プラズマ処理装置、電源システム、及びプラズマ処理装置の制御方法

Abstract

Control the quantity and quality of radicals and ions. [Solution] A plasma processing apparatus comprising: a stage; a first high-frequency power supply that supplies a bias waveform oscillating at a first frequency, the cycle of the bias waveform lasting a predetermined period, the bias waveform being applied to the stage; a second high-frequency power supply that supplies a source waveform oscillating at a second frequency higher than the first frequency, the second frequency being variable during the cycle of the bias waveform; and a control unit that controls the second high-frequency power supply to change the second frequency during the cycle of the bias waveform so that reflected wave power is reduced and to include a period in which the power level of the source waveform is increased.

Description

[Technical Field]

[0001] The present disclosure relates to a plasma processing apparatus, a processor, a control method, a non-transitory computer-readable recording medium, and a power supply system. [Background technology]

[0002] A technique is known in which, during etching, the high-frequency power applied for attracting ions is synchronized with the on / off of the high-frequency power for generating plasma, thereby causing ions to reach the polycrystalline silicon layer and making the etching rate of the polycrystalline silicon layer uniform (see, for example, Patent Document 1). [Prior art documents] [Patent documents]

[0003] [Patent Document 1] Japanese Patent Application Publication No. 10-64915 Summary of the Invention [Problem to be solved by the invention]

[0004] In Patent Document 1, high frequency powers of two different frequencies, namely, source power, which is high frequency power for generating plasma, and bias power, which is high frequency power for attracting ions, are applied into a processing chamber to control the etching rate.

[0005] The present disclosure provides techniques for controlling the quantity and quality of radicals and ions. [Means for solving the problem]

[0006] According to one aspect of the present disclosure, there is provided a plasma processing apparatus comprising: a stage; a first high frequency power supply that supplies a bias waveform oscillating at a first frequency, where a cycle of the bias waveform lasts for a predetermined period of time, and where the bias waveform is applied to the stage; a second high frequency power supply that supplies a source waveform oscillating at a second frequency higher than the first frequency, where the second frequency is variable during the cycle of the bias waveform; and a controller that controls the second high frequency power supply to control the power level of the source waveform during the cycle of the bias waveform. [Effects of the Invention]

[0007] According to one aspect, the quantity and quality of radicals and ions can be controlled. [Brief explanation of the drawings]

