Plasma processing device and electrode consumption measurement method

Abstract

The present invention provides a plasma processing device and an electrode consumption measurement method, which can measure the consumption of an upper electrode with high precision without opening a chamber. The plasma processing device (100) has a shower head (16) and a support table (2) facing each other. A first RF power supply (10a) applies RF power to any electrode of the shower head (16) and the support table (2) in a manner that does not cause plasma ignition. A measuring device (204) measures a physical quantity related to the RF power applied from the first RF power supply (10a). A process controller (91) uses the measured physical quantity related to the RF power in a correlation function between the distance between the electrodes between the shower head (16) and the support table (2) and the physical quantity related to the RF power to calculate the distance between the electrodes.

Description

Technical Field

[0001] The invention relates to a plasma processing device and an electrode consumption measuring method. Background Art

[0002] Plasma processing apparatuses are known for using plasma to perform plasma treatment on a target object, such as a wafer. Such plasma processing apparatuses include, for example, a mounting table for holding the target object, which also serves as an electrode, within a processing chamber capable of forming a vacuum space. The plasma processing apparatus applies a predetermined high-frequency power to the mounting table to perform plasma treatment on the target object, which is placed on the mounting table.

[0003] Prior art literature

[0004] Patent Literature

[0005] Patent Document 1: Japanese Patent Application Laid-Open No. 2015-115541 Summary of the Invention

[0006] Problems to be solved by the invention

[0007] Provided is a technology capable of measuring the consumption of an upper electrode with high precision without opening a chamber.

[0008] Solutions for solving problems

[0009] In one embodiment of the disclosed plasma processing apparatus and electrode consumption measurement method, the plasma processing apparatus includes a first electrode and a second electrode facing each other. A power application unit applies RF (Radio Frequency) power to either the first electrode or the second electrode without igniting plasma. A measurement unit measures a physical quantity related to the RF power applied by the power application unit. A calculation unit applies the measured physical quantity related to the RF power to a correlation function between the inter-electrode distance between the first electrode and the second electrode and a physical quantity related to the RF voltage to determine the inter-electrode distance.

[0010] Effects of the Invention

[0011] According to one embodiment of the disclosed plasma processing apparatus and electrode consumption measurement method, the consumption of the upper electrode can be measured with high accuracy without opening the chamber. BRIEF DESCRIPTION OF THE DRAWINGS

[0012] Figure 1 It is a schematic cross-sectional view showing the structure of a plasma processing apparatus.

[0013] Figure 2 It is a diagram schematically showing the upper electrode and the lower electrode.

[0014] Figure 3 This is a diagram schematically showing a plasma processing apparatus.

[0015] Figure 4 This is a diagram showing the relationship between reflected waves, Vpp, and RF power when RF is totally reflected.

[0016] Figure 5 This is a diagram showing an example of a measurement process.

[0017] Figure 6 This is a diagram for explaining the outline of calculation of consumption amount.

[0018] Figure 7 This is a flow chart of a plasma etching process performed using a plasma processing apparatus and utilizing consumption measurement.

[0019] Figure 8 It is a diagram showing the arrangement state of the upper top plate when acquiring reference data.

[0020] Figure 9 This is a graph showing a correlation straight line obtained based on the reference Vpp.

[0021] Figure 10 This is a diagram showing the configuration state of the upper top plate when acquiring consumption time data.

[0022] Figure 11 FIG. 1 is a diagram showing the state of Vpp during consumption.

[0023] Figure 12 This figure shows the measurement results when the consumption is measured by changing the state using an unused upper top plate.

[0024] Figure 13 This figure shows the measurement results when the consumption amount is measured by changing the state using an upper top plate with a simulated consumption of 2 mm.

[0025] Figure 14 It is a graph showing the measurement results when metal etching was performed.

[0026] Figure 15 This is a diagram for explaining the measurement of consumption when Vpp is measured at a high resolution.

[0027] Description of Reference Numerals

[0028] W: wafer; 1: processing container; 2: mounting table; 16: shower head; 16a: main body; 16b: upper top plate; 91: process controller; 92: user interface; 93: storage unit; 100: plasma processing device. DETAILED DESCRIPTION

[0029] Below, embodiments of the plasma processing apparatus disclosed in this application are described in detail with reference to the accompanying drawings. In the various drawings, identical or corresponding parts are denoted by the same reference numerals. Furthermore, the disclosed invention is not limited by this embodiment. The various embodiments can be appropriately combined to the extent that no conflicting treatment contents are present. In the various drawings, identical or corresponding parts are denoted by the same reference numerals. The terms "upper" and "lower" are used for convenience based on the states shown in the drawings.

[0030] In recent years, attempts have been made to mitigate the effects of process variation caused by component consumption by varying various control parameters through feedback control. Among the components that make up the chamber, the characteristic values ​​of the upper electrode (CEL) are particularly affected by consumption. Therefore, it is desirable to accurately understand the consumption of the upper electrode and control it with an appropriate amount to recover from etching rate variations caused by consumption and improve deteriorated product uniformity.

[0031] Typically, the exact consumption of the component is measured using a vernier or other tool after the chamber is opened and the component is removed, or measurement is outsourced to an external technician, which risks reducing productivity. Alternatively, there are methods for measuring without opening the chamber, such as estimating consumption based on RF accumulation time. However, since component consumption varies depending on conditions, adjusting the control amount of the knob is difficult.

[0032] (First embodiment)

[0033] [Structure of Plasma Processing Apparatus]

[0034] Figure 1This is a schematic cross-sectional view showing the structure of a plasma processing apparatus. The plasma processing apparatus 100 is constructed in an airtight manner and has a processing container 1 that is electrically set to a ground potential. The processing container 1 is cylindrical and is made of, for example, aluminum. The processing container 1 divides a processing space for generating plasma. A mounting table 2 is provided in the processing container 1, and the mounting table 2 horizontally supports a semiconductor wafer (hereinafter referred to as "wafer") W serving as a substrate (work-piece). The mounting table 2 includes a base (base) 2a and an electrostatic chuck (ESC: Electrostatic chuck) 6. The base 2a is made of a conductive metal, such as aluminum, and has a function as a lower electrode. The electrostatic chuck 6 has a function for electrostatically adsorbing the wafer W. The mounting table 2 is supported by a support table 4. The support table 4 is supported by a support member 3 made of, for example, quartz. In addition, a focusing ring 5 made of, for example, single crystal silicon is provided on the outer periphery above the mounting table 2. Furthermore, a cylindrical inner wall member 3 a made of, for example, quartz is provided in the processing chamber 1 so as to surround the mounting table 2 and the support table 4 .

