired. The item "ESC Voltage 1" indicates the voltage applied to the first ESC unit 108. The item "ESC Current 1" indicates the current (first ESC current) measured by the first ammeter 114. The item "ESC Voltage 2" indicates the voltage applied to the second ESC unit 110. The item "ESC Current 2" indicates the current (second ESC current) measured by the second ammeter 118. Data in the item format shown in Fig. 5 is the sensor data D acquired by the data collection unit 22. ０ Alternatively, the sensor data D extracted by the data extraction processing unit 24 may be ２ Although the case where the ESC current and ESC voltage are detected every 0.1 seconds is shown, the detection interval for acquiring data can be set as appropriate.

[0028] FIG. 6 shows an example of a graph of sensor data. FIG. 6(a) shows the relationship between the voltage applied to the first ESC unit 108 and time, and FIG. 6(b) shows the relationship between the current flowing through the first ESC unit 108 and time. For example, if a voltage of 300 V is applied to the first ESC unit 108 at time t0, the current flowing through the first ESC unit 108 changes exponentially like an RC circuit and reaches a stable state after a certain time. A stable state is, for example, a state in which the change in current value over a predetermined time interval is less than a predetermined threshold. Also, if the voltage applied to the first ESC unit 108 changes from 300 V to 0 V at time t0, the direction of the current flowing through the first ESC unit 108 reverses (discharges), and the current value changes exponentially and reaches zero after a certain time. The change in current at time t0 occurs, for example, during the adsorption process, and the change in current at time t1 occurs, for example, during the static elimination process.

[0029] (Feature Calculation) Next, the calculation of the feature will be described with reference to Fig. 7 and Fig. 8. Fig. 7 is a flowchart showing the process of calculating the feature. First, the data collection unit 22 collects time-series data D(t) of the ESC current from the sensor data (step S1).

[0030] Next, the data extraction processing unit 24 records the time t1 when the ESC voltage is turned off (step S2). The data extraction processing unit 24 extracts data S(t) from the time series data D(t) for a predetermined period (first predetermined period) from time t1 (step S3). The predetermined period corresponds to the static elimination process and includes the transient change period until the ESC current stabilizes. Various methods can be used to set the predetermined period. For example, assuming that the transient change lasts for two seconds, data from the period from time t1 to two seconds can be extracted as the predetermined period data S(t). In this way, data from a period shorter or longer than the strict transient change time can be extracted as the predetermined period data S(t). The predetermined period should at least include the period during which the ESC current changes. As another method, for example, the difference between the ESC current at time (t1+x×n) (x is the detection interval of the sensor data (time series data D(t)), and n is an integer equal to or greater than 0) and time (t1+x×(n+1)) is calculated. ｅ If the difference in the ESC current is equal to or less than the predetermined threshold value at time t1 to time (t1+x×(n ｅ +1)) may be defined as a transient change interval, and data included in this interval may be extracted as data S(t) for a predetermined period.

[0031] Next, the feature calculation unit 26 calculates coefficients A and B of the function in Equation 1 using a regression method using data S(t) from a predetermined period (step S4). In Equation 1, i(t) is the value of the ESC current, A is a coefficient indicating the range of the ESC current value, and B is the time constant of the transient change. The value of coefficient B is used as feature F1 in the abnormality determination described below. While the regression method is used to calculate the coefficient, other methods for estimating the coefficient may also be used. Furthermore, Equation 1 reflects transient changes based on the configuration of the ESC unit shown in FIG. 4 (particularly the first contact portion 1080 and first gap portion 1086 in FIG. 4(a) (the resistor portion 1080R and capacitor portion 1086C in FIG. 4(b))), but it is possible to modify Equation 1 depending on changes to the configuration of the ESC unit. i(t) = A × EXP(-t ÷ B) Equation 1

[0032] Next, the anomaly detection unit 28 calculates the change in the feature value F1 over time and determines that an anomaly has occurred if the change exceeds a threshold value (step S5). The change in the feature value F1 over time can be understood by calculating the feature value F1 while the plasma processing apparatus 10 is operating to manufacture semiconductor devices and analyzing the change in the calculated feature value F1. The threshold value may be manually set by a user within an expected normal range, or may be automatically set by estimating the normal range from data obtained during past normal times.