[0008] [Figure 1] FIG. 1 is a diagram showing an example of a plasma processing apparatus according to an embodiment. [Figure 2A] FIG. 2 is a diagram showing an example of the configuration of a control unit according to an embodiment. [Figure 2B] 10A and 10B are diagrams illustrating a case where control is performed using a phase signal of a sensor attached to a power supply system according to an embodiment, or a case where control is performed using a signal synchronized with the high frequency period of bias power. [Figure 3] FIG. 10 is a diagram showing an example of HF supply timing according to the phase within one cycle of LF according to an embodiment. [Figure 4] FIG. 10 is a diagram showing an example of HF supply timing according to the phase within one cycle of LF according to an embodiment. [Figure 5] FIG. 10 is a diagram showing an example of the relationship between the phase within one LF period and the plasma density Ne and the self-bias Vdc according to an embodiment. [Figure 6] FIG. 10 is a diagram showing an example of reflected wave power according to an embodiment. [Figure 7] FIG. 10 is a diagram showing an example of reflected wave power according to an embodiment. [Figure 8] FIG. 10 is a diagram for explaining a control method according to a first modified example of an embodiment. [Figure 9]FIG. 10 is a diagram for explaining a control method according to a second modification of the embodiment. [Figure 10] FIG. 10 is a diagram for explaining a control method according to a third modification of the embodiment. [Figure 11] FIG. 10 is a diagram for explaining a control method according to a fourth modified example of an embodiment. [Figure 12] FIG. 2 is a diagram showing an example of IMD (intermodulation distortion) according to an embodiment. [Figure 13A] 10 is a timing chart showing a control method according to Modification 5-1 of the embodiment. [Figure 13B] 10 is a timing chart showing a control method according to Modification 5-2 of the embodiment. [Figure 13C] 10 is a timing chart showing a control method according to Modification 5-3 of the embodiment. [Figure 13D] 10 is a timing chart showing a control method according to Modification 5-4 of the embodiment. [Figure 14] 10 is a timing chart showing a control method according to a sixth modified example of the embodiment. [Figure 15A] 10 is a timing chart showing a control method according to Modification 7-1 of the embodiment. [Figure 15B] 10 is a timing chart showing a control method according to Modification 7-2 of the embodiment. [Figure 15C] 10 is a timing chart showing a control method according to Modification 7-3 of the embodiment. [Figure 15D] 10 is a timing chart showing a control method according to Modification 7-4 of the embodiment. [Figure 16] FIG. 10 is a diagram showing an example of reflected wave power of source power according to an embodiment. [Figure 17] 13 is a timing chart for explaining a control method according to Modification 8 of the embodiment. [Figure 18] 13 is a timing chart for explaining a control method according to Modification 9 of the embodiment. [Figure 19] 13 is a timing chart for explaining a control method according to a tenth modification of the embodiment. [Figure 20]13 is a timing chart for explaining a control method according to Modification 11 of the embodiment. [Figure 21] 13 is a timing chart showing a control method according to a twelfth modification of the embodiment. [Figure 22] 13 is a timing chart showing a control method according to a thirteenth modification of the embodiment. DETAILED DESCRIPTION OF THE INVENTION

[0009] DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS The present invention will be described in detail below with reference to the accompanying drawings. In this specification and the drawings, substantially identical components are designated by the same reference numerals, and redundant description will be omitted.

[0010] Hereinafter, the frequency (high frequency) of the source power is also referred to as "HF" (High Frequency), and the source power is also referred to as "HF power." Also, the frequency (high frequency) of the bias power, which is lower than the frequency of the source power, is also referred to as "LF" (Low Frequency), and the bias power is also referred to as "LF power."

[0011] [Introduction] When two different frequency high-frequency powers, namely source power for plasma generation and bias power for ion attraction, are applied to a processing chamber, IMD (Intermodulation Distortion) may occur as reflected wave power.

[0012] IMD not only causes mismatching, but also requires a high-frequency power supply with a capacity that can input more power than is actually required to withstand reflections and maintain the plasma. For this reason, optimizing the length of the coaxial cable used in the power supply line of the high-frequency power supply has been a conventional method of reducing the occurrence of IMD.

[0013] However, IMD occurs at the frequency of the sum or difference between the fundamental wave and / or harmonics of the HF power and the fundamental wave and / or harmonics of the LF power. Therefore, while optimizing the length of a coaxial cable can reduce the reflected wave power of high-frequency power at a specific frequency, it cannot eliminate the reflected wave power of other frequencies that arise from the sum or difference between the fundamental wave and / or harmonics of the HF power and the LF power contained in IMD.

[0014] Furthermore, the lower the frequency of the LF power, the closer IMD occurs to the fundamental frequency of the HF power. Therefore, it is conceivable to increase the frequency of the LF power as much as possible to suppress IMD at frequencies close to the fundamental frequency of the HF power. However, in recent years, particularly for high-aspect-ratio etching processes, lowering the frequency of the LF power has been shown to provide better process results. In other words, the deeper a hole with a high aspect ratio is etched, the lower the etch rate becomes. Therefore, the LF power frequency is lowered and the power is increased. This increases the etch rate in high-aspect-ratio etching. However, these process conditions further increase IMD, and the recent trend toward higher and lower LF power has led to an increase in the reflected power of the RF power. The reflected power of the RF power increases, particularly when LF power and HF power are applied to the same electrode.