[0035] The susceptor 2a is connected to a first RF power source 10a via a first matching box 11a and to a second RF power source 10b via a second matching box 11b. The first RF power source 10a is primarily used to generate plasma and is configured to supply high-frequency power of a predetermined frequency selected from the range of 150 MHz to 10 MHz to the susceptor 2a of the mounting table 2. Furthermore, the second RF power source 10b is primarily used to attract ions (for biasing) and is configured to supply high-frequency power of a predetermined frequency selected from the range of 40 MHz to 100 MHz, lower than that of the first RF power source 10a, to the susceptor 2a of the mounting table 2. In this manner, the mounting table 2 is configured to be capable of voltage application.

[0036] Furthermore, the frequency of the first RF power source 10a simultaneously attracts ions while generating plasma, with lower frequencies resulting in a greater proportion of ion attraction. Furthermore, the frequency of the second RF power source 10b simultaneously attracts ions while generating plasma, with higher frequencies resulting in a greater proportion of plasma generation.

[0037] On the other hand, a shower head 16 functioning as an upper electrode is provided above the mounting table 2 so as to be parallel to and face the mounting table 2. The shower head 16 and the mounting table 2 function as a pair of electrodes (an upper electrode and a lower electrode).

[0038] Electrostatic chuck 6 is configured such that electrode 6a is interposed between insulators 6b, and electrode 6a is connected to DC power supply 12. When DC voltage is applied from DC power supply 12 to electrode 6a, wafer W is attracted by Coulomb force.

[0039] A refrigerant flow path 2d is formed inside the mounting table 2 and is connected to the refrigerant inlet pipe 2b and the refrigerant outlet pipe 2c. Furthermore, the mounting table 2 is configured to be controlled to a predetermined temperature by circulating a suitable refrigerant, such as cooling water, through the refrigerant flow path 2d. Furthermore, a gas supply pipe 30 is provided that passes through the mounting table 2, etc., for supplying a heat transfer gas (backside gas) such as helium to the back side of the wafer W. The gas supply pipe 30 is connected to a gas supply source (not shown). Through these structures, the wafer W, which is held by the electrostatic chuck 6 on the upper surface of the mounting table 2, is controlled to a predetermined temperature.

[0040] The focus ring 5 provided on the outer periphery of the mounting table 2 is also controlled to a predetermined temperature. Alternatively, a heater may be provided inside the mounting table 2 or the electrostatic chuck 6 to heat the wafer W and the focus ring 5 to a predetermined temperature.

[0041] The mounting table 2 is provided with a plurality of, for example, three, pin through holes 200 (in Figure 1 Only one is shown in the figure. ), lift pins 61 are respectively provided inside these pin through holes 200. The lift pins 61 are connected to a drive mechanism 62 and are moved up and down by the drive mechanism 62.

[0042] The shower head 16 is mounted on the ceiling of the processing container 1. The shower head 16 includes a main body 16a and an upper top plate 16b constituting an electrode plate, and is supported on the upper portion of the processing container 1 via an insulating member 95. The main body 16a is made of a conductive material, such as anodized aluminum, and the upper top plate 16b is detachably supported on the lower portion of the main body 16a.

[0043] A gas diffusion chamber 16c is provided within the main body 16a. Furthermore, a large number of gas flow holes 16d are formed at the bottom of the main body 16a, positioned below the gas diffusion chamber 16c. Furthermore, gas inlet holes 16e are provided in the upper top plate 16b, overlapping with the aforementioned gas flow holes 16d, extending through the thickness of the upper top plate 16b. With this structure, the processing gas supplied to the gas diffusion chamber 16c is distributed and sprayed into the processing vessel 1 via the gas flow holes 16d and the gas inlet holes 16e.

[0044] A gas inlet port 16g for introducing a processing gas into the gas diffusion chamber 16c is formed in the main body 16a. The gas inlet port 16g is connected to one end of a gas supply pipe 15a. The other end of the gas supply pipe 15a is connected to a processing gas supply source (gas supply unit) 15 that supplies processing gas. In the gas supply pipe 15a, a mass flow controller (MFC) 15b and an on-off valve V2 are provided in sequence from the upstream side. Processing gas for plasma etching is supplied from the gas supply source 15 to the gas diffusion chamber 16c via the gas supply pipe 15a. Processing gas is supplied to the processing container 1 in a dispersed manner in a spray shape from the gas diffusion chamber 16c via the gas flow hole 16d and the gas inlet port 16e.

[0045] The shower head 16, which serves as the upper electrode, is electrically connected to a variable DC power supply 72 via a low-pass filter (LPF) 71. The variable DC power supply 72 is configured to be able to be turned on / off by an on / off switch 73. The current / voltage of the variable DC power supply 72 and the on / off operation of the on / off switch 73 are controlled by a control unit 90, which will be described later. Furthermore, as will be described later, when a high frequency is applied from the first RF power supply 10a and the second RF power supply 10b to the mounting table 2 to generate plasma in the processing space, the on / off switch 73 is turned on by the control unit 90 as needed. Thus, a predetermined DC voltage is applied to the shower head 16, which serves as the upper electrode.

[0046] A cylindrical ground conductor 1a is provided so as to extend from the side wall of the processing container 1 to a position above the height position of the shower head 16. The cylindrical ground conductor 1a has a ceiling at its upper portion.

[0047] An exhaust port 81 is formed at the bottom of the processing container 1. Exhaust port 81 is connected to a first exhaust unit 83 via an exhaust pipe 82. The first exhaust unit 83 includes a vacuum pump, and is configured to reduce the pressure within the processing container 1 to a predetermined vacuum level by operating the vacuum pump. Meanwhile, a wafer W loading / unloading port 84 is provided on a side wall of the processing container 1. A gate valve 85 is provided at this loading / unloading port 84 for opening and closing the port.

[0048] A sediment shield 86 is installed along the inner wall of the processing container 1. Sediment shield 86 prevents etching byproducts (sediments) from adhering to the processing container 1. A conductive member (GND block) 89, connected to a ground potential that can be controlled, is installed at the same height as the wafer W on sediment shield 86 to prevent abnormal discharge. Sediment shield 87 is also installed at the lower end of sediment shield 86, extending along the inner wall member 3a. Sediment shields 86 and 87 are detachable.