[0033] 8A and 8B are diagrams showing a specific example of calculation of the feature amount F1 during static elimination processing and an example of change over time of the feature amount F1. Fig. 8A is a graph showing the relationship between time and the current flowing through the first ESC unit 108, Fig. 8B is a graph showing the relationship between time and the voltage applied to the first ESC unit 108, and Fig. 8C is an enlarged view of the transient change section of Fig. 8B.

[0034] As shown in FIG. 8A, a voltage is applied at time t0, and an adsorption process is initiated. After a period of voltage application, a static elimination process is initiated, and the application of the voltage is stopped at time t1. In the feature calculation process, first, as shown in FIG. 8A, the time when the voltage applied to the first ESC unit 108 is turned off is recorded as time t1 (corresponding to step S2 in FIG. 7). Next, as shown in FIG. 8B, time-series data of the current from time t1 to time (t1+2 seconds) is extracted as data S(t) for a predetermined period (corresponding to step S3 in FIG. 7).

[0035] 8(b) and 8(c), for example, the solid line indicates the current value when the ESC surface state is before degradation, and the dotted line indicates the current value when the ESC surface state is after degradation. At time t1, the current indicated by the dashed line is larger than the current indicated by the solid line, and after time t1, the current indicated by the dashed line has a steeper slope than the current indicated by the solid line and takes longer to reach zero. Thus, when the ESC surface state is deteriorated, the time constant of the transient change in the current value becomes large during the period corresponding to the static elimination process.

[0036] By the above calculation, a time constant F1 of the transient change in the current value during the static elimination process of the first ESC unit 108 is calculated as a feature amount (corresponding to step S4 in FIG. 7). The same process as that of the first ESC unit 108 is also performed for the second ESC unit 110, and a time constant F2 of the transient change in the current value during the static elimination process of the second ESC unit 110 is also calculated.

[0037] Although the case where the time constant is used as the feature has been described, the present disclosure is not limited to this case. The feature calculation unit 26 may calculate a statistic of the time-series data of the current or the slope of the trajectory of the time-series data of the current as the feature. Furthermore, when calculating the time constant, a general statistical processing method may be added to the feature calculation method for the purpose of noise reduction, etc.

[0038] (Abnormality Determination) Next, the abnormality determination process based on the feature values ​​will be described with reference to FIG. 9 (corresponding to step S5 in FIG. 7). FIG. 9 is a diagram showing the change over time of the feature values. FIG. 9(a) shows the change over time of the feature value F1, and FIG. 9(b) shows the change over time of the feature value F2. In FIG. 9, the horizontal axis shows the cumulative time of the etching process performed in the plasma processing apparatus 10 as an index showing the change over time.

[0039] Specifically, the abnormality determination process is performed by analyzing the time series of the feature quantities, as shown in FIG. 9 . The threshold value for determining whether or not the sample stage 106 is abnormal is a value determined based on the time constant calculated as the feature quantity when the sample stage 106 is in a normal state, or a value determined based on the type of sample. For example, in FIG. 9( a), two upper and lower threshold values ​​th1 and th2 are set for the feature quantity F1. If the feature quantity F1 exceeds either threshold, the feature quantity F1 is determined to be abnormal. Also, in FIG. 9( b), one threshold value th3 is set for the feature quantity F2. If the feature quantity F2 exceeds the threshold value th3, the feature quantity F2 is determined to be abnormal. If either F1 or F2 is determined to be abnormal overall, it is determined that an abnormality has occurred in the ESC surface condition of the plasma processing apparatus 10. The threshold value setting method can be set as appropriate. The sensor data output from the plasma processing apparatus 10 may include the type of wafer 104, or an ID number or the like may be added to identify the sample.