[0015] For example, Figure 12 shows an example of the reflected wave power generated when HF power of a specified frequency is applied to an electrode to which LF power has also been applied. The intensity of IMD changes periodically in synchronization with the phase of the LF Vpp (Peak to Peak). For example, in the example of Figure 12, when the LF potential is near its maximum positive value, the IMD is almost 0 W, meaning that no reflection occurs. Also, when the LF potential is in the negative range, the IMD is relatively low. When the LF potential exceeds its maximum positive value and then turns negative, the maximum reflected wave power occurs, and the IMD is at its maximum.

[0016] Therefore, the inventors propose a control method for suppressing the occurrence of IMD according to the phase of LF, taking into account the timing of IMD occurrence, and a plasma processing apparatus for executing the control method.Furthermore, the inventors propose a control method for controlling the quantity and quality of radicals and ions by controlling high-frequency power at two different frequencies, LF and HF.

[0017] [Overall configuration of plasma processing device] First, an example of a plasma processing apparatus 1 according to an embodiment will be described with reference to Fig. 1. Fig. 1 is a diagram showing an example of a plasma processing apparatus according to an embodiment.

[0018] The plasma processing apparatus 1 according to one embodiment is a capacitively coupled parallel plate plasma processing apparatus, and includes a cylindrical processing chamber 10 made of, for example, aluminum whose surface is anodized. The processing chamber 10 is grounded.

[0019] A cylindrical support table 14 is disposed at the bottom of the processing vessel 10 via an insulating plate 12 made of ceramics or the like, and a mounting table 16 made of, for example, aluminum is provided on the support table 14. The mounting table 16 constitutes a lower electrode, and a wafer W, which is an example of an object to be processed, is mounted on the mounting table 16 via an electrostatic chuck 18.

[0020] An electrostatic chuck 18 that attracts and holds the wafer W by electrostatic force is provided on the upper surface of the mounting table 16. The electrostatic chuck 18 has a structure in which an electrode 20 made of a conductive film is sandwiched between a pair of insulating layers or insulating sheets. A DC power supply 22 is connected to the electrode 20. A DC voltage output from the DC power supply 22 is applied to the electrode 20. The wafer W is attracted and held by the electrostatic chuck 18 by the electrostatic force generated thereby, such as Coulomb force.

[0021] A conductive edge ring 24 made of, for example, silicon is disposed on the mounting table 16 around the wafer W. The edge ring 24 is also called a focus ring. A cylindrical inner wall member 26 made of, for example, quartz is provided on the side surfaces of the mounting table 16 and the support table 14.

[0022] A coolant chamber 28 having, for example, an annular shape is provided inside the support table 14. A coolant, such as cooling water, at a predetermined temperature is circulated and supplied to the coolant chamber 28 from an externally provided chiller unit via pipes 30a and 30b, and the processing temperature of the wafer W on the mounting table 16 is controlled by the temperature of the coolant. Note that the coolant is an example of a temperature-adjusting medium circulated and supplied through the pipes 30a and 30b, and the temperature-adjusting medium may not only cool the mounting table 16 and the wafer W but also heat them.

[0023] Furthermore, a heat transfer gas, for example, He gas, is supplied from a heat transfer gas supply mechanism to between the upper surface of the electrostatic chuck 18 and the back surface of the wafer W via a gas supply line 32 .

[0024] An upper electrode 34 is provided above the mounting table 16 in parallel to and facing the mounting table 16. A plasma processing space is formed between the upper electrode 34 and the lower electrode. The upper electrode 34 forms a surface that faces the wafer W on the mounting table 16 and comes into contact with the plasma processing space, i.e., a facing surface.