[0049] The operation of the plasma processing apparatus 100 having the above-described structure is centrally controlled by a control unit 90. The control unit 90 includes a process controller 91 having a CPU and controlling various components of the plasma processing apparatus 100, a user interface 92, and a storage unit 93.

[0050] The user interface 92 includes a keyboard for the process manager to input commands for managing the plasma processing apparatus 100 , a display for visually displaying the operating status of the plasma processing apparatus 100 , and the like.

[0051] The storage unit 93 stores recipes, including control programs (software) and processing condition data, for implementing various processes executed by the plasma processing apparatus 100 under the control of the process controller 91. Furthermore, as needed, any recipe can be retrieved from the storage unit 93 in accordance with instructions from the user interface 92 and executed by the process controller 91, thereby performing the desired process using the plasma processing apparatus 100 under the control of the process controller 91. Furthermore, recipes such as control programs and processing condition data can be used while stored in a computer-readable storage medium (e.g., a hard disk, CD, floppy disk, semiconductor memory, etc.), or can be transferred from another device via, for example, a dedicated line and used online.

[0052] [Measurement of upper electrode consumption]

[0053] Next, we'll explain how to measure the consumption of the upper electrode. During plasma etching, the electrode plate of the showerhead 16, namely the upper top plate 16b, which actually serves as the upper electrode, is consumed. Therefore, the following description focuses on the consumption of the upper top plate 16b. However, when considering the main body 16a and upper top plate 16b as an integrated showerhead, the consumption of the upper top plate 16b can also be considered the consumption of the upper electrode (CEL).

[0054] Figure 2 : is a diagram schematically showing an upper electrode and a lower electrode. The inter-electrode distance d is the distance between the facing surfaces of the shower head 16 as the upper electrode and the mounting table 2 as the lower electrode.

[0055] in addition, Figure 3 FIG is a diagram schematically showing a plasma processing apparatus. Figure 3 As shown, the plasma processing apparatus 100 is generally divided into a power supply 201, a matching circuit 202, and a chamber space 203. The power supply 201 includes, for example, Figure 1 In addition, the matching circuit 202 includes, for example, Figure 1The first matching device 11a in the process chamber 10 is provided, and includes variable capacitors C1 and C2 as the circuit structure of the first matching device 11a. The chamber space 203 includes various components arranged inside the process container 1. The measuring device 204 measures and outputs the peak-to-peak voltage (Vpp) of the RF power supplied from the first RF power source 10a to the showerhead 16 and the mounting stage 2 via the first matching device 11a.

[0056] When plasma etching is performed, the upper plate 16b of the shower head 16, which serves as the upper electrode (CEL), is consumed. When the upper plate 16b is consumed, the inter-electrode distance d becomes longer than before consumption, even if the shower head 16 and the mounting table 2 are arranged at the same position. Figure 3 The impedance of the capacitive component C between the shower head 16 and the mounting table 2 in the plasma processing apparatus 100 shown in FIG is Z = 1 / jωC. Furthermore, C = ε(S / d), so when d changes, the electrostatic capacitance changes, and accordingly, the impedance between the shower head 16 and the mounting table 2 changes.

[0057] When plasma is generated, the RF power output from the first RF power source 10a is absorbed by the load (plasma), resulting in no reflected wave or a small observed value. Furthermore, since the magnitude of the reflected wave is determined by changes in the load impedance, it varies with changes in the plasma state. Similar to the reflected wave, the Vpp value also fluctuates due to changes in the load impedance caused by interference. Therefore, the reliability of the Vpp measurement during plasma ignition is low, and using Vpp for control carries the risk of inaccurate control.

[0058] On the other hand, when RF radiation emitted from the mounting table 2, serving as the lower electrode, is totally reflected by the upper ceiling 16b of the showerhead 16, serving as the upper electrode, the power provided without generating plasma returns to the mounting table 2 as a reflected wave. During non-plasma ignition, the influence of conditions within the processing chamber 1, such as gas flow rate and pressure, is mitigated, suppressing impedance fluctuations. Furthermore, since RF radiation is totally reflected rather than absorbed on the load side, the magnitude of the reflected wave remains constant. Furthermore, when RF radiation is totally reflected, the Vpp value varies due to changes in capacitance associated with changes in the inter-electrode distance d caused by depletion.

[0059] Figure 4 This is a diagram showing the relationship between the reflected wave and Vpp and RF power when RF is totally reflected. Figure 4In graphs 301 and 302, the horizontal axis represents RF power, and the vertical axis represents reflected wave power. Graph 301 shows the RF output at a relatively high frequency of 40 MHz, while graph 302 shows the RF output at a relatively low frequency of 13 MHz. In graphs 303 and 304, the horizontal axis represents RF power, while the vertical axis represents Vpp squared. Graph 303 shows the RF output at a relatively high frequency of 40 MHz, while graph 304 shows the RF output at a relatively low frequency of 13 MHz.

[0060] exist Figure 4 In the graphs 301 and 302, even if the distance between electrodes is changed stepwise to 30 mm, 40 mm, and 50 mm, the same straight line is obtained in all cases. That is, even if the distance between electrodes d changes, the power of the reflected wave remains the same. Figure 4 As shown in the graphs 301 and 302 , it is fixed.

[0061] In contrast, line 331 in graph 303 shows the change in Vpp when the inter-electrode distance is 50 mm, line 332 shows the change in Vpp when the inter-electrode distance is 40 mm, and line 333 shows the change in Vpp when the inter-electrode distance is 30 mm. Furthermore, line 341 in graph 304 shows the change in Vpp when the inter-electrode distance is 50 mm, line 342 shows the change in Vpp when the inter-electrode distance is 40 mm, and line 343 shows the change in Vpp when the inter-electrode distance is 30 mm.

[0062] As shown in graphs 303 and 304, Vpp changes with capacitance when the inter-electrode distance is varied to 30 mm, 40 mm, and 50 mm. The plasma processing apparatus 100 of this embodiment measures this change in Vpp to determine the change in the distance between the showerhead 16 and the mounting table 2, and thereby calculates the consumption of the upper top plate 16b. The consumption measurement process is described in detail below.