[0040] Although the cumulative etching time has been used as an indicator of change over time, the present disclosure is not limited to this. For example, the number of wafers etched in the plasma processing apparatus 10 may be used as an indicator of change over time. Each time a wafer is etched, sensor data is acquired, a feature value is calculated, and an abnormality determination can be performed.

[0041] (Output of Determination Result) Next, an example of output of the determination result of the abnormality determination will be described with reference to Fig. 10 . Fig. 10 is a diagram showing an example of a GUI (Graphic User Interface) screen 200 that displays the determination result. The GUI screen 200 can be presented to the user by, for example, the abnormality detection unit 28 via an input / output device such as a display. The abnormality detection unit 28 can display feature amounts (for example, in list form), changes over time in the feature amounts, calculation results, abnormality determination results, etc. in the GUI (Graphic User Interface), and can present actions (countermeasures) if an abnormality occurs.

[0042] First, a form on the GUI screen 200 that the user uses to input and transmit information to the diagnostic device 20 will be described. Device ID 202 indicates information identifying the plasma processing device 10 from which sensor data was acquired. Start time 204 indicates the start time of the period for which abnormality determination is to be performed. End time 206 indicates the end time of the period for which abnormality determination is to be performed. Wafer ID 208 indicates information identifying the wafer for which feature quantities are to be calculated. Threshold value 210 indicates information identifying the method for setting the threshold value. Methods for setting the threshold value include, for example, manual setting by the user and automatic setting by the diagnostic device 20 based on predetermined information. In either form, each piece of information can be set by text input, or each piece of information can be selected from a drop-down list.

[0043] For example, a plasma processing apparatus to be diagnosed can be selected by setting information using an apparatus ID 202, a start time 204, and an end time 206. A threshold value used to determine an abnormality of a feature can be controlled by setting information using a wafer ID 208 and a threshold value 210.

[0044] The feature list 212 indicates thresholds of the feature values ​​used for abnormality determination. If the threshold setting method in the threshold value setting 210 is set to manual, the user can set the feature values, parameter values, and threshold values ​​used for diagnosis from the feature value list 212.

[0045] The time-dependent changes of each calculated feature amount are displayed in the area of ​​the abnormality determination 214. In Fig. 10, the time-dependent changes of the feature amounts F1 and F2 are shown.

[0046] If it is determined that there is an abnormality in the feature, the abnormal feature is displayed in the alarm section 216. In the action section 218, work such as performing maintenance or adjusting process conditions is displayed as a countermeasure against the abnormality.

[0047] The second embodiment differs from the first embodiment in that an abnormality is determined using the ESC current during the adsorption process instead of the ESC current during the static elimination process. In the following description, the same or equivalent components as those in the first embodiment are denoted by the same reference numerals, and their description will be simplified or omitted.

[0048] 11 is a flowchart showing the feature calculation process. First, the data collection unit 22 collects time-series data D(t) of the ESC current from the sensor data (step S21).

[0049] Next, the data extraction processing unit 24 extracts data for a second predetermined period corresponding to the adsorption process from the time-series data D(t) of the ESC current. Specifically, the data extraction processing unit 24 records the time t0 when the ESC voltage is turned on (voltage is applied) (step S22). The data extraction processing unit 24 extracts data R(t) for a predetermined period (second predetermined period) from the time t0 from the time-series data D(t) (step S23).

[0050] Next, the feature calculation unit 26 calculates the time constant of the transient change of the data for the second predetermined period as a feature. Specifically, the feature calculation unit 26 calculates coefficients A and B of the function of Equation 1 by a regression method using the data R(t) for the predetermined period (step S24). The value of coefficient B is used as feature F3 in abnormality determination.

[0051] 12A and 12B are diagrams showing a specific example of calculation of the feature quantity F3 during adsorption processing and an example of change over time in the feature quantity F3. Fig. 12A is a graph showing the relationship between time and the current flowing through the first ESC unit 108, Fig. 12B is a graph showing the relationship between time and the voltage applied to the first ESC unit 108, and Fig. 12C is an enlarged view of the transient change section of Fig. 12B.