[0025] The upper electrode 34 is supported on the top of the processing chamber 10 via an insulating shielding member 42. The upper electrode 34 includes an electrode plate 36 that faces the mounting table 16 and has numerous gas discharge holes 37, and an electrode support 38 that detachably supports the electrode plate 36 and is made of a conductive material, such as aluminum with an anodized surface. The electrode plate 36 may be made of silicon or SiC, for example. A gas diffusion chamber 40 is provided inside the electrode support 38, and numerous gas flow holes 41 that communicate with the gas discharge holes 37 extend downward from the gas diffusion chamber 40.

[0026] The electrode support 38 is formed with a gas inlet 62 for introducing a processing gas into the gas diffusion chamber 40. The gas inlet 62 is connected to a gas supply pipe 64, which is connected to a processing gas supply source 66. A mass flow controller (MFC) 68 and an open / close valve 70 are provided in the gas supply pipe 64, in that order from the upstream side. Processing gas for etching is supplied from the processing gas supply source 66. The processing gas flows from the gas supply pipe 64 to the gas diffusion chamber 40, passes through the gas flow holes 41, and is then discharged in a shower-like manner from the gas discharge holes 37 into the plasma processing space. In this way, the upper electrode 34 functions as a showerhead for supplying processing gas.

[0027] A variable DC power supply 50 is connected to the upper electrode 34, and a DC voltage from the variable DC power supply 50 is applied to the upper electrode 34. The polarity, current, and voltage of the variable DC power supply 50, as well as the control of the electronic switches that turn the current and voltage on and off, are controlled by a control unit 200.

[0028] A first high frequency power supply 48 is connected to the mounting table 16 via a power feed rod 47 and a matching box 46. The first high frequency power supply 48 applies LF power to the mounting table 16. This causes ions to be attracted to the wafer W on the mounting table 16. The first high frequency power supply 48 outputs LF power with a frequency in the range of 200 kHz to 13.56 MHz. The matching box 46 matches the internal impedance of the first high frequency power supply 48 with the load impedance.

[0029] A second high-frequency power supply 90 is connected to the mounting table 16 via a power feed rod 89 and a matching box 88. The second high-frequency power supply 90 applies HF power to the mounting table 16. The HF frequency is higher than the LF frequency, and the second high-frequency power supply 90 outputs HF power with a frequency of 13.56 MHz or higher. For example, HF power with a higher frequency of 100 MHz may be output in comparison with LF power of 400 kHz. The matching box 88 matches the internal impedance of the second high-frequency power supply 90 with the load impedance. A filter 94 for passing a predetermined high frequency to ground may be connected to the mounting table 16. The HF power supplied from the second high-frequency power supply 90 may be applied to the upper electrode 34.

[0030] An exhaust port 80 is provided at the bottom of the processing vessel 10, and an exhaust device 84 is connected to the exhaust port 80 via an exhaust pipe 82. The exhaust device 84 has a vacuum pump such as a turbomolecular pump, and the inside of the processing vessel 10 can be depressurized to a desired vacuum level. A wafer W loading / unloading port 85 is provided at the sidewall of the processing vessel 10, and the loading / unloading port 85 can be opened and closed by a gate valve 86. A deposit shield 11 is detachably provided along the inner wall of the processing vessel 10 to prevent etching by-products (deposits) from adhering to the processing vessel 10. That is, the deposit shield 11 constitutes the wall of the processing vessel. The deposit shield 11 is also provided on the outer periphery of the inner wall member 26. An exhaust plate 83 is provided between the deposit shield 11 on the wall side of the processing vessel 10 at the bottom and the deposit shield 11 on the inner wall member 26 side. The deposit shield 11 and the exhaust plate 83 can be made of aluminum coated with ceramics such as Y2O3.