[0063] The unused upper top plate 16 b of the shower head 16 is mounted on the plasma processing apparatus 100 . Then, a wafer is mounted on the mounting table 2 .

[0064] The process controller 91 then receives instructions from the user interface 92 and executes a reference data acquisition process for acquiring reference data used for calculating consumption. Specifically, the process controller 91 uses the measurement recipe stored in the storage unit 93 to apply RF power that does not generate plasma and causes total reflection to the mounting table 2 on which the wafer is mounted, thereby causing total reflection of the RF power and thereby acquiring the reference data.

[0065] At this time, the process controller 91 measures Vpp after leaving an idle time of at least one hour before executing the reference data acquisition process to cool the internal temperature of the processing container 1. This is to reduce the change in impedance due to the influence of temperature change.

[0066] Figure 5 This is a diagram showing an example of a measurement process. For example, as a processing condition registered in the measurement process, Figure 5 The process shown. The output of the RF power in the measurement process is an output that does not generate plasma and causes total reflection. In this embodiment, regarding RF, when the lower 13 MHz is used, the RF output is 700 W. In addition, the minimum value of the pressure of the processing container 1 is set to 0 mT, and the processing gas is set to 0 sccm, that is, a state where the processing gas is not flowing. In addition, the variable capacitors C1 and C2 of the first matching unit 11a are fixed to 0 / 0 [%], which represents the lower limit value of the range of variable electrostatic capacitance. In addition, the temperature of the electrostatic chuck 6 and the focusing ring 5 is set to 30°C, and the temperature of the shower head 16 as the upper electrode of the processing container 1 and the side wall of the processing container 1, namely the deposition shield 86 and 87, is set to 150°C. In addition, the electrode distance d is set to a value between 30 mm and 50 mm, which varies in increments of 1 mm. Figure 5 The Gap in the figure represents the value of the inter-electrode distance d during the reference data acquisition process. Furthermore, the set values ​​for the electrostatic capacitance of variable capacitors C1 and C2, and the temperatures of electrostatic chuck 6, focus ring 5, showerhead 16, and deposit shields 86 and 87 are not limited to these values. For example, the values ​​of variable capacitors C1 and C2 may be 50 / 50 [%] or 100 / 100 [%]. Furthermore, to improve measurement accuracy, it is desirable that the values ​​of variable capacitors C1 and C2 be fixed.

[0067] Here, when measuring Vpp for obtaining the consumption, the plasma is not ignited, so Figure 5 As shown in the measurement process, the pressure, gas, and capacitance of the variable capacitor can be minimized. This reduces the influence of interference from pressure, gas, etc. compared to plasma ignition, allowing for more accurate Vpp measurement.

[0068] The process controller 91 obtains the Vpp at each position after varying the inter-electrode distance d within the range of 30 mm to 50 mm in 1 mm intervals, and uses this as the reference Vpp. The reference Vpp information is stored in the storage unit 93. Hereinafter, the inter-electrode distance d used to calculate each reference Vpp is referred to as the set value Gap, and the position of the showerhead 16 and mounting stage 2 when the inter-electrode distance d is at the set value is referred to as the reference position.

[0069] The process controller 91 then uses the reference Vpp value at each set value Gap to determine a correlation function representing the relationship between Vpp and the inter-electrode distance d, and stores the determined correlation function in the storage unit 93 .

[0070] For example, the process controller 91 sets the approximate straight line obtained by plotting the reference Vpp and the set value Gap on a two-dimensional plane with Vpp and the distance d between electrodes as the respective dimensions as a correlation line between Vpp and the distance d between electrodes, and calculates a correlation function representing the correlation line. Specifically, the process controller 91 can calculate the correlation function and the correlation line using the approximate formula Vpp = a × distance d between electrodes + b. Here, a is the slope (Slope) in the two-dimensional plane with Vpp and the distance d between electrodes as the respective dimensions, and b is the intercept (Intercept). That is, the correlation function between the distance d between electrodes and Vpp is expressed by the following formula (1).

[0071]

Number 1

[0072]

[0073] After that, the process controller 91 performs plasma etching. Figure 5 The measurement process shown in the figure performs a consumption data acquisition process. This process uses the upper top plate 16b, which has been installed and used in plasma etching, to acquire consumption data. In this case, the process controller 91 acquires consumption data while the wafer is mounted on the mounting table 2. Furthermore, in this case, the process controller 91 allows an idle time of at least one hour before executing the consumption data acquisition process. This stabilizes the temperature within the processing vessel 1.

[0074] When acquiring data during consumption, the shower head 16 and stage 2 are sequentially positioned at reference positions corresponding to the respective set values ​​Gap registered in the measurement process. In each configuration, the process controller 91 applies RF power, which does not generate plasma but causes total reflection, to the stage 2 on which the wafer is mounted, using the measurement process to cause total reflection of the RF. Furthermore, the process controller 91 acquires the Vpp measured when the consumed shower head 16 and stage 2 are placed at each reference position as the consumption Vpp.

[0075] Figure 6 This is a diagram for explaining the outline of calculation of consumption amount. Figure 6The black circle in represents the reference Vpp. Furthermore, the white circle represents the Vpp during consumption. Furthermore, the correlation line 310 is a line showing the correlation between the Vpp obtained from the reference Vpp and the set value Gap and the inter-electrode distance d.

[0076] The process controller 91 obtains the inter-electrode distance d corresponding to the consumption Vpp on the correlation line 310. Next, the process controller 91 calculates the difference between the set value Gap for the measured consumption Vpp and the inter-electrode distance d estimated based on the consumption Vpp, and sets this difference as the consumption amount Δd. Hereinafter, the inter-electrode distance d estimated based on the consumption Vpp is referred to as the estimated Gap.

[0077] For example, the process controller 91 measures the shower head 16 and the mounting table 2, which have consumed the shower head 16, by placing them at the reference position when the set value Gap is 40 mm. The process controller 91 obtains the value represented by point 311 as the Vpp at the time of consumption. Next, the process controller 91 obtains the value represented by point 312 as the inter-electrode distance d corresponding to the Vpp represented by point 311. For example, when the correlation function is expressed as equation (1), the process controller 91 substitutes the Vpp at the time of consumption into the Vpp in equation (1) to obtain the inter-electrode distance d as the estimated Gap.