[0052] As shown in Fig. 12(a), voltage is applied at time t0, the adsorption process is initiated, and the voltage application state continues. In the feature calculation process, first, as shown in Fig. 12(a), the time when the voltage of the first ESC unit 108 is turned on is recorded as time t0 (corresponding to step S22 in Fig. 11). Next, as shown in Fig. 12(b), time-series data of the current from time t0 to time (t0 + 2 seconds) is extracted as data R(t) for a predetermined period (corresponding to step S23 in Fig. 12).

[0053] 12(b) and 12(c), for example, the solid line indicates the current value before the ESC surface state deteriorates, and the dotted line indicates the current value after the ESC surface state deteriorates. For example, at time t0, the current indicated by the dashed line has a larger current value than the current indicated by the solid line, and after time t0, the current indicated by the dashed line has a larger slope than the current indicated by the solid line and takes longer to reach zero. Thus, when the ESC surface state deteriorates, the time constant of the transient change in the current value becomes large during the period corresponding to the adsorption process.

[0054] By the above calculation, a time constant F1 of the transient change in the current value during the static elimination process of the first ESC unit 108 is calculated as a feature amount (corresponding to step S24 in FIG. 11 ). The same process as that of the first ESC unit 108 is also performed for the second ESC unit 110, and a time constant F2 of the transient change in the current value during the adsorption process of the second ESC unit 110 is also calculated.

[0055] Example 3 differs from Examples 1 and 2 in that analysis during static elimination processing and analysis during adsorption processing are performed simultaneously, and abnormalities are determined using feature quantities from both processings. In the following description, components that are the same as or equivalent to those in Example 1 above are denoted by the same reference numerals, and their description will be simplified or omitted.

[0056] FIG. 13 is a flowchart showing the feature calculation process. Steps S32 to S34 correspond to steps S2 to S4 in FIG. 7 and are equivalent to analysis during the static elimination process. Steps S35 to S37 correspond to steps S22 to S24 in FIG. 11 and are equivalent to analysis during the power supply process. The data collection unit 22 collects time-series data D(t) of the ESC current from the sensor data (step S31). The data extraction processing unit 24 extracts time-series data of the current during the static elimination process and time-series data of the current during the power supply process (steps S32, S33, S35, and S36). The feature calculation unit 26 calculates feature F3 from feature F1 (steps S34 and S37).

[0057] The abnormality detection unit 28 uses a plurality of feature amounts calculated from the time-series data of the current in the static elimination process and the time-series data of the current in the adsorption process to determine an abnormality in the sample stage 106. Specifically, the abnormality detection unit 28 calculates the changes over time in the feature amounts F1 and F2, and determines that an abnormality has occurred in the ESC unit if either or both of the feature amounts exceed a threshold value (step S38).

[0058] (Actions and Effects) According to the present disclosure, it is possible to grasp the surface condition of the sample stage and diagnose the occurrence of an abnormality. Even if the amount of charge on the surface condition of the sample stage (ESC) cannot be calculated due to the occurrence of leakage current, it is possible to grasp the surface condition of the sample stage by using the time constant of the transient change as a feature and analyzing the change in the feature over time.

[0059] Although the embodiments of the present invention have been described above, the present invention is not limited to the above-described embodiments, and various modifications are possible within the scope of the gist of the present invention.