[0031] When performing an etching process using the plasma processing apparatus 1 configured as described above, first, the gate valve 86 is opened, and the wafer W to be etched is loaded into the processing chamber 10 through the loading / unloading port 85 and placed on the mounting table 16. Then, a processing gas for etching is supplied from the processing gas supply source 66 to the gas diffusion chamber 40 at a predetermined flow rate, and then supplied into the processing chamber 10 through the gas flow holes 41 and the gas outlet holes 37. The processing chamber 10 is then evacuated using the exhaust device 84, and the internal pressure is set to a set value within a range of, for example, 0.1 to 150 Pa. Various conventional processing gases can be used, and a gas containing a halogen element, such as C4F8 gas, is preferably used. Other gases, such as Ar gas and O2 gas, may also be included.

[0032] With the etching gas introduced into the processing chamber 10 in this manner, HF power is applied to the mounting table 16 from the second high frequency power supply 90. LF power is also applied to the mounting table 16 from the first high frequency power supply 48. A DC voltage is also applied to the upper electrode 34 from the variable DC power supply 50. A DC voltage is also applied to the electrode 20 from the DC power supply 22, and the wafer W is attracted and held on the mounting table 16.

[0033] The processing gas discharged from the gas discharge holes 37 of the upper electrode 34 is dissociated and ionized mainly by the HF power to generate plasma. The surface to be processed of the wafer W is etched by radicals and ions in the plasma. In addition, by applying LF power to the mounting table 16, the ions in the plasma can be controlled, thereby widening the plasma control margin, such as enabling etching of holes with a high aspect ratio.

[0034] The plasma processing apparatus 1 is provided with a control unit 200 that controls the overall operation of the apparatus. The control unit 200 performs a desired plasma process, such as etching, according to a recipe stored in a memory such as a read-only memory (ROM) and a random-access memory (RAM). The recipe contains control information for the apparatus relative to process conditions, such as process time, pressure (gas exhaust), high-frequency power and voltage, various gas flow rates, temperatures inside the processing chamber (upper electrode temperature, sidewall temperature of the processing chamber, wafer W temperature, electrostatic chuck temperature, etc.), and the temperature of the coolant output from the chiller. These programs and recipes indicating the processing conditions may be stored on a hard disk or semiconductor memory. Alternatively, the recipe may be stored on a portable, computer-readable storage medium such as a CD-ROM or DVD, set in a predetermined location, and read out.

[0035] The on / off or high / low of the HF power may be controlled to synchronize with a signal synchronized with the high frequency cycle of the bias power, or with the phase within one cycle of the voltage, current, or electromagnetic field measured in the bias power supply system. For example, the control unit 200 may control the on / off or high / low of the HF power to synchronize with the phase within one cycle of the LF voltage or current. This allows the quantity and quality of ions and radicals to be controlled, and also reduces the occurrence of IMD.

[0036] The bias power feed system refers to the first high-frequency power supply 48 → matching box 46 → power feed rod 47 → mounting table 16 → (plasma) → upper electrode 34 → (ground). The voltage, current, or electromagnetic field measured in the bias power feed system refers to the voltage, current, or electromagnetic field measured at the parts from the first high-frequency power supply 48 through the inside of the matching box 46 and the power feed rod 47 to the mounting table 16, and at the upper electrode 34.

[0037] The state of the signal synchronized with the high frequency cycle of the bias power, or any of the voltage, current, or electromagnetic field measured in the bias power supply system, is also referred to as the “reference electrical state.” The HF power (source power) is controlled so that it is applied alternately between a first state and a second state, which will be described later, in synchronization with the phase within one cycle of the reference electrical state.

[0038] However, when the "reference electrical state" is any of the voltage, current, or electromagnetic field measured in the bias power supply system, it is preferable that the reference electrical state is any of the voltage, current, or electromagnetic field measured in any of the components from the mounting table 16 to the inside of the matching box connected via the power supply rod 47.

[0039] One example of a method for measuring the reference electrical state in a bias power supply system is to measure the voltage, current, or induced magnetic field of each part by placing a voltage probe, a current probe, or a BZ probe (a probe that measures induced magnetic fields) near one of the parts in the bias power supply system.