[0078] Next, the process controller 91 calculates the difference between the value at point 312 and 40 mm as the consumption Δd. That is, the process controller 91 calculates the consumption Δd = estimated Gap - set value Gap as the consumption Δd. Figure 6 As shown, the value of point 312 is expressed as 40+Δd (mm). For example, when the set value Gap is 40 mm and the estimated Gap is 42 mm, the consumption Δd is 42-40=2 mm.

[0079] Next, the process controller 91 calculates the average value of the consumption Δd calculated for each consumption period Vpp. For example, when the consumption Δd corresponding to each set value Gap between 30 mm and 50 mm with an increment of 1 mm is expressed as Δd(30) to Δd(50), the process controller 91 calculates the average value of the consumption Δd using the following equation (2).

[0080]

Number 2

[0081] Δdavg=Average[Δd(30)+Δd(31)+…+Δd(49)+Δd(50)]…(2)

[0082] In the equation (2), Δdavg represents the average value of the consumption Δd. Furthermore, the function represented by Average[] is a function for finding the average value. Furthermore, the process controller 91 sets the average value of the consumption Δd as the consumption of the upper top plate 16b.

[0083] In this embodiment, the consumption of the upper top plate 16b is calculated using the consumption Δd at 1 mm intervals within a range of 30 mm to 50 mm. However, the number of points used depends on the resolution of Vpp. Specifically, a higher resolution of Vpp can further reduce the number of points used to calculate the consumption Δd.

[0084] Next, the process controller 91 compares the calculated consumption threshold of the upper top plate 16b with a predetermined consumption threshold to determine whether the consumption falls within the allowable value. For example, the process controller 91 stores 2 mm as the consumption threshold.

[0085] If the calculated consumption threshold of the upper top plate 16b exceeds the consumption threshold, the process controller 91 determines that the consumption exceeds the allowable value. Furthermore, the process controller 91 notifies the administrator of the plasma processing apparatus 100 of the abnormality by displaying an alarm on the display of the user interface 92, for example. Upon receiving the alarm, the administrator replaces the upper top plate 16b. After replacing the upper top plate 16b, the process controller 91 measures the consumption time Vpp and calculates the consumption. If the calculated consumption falls within the allowable value, the process controller 91 determines that the process is normal. The process controller 91 then waits for instructions from the user interface 92 and, if so, executes plasma etching accordingly.

[0086] In contrast, if the calculated consumption threshold of the upper top plate 16b is less than the consumption threshold, the process controller 91 determines that the consumption is within the allowable value. Furthermore, the process controller 91 calculates the knob control amount based on the calculated consumption of the upper top plate 16b and incorporates this value into the process during plasma etching. The process controller 91 then executes plasma etching using the corrected process.

[0087] In this embodiment, Vpp is measured while the wafer is placed on the mounting table 2 as a protective cover. A measurement process based on the absence of plasma is used for Vpp measurement. However, if plasma is generated, the wafer is placed to prevent ignition of components. Since plasma etching is not actually performed, consumption measurement can be performed without placing the wafer. Regardless of whether the wafer is placed or not, both the baseline Vpp measurement and the consumption Vpp measurement are performed under the same wafer placement conditions.

[0088] Next, refer to Figure 7 A flow of plasma etching processing using consumption measurement performed by the plasma processing apparatus 100 according to this embodiment will be described. Figure 7 This is a flow chart of a plasma etching process performed by a plasma processing apparatus using consumption measurement.

[0089] The process controller 91 acquires reference data including information on a reference Vpp corresponding to each set value Gap, and calculates a correlation function between Vpp and the inter-electrode distance d (step S1 ).

[0090] Thereafter, the process controller 91 executes plasma etching using the plasma processing apparatus 100 (step S2), and determines, for example, whether the accumulated time of the plasma etching process has elapsed for a predetermined time (step S3). If the predetermined time has not elapsed (step S3: No), the process controller 91 returns to step S2 and continues the plasma etching process.

[0091] On the other hand, if the fixed time has elapsed (step S3: Yes), the process controller 91 determines to execute consumption calculation and measures the consumption time Vpp of the currently mounted upper plate 16b using the measurement recipe (step S4).

[0092] Next, the process controller 91 uses the consumption Vpp for the correlation function between Vpp and the inter-electrode distance d to determine an estimated Gap relative to each set Gap. The process controller 91 then subtracts the set Gap from the estimated Gap to calculate the consumption Δd. Furthermore, the process controller 91 calculates the average of the consumption Δd corresponding to each set Gap to calculate the consumption of the currently installed upper plate 16b (step S5).

[0093] Then, the process controller 91 compares the calculated consumption with the consumption threshold value to determine whether the consumption falls within the allowable value (step S6).

[0094] On the other hand, when the consumption amount is within the allowable value (step S6: Yes), the process controller 91 adjusts the control amount of the knob control according to the consumption amount (step S7).

[0095] Then, the process controller 91 returns to step S2 and performs plasma etching using the process in which the control amount of the knob control is corrected.

[0096] When the consumption exceeds the permissible value (step S6 : No), the process controller 91 notifies the manager of an alarm (step S8 ).

[0097] Thereafter, the process controller 91 determines whether to stop the operation of the plasma processing apparatus 100 based on whether an operation stop instruction is received from the user interface 92 (step S9 ).

[0098] If the operation is not stopped (step S9: No), the manager who receives the alarm replaces the upper top plate 16b (step S10). Thereafter, the plasma etching process using consumption measurement performed by the process controller 91 returns to step S4.

[0099] On the other hand, when it is determined that the operation is to be stopped (step S9 : Yes), the process controller 91 stops the operation of the plasma processing apparatus 100 .

[0100] [Simulation of consumption calculation]

[0101] Next, a simulation of consumption calculation will be described. Figure 8 It is a diagram showing the arrangement state of the upper top plate when acquiring reference data. Figure 9 This is a graph showing the correlation straight line obtained based on the reference Vpp. Figure 9 In FIG, the vertical axis represents Vpp, and the horizontal axis represents the distance between electrodes.

[0102] In the benchmark data acquisition process, such as Figure 8 As shown, the unused upper top plate 16b is arranged so that the distance d between electrodes is gradually increased by 1 mm from 30 mm to 50 mm. Figure 8 The figure shows a state where the inter-electrode distance d changes from its narrowest width of 30 mm to its widest width of 50 mm. The inter-electrode distance d at this time is the set value Gap. The position of the top plate 16b in each state is the reference position. Furthermore, the process controller 91 measures the reference Vpp for each inter-electrode distance d.