[0060] The following are some possible aspects that may be included in the present invention, but the present invention is not limited to these. (Aspect 1) A diagnostic device that diagnoses the presence or absence of an abnormality in a sample stage on which a sample is placed, wherein the presence or absence of an abnormality in the sample stage is determined based on a time constant in a transient change in current when a static elimination process is performed, wherein the static elimination process is a process for detaching the sample electrostatically attracted to the sample stage from the sample stage, and wherein the current is a current that flows through the sample stage when a voltage for electrostatically attracting the sample to the sample stage is applied to the sample stage. (Aspect 2) A diagnostic device that diagnoses the presence or absence of an abnormality in the sample stage on which a sample is placed, wherein the presence or absence of an abnormality in the sample stage is determined based on a time constant in a transient change in current when the sample is electrostatically attracted to the sample stage, and wherein the current is a current that flows through the sample stage when a voltage for electrostatically attracting the sample to the sample stage is applied to the sample stage. (Aspect 3) The diagnostic device according to Aspect 1 or Aspect 2, wherein the presence or absence of an abnormality in the sample stage is determined based on a time constant in a transient change in the current when the sample is electrostatically attracted to the sample stage. (Aspect 4) The diagnostic device according to any one of Aspects 1 to Aspect 3, wherein a statistical quantity in the time-series data of the current or a slope of a trajectory in the time-series data of the current is calculated as a feature amount, and the presence or absence of an abnormality in the sample stage is determined based on the calculated feature amount. (Aspect 5) The diagnostic device according to any one of Aspects 1 to Aspect 4, wherein the threshold value for determining the presence or absence of an abnormality in the sample stage is a value defined based on the time constant calculated as a feature amount when the sample stage is in a normal state, or a value defined based on the type of the sample. (Aspect 6) The diagnostic device according to any one of Aspects 1 to Aspect 5, wherein the time constant is calculated as a feature amount, and wherein the diagnostic device comprises a GUI that displays the feature amount, changes in the feature amount over time, and the result of the determination.(Aspect 7) A diagnostic method for diagnosing the presence or absence of an abnormality in a sample stage on which a sample is placed, comprising: a step of determining the presence or absence of an abnormality in the sample stage based on a time constant in a transient change in current when a static elimination process is performed, wherein the static elimination process is a process of detaching the sample electrostatically attracted to the sample stage from the sample stage, and wherein the current is a current that flows through the sample stage when a voltage for electrostatically attracting the sample to the sample stage is applied to the sample stage. (Aspect 8) A diagnostic method for diagnosing the presence or absence of an abnormality in a sample stage on which a sample is placed, comprising: a step of determining the presence or absence of an abnormality in the sample stage based on a time constant in a transient change in current when the sample is electrostatically attracted to the sample stage, and wherein the current is a current that flows through the sample stage when a voltage for electrostatically attracting the sample to the sample stage is applied to the sample stage. (Aspect 9) A semiconductor device manufacturing system equipped with a platform on which an application is implemented that diagnoses the presence or absence of an abnormality in a sample stage on which a sample is placed, the application executing a step of determining the presence or absence of an abnormality in the sample stage based on a time constant in a transient change in current when a static elimination process is performed, the static elimination process being a process of detaching the sample electrostatically attracted to the sample stage from the sample stage, the current being a current that flows through the sample stage when a voltage for electrostatically attracting the sample to the sample stage is applied to the sample stage. (Aspect 10) A semiconductor device manufacturing system equipped with a platform on which an application is implemented that diagnoses the presence or absence of an abnormality in a sample stage on which a sample is placed, the application executing a step of determining the presence or absence of an abnormality in the sample stage based on a time constant in a transient change in current when the sample is electrostatically attracted to the sample stage, the current being a current that flows through the sample stage when a voltage for electrostatically attracting the sample to the sample stage is applied to the sample stage. (Aspect 11) The semiconductor device manufacturing system according to aspect 9 or aspect 10, wherein the platform is a server.

[0061] 1 Semiconductor device manufacturing system (diagnostic system), 10 Plasma processing apparatus, 20 Diagnostic apparatus, 22 Data collection unit, 24 Data extraction processing unit, 26 Feature calculation unit, 28 Abnormality detection unit, 100 Chamber, 102 Plasma, 104 Wafer (sample), 106 Sample stage, 108 First electrostatic chuck unit (first ESC unit), 110 Second electrostatic chuck unit (second ESC unit), 112 Substrate, 114 First ammeter, 116 First power supply, 118 Second ammeter, 120 Second power supply, 200 GUI screen, 1080: First contact portion, 1080R Resistor portion, 1082 First bulk portion, 1082C Capacitor portion, 1082R Resistor portion, 1084 First electrode portion, 1084R Resistor portion, 1086 First gap portion, 1086C Capacitor portion, 1100 Second contact portion, 1102, second bulk portion, 1106, second gap portion