[0040] For example, Figure 2B(a) shows an example in which the "reference electrical state" is any one of the voltage, current, or electromagnetic field measured in the bias power feed system. In Figure 2B(a), processor 100 receives an HF voltage or current, an LF voltage or current, an HF phase signal, or an LF phase signal from a sensor such as a VI probe attached to the power feed system. Processor 100 alternately applies source power between a first state and a second state in synchronization with the phase within one cycle of the reference electrical state that indicates the input HF voltage or current, LF voltage or current, HF phase signal, or LF phase signal.

[0041] The processor 100 may generate a signal synchronized with the high-frequency period of the bias power output from the first high-frequency power supply 48 without relying on a signal from a sensor. In this case, the state of the signal can be used as the reference electrical state. Furthermore, the step of measuring the reference electrical state in the bias power supply system can be omitted. For example, in FIG. 2B(b), the processor 100 receives an LF phase signal (low-power waveform) or a signal related to bias power information from the first high-frequency power supply 48, and generates a signal synchronized with the high-frequency period of the bias power based on this input signal. The processor 100 outputs the generated signal to the second high-frequency power supply 90. Based on this signal, the second high-frequency power supply 90 alternately applies source power between a first state and a second state.

[0042] The processor 100 may generate a signal synchronized with the high frequency cycle of the bias power, without being based on a signal from the first high frequency power supply 48. In this case, the processor 100 generates a signal having a cycle, for example, as shown by LF in FIG. 3, and also generates an on / off signal synchronized with this signal, for example, as shown by HF in FIG. 3. The processor 100 outputs the generated signal to the first high frequency power supply 48 and the second high frequency power supply 90. The first high frequency power supply 48 outputs bias power based on this signal. The second high frequency power supply 90 alternately applies source power between the first state and the second state based on this signal.

[0043] The mounting table 16 is an example of a first electrode on which the wafer W is mounted. The upper electrode is an example of a second electrode facing the first electrode. The first high frequency power supply 48 is an example of a bias power supply that supplies LF power to the first electrode. The second high frequency power supply 90 is an example of a source power supply that supplies HF power of a higher frequency than the LF power to the first electrode or the second electrode. The control unit 200 is an example of a control unit that controls the bias power supply and the source power supply. The potential of the lower electrode (mounting table 16) to which bias power is applied is also referred to as the electrode potential.

[0044] [Controller configuration] 2A, a specific configuration of the control unit 200 will be described. The control unit 200 includes a processor 100, a signal generating circuit 102, directional couplers 105 and 108, a reflection detector 111, and an oscilloscope 112.

[0045] A directional coupler 105 is connected to the feed line of the first high frequency power supply 48 between the first high frequency power supply 48 and the matching box 46. A directional coupler 108 is connected to the feed line of the second high frequency power supply 90 between the second high frequency power supply 90 and the matching box 88.

[0046] The directional coupler 105 provides a portion of the LF forward wave power (Pf) to the oscilloscope 112. The directional coupler 108 also provides a portion of the HF forward wave power and reflected wave power to the oscilloscope 112.

[0047] In one embodiment, the LF frequency displayed on the oscilloscope 112 is, for example, 400 kHz, and the HF frequency is, for example, 100 MHz, so that the waveform of the LF forward wave power, the waveform of the HF forward wave power, and the waveform of the HF reflected wave power can be observed on the oscilloscope 112.

[0048] In addition, directional coupler 108 separates a certain percentage of the reflected HF wave and provides it to reflection detector 111. Reflection detector 111 is composed of, for example, a spectrum analyzer, a power meter, etc., and measures the amount of IMD (Intermodulation Distortion) occurring at what wavelength and how much of it is being reflected, as well as the power of the reflected wave. IMD refers to the reflected wave power from the plasma side that is generated according to the sum or difference frequency of the LF fundamental wave and / or harmonics and the HF fundamental wave and / or harmonics, which is generated by applying HF power to the upper electrode or lower electrode (the lower electrode in one embodiment) of plasma processing apparatus 1 and applying LF power to the lower electrode.