[0103] When the measured reference Vpp and the set value Gap are marked on a two-dimensional plane, it becomes Figure 9 The white circles in the figure represent points. Figure 9 The approximate straight line of each of the points shown is the correlation line 320. The process controller 91 also stores a correlation function representing the acquired correlation line 320.

[0104] Figure 10 This is a diagram showing the configuration state of the upper top plate when acquiring consumption time data. Figure 11 It is a diagram that marks the state of Vpp during consumption. Figure 11 In FIG, the vertical axis represents Vpp, and the horizontal axis represents the distance between electrodes.

[0105] In the consumption data acquisition process, the upper top plate 16b with a simulated consumption of 2 mm is given as shown in FIG. Figure 10 In this case, the distance d between electrodes is 2 mm longer than the set value Gap. Figure 10 As shown, the upper top plate 16b is arranged so that the inter-electrode distance d varies from a narrowest width of 30+2 mm to a widest width of 50+2 mm. The process controller 91 measures the consumption time Vpp of each inter-electrode distance d.

[0106] When the measured consumption Vpp is marked in a two-dimensional plane, it becomes Figure 11 The black circles in the figure represent the points. Moreover, the process controller 91 uses the consumption time Vpp and the related straight line 320 to calculate the distance d between the electrodes corresponding to each consumption time Vpp as the estimated Gap. Next, the process controller 91 calculates the difference between the set value Gap and the estimated Gap, that is, the consumption Δd. Next, the process controller 91 calculates the average of the consumption Δd corresponding to each set value Gap as the consumption of the upper top plate 16b. Here, the process controller 91 calculates the consumption of the upper top plate 16b to be 1.98mm. Here, the upper top plate 16b in the consumption time data acquisition process is given a simulated consumption of 2mm. In contrast, the consumption estimated by the process controller 91 is 1.98mm, which can be said to be able to estimate the consumption with sufficient accuracy.

[0107] [Consumption calculation results]

[0108] Next, the results of calculating the actual consumption using the plasma processing apparatus 100 according to this embodiment will be described. Figure 12 This figure shows the measurement results when the consumption is measured by changing the state using an unused upper top plate. Figure 13 This graph shows the measurement results when the consumption was measured by changing the state using an upper top plate with a simulated consumption of 2 mm. Figure 14 It is a graph showing the measurement results when metal etching was performed.

[0109] When a silicon wafer is mounted and an unused upper plate 16b is used to measure a reference state, plasma processing is performed multiple times, and then measurement is performed. Figure 12 As shown in FIG, the measurement result of the consumption amount is 0.15 mm. In this case, it can be confirmed that the upper top plate 16 b is consumed by performing the plasma treatment multiple times.

[0110] Afterward, after 30 seconds of treatment under conditions such that sediment was deposited on the inner wall of processing vessel 1, namely, on sediment shields 86 and 87 and upper ceiling 16b, the consumption was measured to be 0.11 mm. This indicates that even if sediment was deposited on the inner wall of processing vessel 1, the effect of the sediment was minimal, ensuring sufficient measurement accuracy.

[0111] Next, when the plasma treatment was performed using the cleaning process and the measurement was performed immediately after the cleaning process was performed once, the consumption was measured to be 0.50 mm. It can be seen that although the consumption of the upper top plate 16b was not much after the cleaning process was performed once, the consumption increased. Since high-output RF was used in the cleaning process, the amount of plasma generated increased, and the interior of the processing container 1 became hot due to the heat input from the plasma. In other words, it can be seen that when the consumption is measured under high temperature, the measurement accuracy becomes poor. This is believed to be because Figure 3 In the plasma processing apparatus 100 shown in FIG, the impedance of the showerhead 16, which is synthesized in addition to the impedance between the showerhead 16 and the mounting table 2, such as the impedance component of the showerhead 16, varies with temperature, causing deviations in the measurement results. Generally speaking, the impedance based on the resistance component is proportional to temperature. Therefore, it can be said that it is preferable to allow sufficient idle time before performing consumption measurement to allow the internal temperature of the processing container 1 to cool down, that is, stabilize at the temperature used for the measurement reference state.

[0112] In addition, if Figure 13 As shown, when the upper top plate 16b, which was designed to be thinner by 2 mm in advance and simulated by consuming 2 mm, was measured, the consumption amount was measured to be 1.98 mm. This measurement accuracy is within the allowable range, and it can be said that sufficient measurement accuracy can be ensured.

[0113] Next, after performing a plasma treatment that, based on past experience, would consume approximately 0.2 to 0.3 mm of the upper top plate 16b, the measured consumption result was 2.30. In this case, using the simulated consumption of 2 mm of the upper top plate 16b as a benchmark, the measured consumption increased by 0.32 mm, resulting in a result that is almost identical to the empirical value. In this case, the measurement accuracy is within the allowable range, ensuring sufficient accuracy. Furthermore, the consumption of 2 mm in this case is an error, and an alarm is issued.

[0114] Next, when the consumption measurement result was measured immediately after cleaning, which had increased by 0.32 mm, the consumption measurement result was 3.08 mm. In this case, using the upper top plate 16b, which had simulated consumption of 2 mm, as a reference, the consumption measurement result increased by 0.78 mm. It can be seen that in this case, too, the temperature increase caused by the cleaning process deteriorated the measurement accuracy.

[0115] Next, when the internal temperature of the processing chamber 1 was returned to a cold state and the focus ring 5 was set to 90°C for measurement, the consumption result was 2.12. In this case, using the simulated consumption of 2 mm of the upper top plate 16b as a reference, the consumption result increased by 0.14 mm. It can be seen that in this case, the accuracy deteriorated due to the temperature increase, but at this temperature, the deterioration in measurement accuracy can be minimized.

[0116] Next, when the wafer was changed from a silicon wafer to a silicon dioxide (SiO2) wafer and measured, the consumption result was 2.29mm. In this case, when the upper plate 16b was simulated to have consumed 2mm as a reference, the consumption result increased by 0.31mm. This is almost equivalent to the consumption result increasing by 0.32mm. Therefore, even if the wafer is replaced, the impact on measurement accuracy is minimal, and it can be said that sufficient measurement accuracy is maintained.