Claims

1. A diagnostic device for diagnosing the presence or absence of an abnormality in a sample stage on which a sample is placed, characterized in that the presence or absence of an abnormality in the sample stage is determined based on the time constant in the transient change in current when a static elimination process is performed, the static elimination process is a process for detaching the sample electrostatically adsorbed to the sample stage from the sample stage, and the current is a current that flows through the sample stage when a voltage for electrostatically adsorbing the sample to the sample stage is applied to the sample stage.

2. A diagnostic device for diagnosing the presence or absence of an abnormality in a sample stage on which a sample is placed, characterized in that the presence or absence of an abnormality in the sample stage is determined based on the time constant of the transient change in current when the sample is electrostatically attracted to the sample stage, and the current is the current that flows through the sample stage when a voltage for electrostatically attracting the sample to the sample stage is applied to the sample stage.

3. A diagnostic device according to claim 1, characterized in that the presence or absence of an abnormality in the sample stage is determined based on the time constant in the transient change in the current when the sample is electrostatically attracted to the sample stage.

4. A diagnostic device according to claim 1, wherein a statistical quantity in the time series data of the current or a slope of the trajectory in the time series data of the current is calculated as a feature, and the presence or absence of an abnormality in the sample stage is determined based on the calculated feature.

5. A diagnostic device according to claim 1, wherein the threshold value for determining whether or not there is an abnormality in the sample stage is a value defined based on the time constant calculated as a characteristic quantity when the sample stage is in a normal state, or a value defined based on the type of sample.

6. A diagnostic device according to claim 1, wherein the time constant is calculated as a feature, and the diagnostic device is provided with a GUI that displays the feature, changes in the feature over time, and the results of the judgment.

7. A diagnostic method for diagnosing whether or not there is an abnormality in a sample stage on which a sample is placed, comprising a step of determining whether or not there is an abnormality in the sample stage based on the time constant in the transient change in current when a static elimination process is performed, wherein the static elimination process is a process of removing the sample electrostatically adsorbed to the sample stage from the sample stage, and the current is a current that flows through the sample stage when a voltage for electrostatically adsorbing the sample to the sample stage is applied to the sample stage.

8. A diagnostic method for diagnosing whether or not there is an abnormality in a sample stage on which a sample is placed, comprising a step of determining whether or not there is an abnormality in the sample stage based on the time constant of the transient change in current when the sample is electrostatically attracted to the sample stage, wherein the current is the current that flows through the sample stage when a voltage for electrostatically attracting the sample to the sample stage is applied to the sample stage.

9. A semiconductor device manufacturing system having a platform on which an application is implemented that diagnoses the presence or absence of an abnormality in a sample stage on which a sample is placed, which is provided in semiconductor manufacturing equipment, wherein the application executes a step of determining the presence or absence of an abnormality in the sample stage based on the time constant of the transient change in current when a charge removal process is performed, the charge removal process is a process of detaching the sample electrostatically adsorbed to the sample stage from the sample stage, and the current is a current that flows through the sample stage when a voltage for electrostatically adsorbing the sample to the sample stage is applied to the sample stage.

10. A semiconductor device manufacturing system equipped with a platform on which an application for diagnosing the presence or absence of an abnormality in a sample stage on which a sample is placed, which is provided in semiconductor manufacturing equipment, wherein the application executes a step of determining the presence or absence of an abnormality in the sample stage based on the time constant of the transient change in current when the sample is electrostatically attracted to the sample stage, and the current is the current that flows through the sample stage when a voltage for electrostatically attracting the sample to the sample stage is applied to the sample stage.

11. A semiconductor device manufacturing system according to claim 9 or 10, wherein the platform is a server.

Images