[0117] In addition, if Figure 14 As shown, after measuring the reference state using an unused upper top plate 16b, the following processing was performed to measure the consumption. The measurement was performed after an idle time of one hour.

[0118] When etching five silicon (Si) wafers, the measurement result was 0.14 mm. In this case, etching the silicon wafers resulted in silicon adhering to the inner walls of the processing vessel 1, namely, to the sediment shields 86 and 87, and to the upper ceiling 16b. The measurement result in this case was within the error range of measurement accuracy, demonstrating that even when silicon adhered to the inner walls of the processing vessel 1, the influence of the adhering matter was minimal, allowing accurate measurement.

[0119] Furthermore, when measuring the consumption after etching five titanium (TiN) wafers, the result was 0.05 mm. In this case, etching the titanium wafers resulted in titanium adhering to the inner wall of the processing container 1. The measurement results in this case were within the error range of measurement accuracy, demonstrating that even when titanium adhering to the inner wall of the processing container 1, the influence of the adhering material was minimal, allowing accurate measurement.

[0120] In addition, in the five aluminum (AlO x When the consumption was measured after etching the aluminum wafer, the result was 0.07 mm. In this case, the aluminum wafer was etched, leaving aluminum attached to the inner wall of the processing container 1. The measurement result in this case was within the error range of measurement accuracy, indicating that even when aluminum was attached to the inner wall of the processing container 1, the influence of the attached material was minimal, allowing accurate measurement.

[0121] As described above, the plasma processing apparatus according to this embodiment measures the Vpp when RF power is applied without generating plasma and is totally reflected, and uses the measured Vpp to estimate the inter-electrode distance. This allows for highly accurate measurement of upper electrode consumption without opening the processing container (chamber).

[0122] Furthermore, by automatically measuring the consumption of the upper electrode, it is possible to determine when to replace the upper electrode. Furthermore, for devices that can move the position of the upper electrode to change the distance between electrodes, the consumption measurement results can be used to adjust the distance between electrodes to cope with changes in characteristics.

[0123] (Second embodiment)

[0124] Next, the second embodiment is described. In the first embodiment, the process controller 91 calculates the consumption Δd at multiple points and calculates their average value as the consumption of the upper top plate 16b. If the Vpp measurement is high-resolution, an appropriate approximate straight line can be obtained from two points. The process controller 91 uses two Vpp corresponding to two different inter-electrode distances d to calculate the approximate straight line and can also calculate the consumption of the upper top plate 16b after consumption. High resolution refers to the measurement capability of displaying a large number of digits after the decimal point of the Vpp measurement result. In this case, even for a non-gap-driven plasma processing device 100 with a fixed inter-electrode distance, the consumption can be calculated.

[0125] For example, calculation of consumption using the non-Gap drive type plasma processing apparatus 100 will be described. Hereinafter, the fixed inter-electrode distance used in the plasma processing apparatus 10 according to this embodiment is referred to as a fixed inter-electrode distance. Figure 15 This is a diagram for explaining the measurement of consumption when Vpp is measured at a high resolution. Figure 15 In FIG, the vertical axis represents Vpp, and the horizontal axis represents the distance between electrodes.

[0126] The measurement recipe is stored in the storage unit 93. In this case, various conditions under the fixed inter-electrode distance are registered in the measurement recipe. For example, if the fixed inter-electrode distance in the plasma processing apparatus 100 is 40 mm, then Figure 5 The row with a Gap of 40 mm corresponds to the measurement process in this modification.

[0127] The process controller 91 receives an instruction from the user interface 92 to measure Vpp when the upper plate 16b is unused. In this case, the operator has installed the unused upper plate 16b. The process controller 91 then measures Vpp using the measurement recipe stored in the storage unit 93. For example, if the fixed inter-electrode distance is 40 mm, the process controller 91 obtains Figure 15 The process controller 91 stores the measurement result in the storage unit 93 .

[0128] Next, the process controller 91 receives an instruction from the user interface 92 to measure the Vpp of the upper top plate 16b to which the consumption corresponding to the consumption threshold is assigned. In this case, the upper top plate 16b having the consumption corresponding to the consumption threshold is installed by the operator. For example, if the consumption threshold is 2 mm, the upper top plate 16b having the simulated consumption of 2 mm is installed. Furthermore, the process controller 91 measures the Vpp using the measurement process stored in the storage unit 93. For example, if the fixed electrode distance is 40 mm and the consumption threshold is 2 mm, the process controller 91 obtains Figure 15 The position of the point 402 on the two-dimensional plane in FIG is used as a point indicating the measurement result of Vpp of the upper top plate 16b to which the consumption amount corresponding to the consumption amount threshold value is applied. Thereafter, the process controller 91 stores the measurement result in the storage unit 93.

[0129] Furthermore, the process controller 91 stores the Figure 15 The information of points 401 and 402 in the image is used to generate a correlation line 403 representing the relationship between the inter-electrode distance d and Vpp. In this case, due to the high resolution, an appropriate correlation line 403 is obtained by finding a line connecting the two points. The process controller 91 then stores the correlation function representing the correlation line 403 in the storage unit 93.

[0130] After that, the process controller 91 repeats the plasma etching of the wafer using the newly installed unused upper plate 16b. Moreover, when the fixed period has passed, the process controller 91 performs the consumption time data acquisition process. Thus, the process controller 91 acquires the consumption time Vpp when the upper plate 16b that has been consumed is used. The consumption time Vpp is, for example, Figure 15 The process controller 91 uses the obtained Vpp to calculate the distance d between electrodes using the correlation function stored in the storage unit. The distance d between electrodes is, for example, Figure 15 It is represented by point 405.

[0131] The process controller 91 then calculates the difference between the calculated inter-electrode distance d and the fixed inter-electrode distance and sets it as the consumption Δd. In this embodiment, since the Vpp measurement capability is high resolution, an appropriate consumption can be obtained based on one measured consumption Vpp.

[0132] The process controller 91 then determines whether the consumption of the upper top plate 16b exceeds a consumption threshold. If the consumption of the upper top plate 16b exceeds the consumption threshold, the process controller 91 issues an alarm to the administrator. In contrast, if the consumption of the upper top plate 16b does not exceed the consumption threshold, the process controller 91 continues to perform normal plasma etching of the wafer until the next consumption determination timing. In the plasma processing apparatus 100 of this embodiment, the inter-electrode distance is fixed, so the process controller 91 does not use consumption to calibrate the process during plasma etching.

[0133] Here, in this embodiment, the correlation line is calculated using the Vpp when the upper top plate 16b is in an unused state and the Vpp when the upper top plate 16b is consumed according to the consumption threshold. In other words, with the goal of improving the calculation accuracy of the inter-electrode distance when consumption corresponding to the consumption threshold occurs, the correlation line is obtained using the Vpp when the consumption threshold is applied. However, when measuring the Vpp used to determine the correlation line, Vpp in other states can also be used. For example, if the upper top plate 16b is in a state where the consumption has been determined, the Vpp of each of the two upper top plates 16b with different consumption states can be measured and the correlation line can be obtained using the measured Vpp.

[0134] In addition, in this embodiment, the case of a non-Gap-driven plasma processing apparatus 100 is described as an example, but in a Gap-driven plasma processing apparatus 100 in which the inter-electrode distance can be changed, the Vpp in two states of the upper top plate 16b can be used to obtain a correlation straight line and measure the consumption.

[0135] As described above, the plasma processing apparatus according to this embodiment has the capability to measure Vpp with high resolution. A correlation line is obtained based on two Vpp measurement results when using two upper top plates with different depletion states. Furthermore, the plasma processing apparatus can use the Vpp measurement results of a single position on the upper top plate during depletion to measure consumption. This reduces the processing load for consumption measurement.

[0136] Although one embodiment has been described above, the present invention is not limited to this specific embodiment, and various modifications and changes can be made within the scope of the gist of the present invention described in the claims.

[0137] For example, the plasma processing apparatus 100 may measure Vpp without placing a wafer as long as there is no risk of plasma generation. Furthermore, when acquiring baseline data and consumption data while changing the inter-electrode distance, the plasma processing apparatus 100 may also change the number of data acquisition locations based on the resolution of the Vpp measurement. For example, it is preferable to reduce the number of data acquisition locations when the resolution is high, and to use more data acquisition locations when the resolution is low.

[0138] In the embodiment, the reference state measurement and consumption measurement were performed using the RF power output from the first RF power source 10a. However, the RF power output from the second RF power source 10b may also be used. Furthermore, even if the frequency of the RF power output from the first RF power source 10a differs from the frequency of the RF power output from the second RF power source 10b, the consumption can be measured using each RF power source and the results compared, thereby improving the accuracy of consumption measurement.

[0139] In the embodiment, the first RF power source 10a is connected to the susceptor 2a via the first matching unit 11a, but it may be connected to the shower head 16 as the upper electrode via the first matching unit 11a. In this case, it is possible to use Figure 3 Since it can be represented by such an equivalent circuit, the same measurement can be performed.

[0140] In the embodiment, the measuring device 204 measures and outputs the Vpp (voltage peak-to-peak) value of the supplied RF power, i.e., the RF voltage. However, the measuring device 204 may also measure and output the RF current or the phase difference between the RF voltage and the RF power. Furthermore, the measuring device 204 may output the impedance calculated by dividing the measured RF voltage by the RF current. Furthermore, the measuring device 204 may output the product of the measured RF voltage and RF current, i.e., the RF power at the measurement site. These values ​​are part of the physical quantities related to the supplied RF power.

Claims

1. A plasma processing apparatus comprising a first electrode and a second electrode facing each other, wherein the plasma processing apparatus comprises: a power applying unit for applying radio frequency power to either the first electrode or the second electrode so as not to ignite plasma; a measuring unit that measures a physical quantity related to the radio frequency power applied from the power applying unit; and a calculation unit that uses the measured physical quantity related to the radio frequency power in a correlation function between the inter-electrode distance between the first electrode and the second electrode and the physical quantity related to the radio frequency power to calculate the inter-electrode distance; in, The physical quantity related to the radio frequency power is at least one of radio frequency voltage, radio frequency current, phase difference between radio frequency power and radio frequency current, and impedance. In which, the calculation unit changes the inter-electrode distance between the reference electrode and the second electrode when a reference electrode with a specified thickness is used as the first electrode and measures a physical quantity related to the radio frequency power as a reference, and obtains a correlation function representing the relationship between multiple inter-electrode distances and the physical quantity related to the radio frequency power as a reference.

2. The plasma processing apparatus according to claim 1, wherein The apparatus further includes a notification unit configured to issue an alarm when the inter-electrode distance is equal to or greater than a threshold value.

3. The plasma processing apparatus according to claim 1 or 2, wherein: The physical quantity related to the radio frequency power is measured when the distance between the first electrode and the second electrode is changed multiple times and the first electrode is configured at the position where the reference electrode is configured at the multiple distances between the electrodes. Based on the correlation function, the corresponding distances corresponding to each measured physical quantity related to the radio frequency power are obtained, and the distance between the first electrode and the second electrode is obtained by averaging the differences between each corresponding distance and the multiple distances between the electrodes.

4. The plasma processing apparatus according to claim 1 or 2, wherein: A reference electrode having a specified thickness is used as the first electrode, and a physical quantity related to the first RF power is measured between the reference electrode arranged at a specified position and the second electrode. A consumable electrode having a thickness thinner than the specified thickness is used as the first electrode, and a physical quantity related to the second RF power is measured between the consumable electrode arranged at the specified position and the second electrode, and a correlation function representing the relationship between the physical quantity related to the first RF power and the physical quantity related to the second RF power is obtained.

5. A method for measuring electrode consumption, characterized in that: In a plasma device having a first electrode and a second electrode facing each other, applying radio frequency power to either the first electrode or the second electrode in a manner that does not cause plasma ignition, measuring a physical quantity related to the applied radio frequency power, The inter-electrode distance is obtained by applying the measured physical quantity related to the radio frequency power to a correlation function between the inter-electrode distance between the first electrode and the second electrode and the physical quantity related to the radio frequency power. The physical quantity related to the radio frequency power is at least one of the radio frequency voltage, the radio frequency current, the phase difference between the radio frequency power and the radio frequency current, and the impedance. In which, the inter-electrode distance between the reference electrode and the second electrode is changed when a reference electrode having a specified thickness is used as the first electrode, and a physical quantity related to the radio frequency power as a reference is measured, and a correlation function representing the relationship between multiple inter-electrode distances and the physical quantity related to the radio frequency power as a reference is obtained.

Images