[0049] The directional coupler 105 provides a portion of the LF traveling wave power to the processor 100. The processor 100 generates a synchronization signal for HF that is synchronized with the LF traveling wave power. For example, the processor 100 may generate the synchronization signal for HF by synchronizing it with the positive timing of the LF traveling wave. Note that instead of the directional coupler 105, an LF waveform detected using a sensor such as a VI probe may be provided to the processor 100.

[0050] The processor 100 provides the created synchronization signal to the signal generating circuit 102. The signal generating circuit 102 generates a control signal synchronized with the LF traveling wave power from the provided synchronization signal and provides the control signal to the second high frequency power supply 90 and the first high frequency power supply 48.

[0051] There are two methods for generating the control signal: When the first high frequency power supply 48 is a general power supply, the directional coupler 105 extracts a portion of the LF voltage or current output from the first high frequency power supply 48 as a waveform and inputs it to the processor 100. However, this is not limiting, and the processor 100 may input a portion of the LF power or the like directly from the first high frequency power supply 48. The processor 100 creates an ON signal with a desired delay and width from the input waveform signal and transmits it to the signal generating circuit 102. The ON signal is an example of a synchronization signal.

[0052] The signal generating circuit 102 sends a command signal to the second high frequency power supply 90 to generate HF power during the ON signal. Depending on the input form of the second high frequency power supply 90, the command signal may be a control signal that generates HF power during the ON signal or the ON signal itself.

[0053] If the first high frequency power supply 48 is an amplifier that amplifies LF power, voltage, or current, the signal generating circuit 102 may extract a portion of the LF power or the like output from the first high frequency power supply 48 as a waveform, and create an ON signal with an arbitrary delay and an arbitrary width from the waveform signal, without using the signal from the directional coupler 105. The signal generating circuit 102 transmits the waveform signal and the ON signal to the second high frequency power supply 90.

[0054] However, the above-described method for generating the control signal is merely an example and is not limiting. As long as a control signal can be generated from a given synchronization signal that controls the HF power to be turned on and off or to alternately apply high and low in synchronization with the phase within one cycle of the reference electrical state (the phase within one cycle of the LF voltage or current, the electrode potential, etc.), other hardware or software can be used, not limited to the circuit of control unit 200 shown in FIG. 2A.

[0055] The amplifier of the first high frequency power supply 48 amplifies the amplitude of the 400 kHz LF modulation signal (AM) and supplies it to the lower electrode, while the amplifier of the second high frequency power supply 90 amplifies the amplitude of the 100 MHz HF modulation signal and supplies it to the lower electrode.

[0056] FIG. 3 shows an example of an LF voltage or current waveform and an HF voltage or current applied in response to the timing when the LF voltage or current is positive. When the electrode potential shown in the second waveform from the bottom is positive, the HF voltage or current is controlled to a positive value (ON). When the electrode potential is negative, the HF voltage or current is controlled to 0 (OFF). Basically, since the electrode potential is determined by the LF voltage or current, the HF voltage or current is turned off when the LF voltage or current is negative, and turned on when the LF voltage or current is positive.

[0057] The processor 100 may generate a synchronization signal that controls the HF power during a time period that includes the timing when the electrode potential is positive. However, the processor 100 is not limited to this, and may generate a synchronization signal that controls the HF power during a short period that includes the timing when the electrode potential is most negative.

[0058] [HF power supply timing] Next, the timing of supplying HF power in one embodiment will be described with reference to Fig. 4. Fig. 4 is a diagram showing an example of the timing of supplying HF power in one embodiment.

[0059] The vertical axis in Figures 4(a) to (c) represents the electrode potential. The electrode potential is almost the same as the wafer potential. The electrode potential is the potential when the LF and HF voltages are superimposed. Here, the Vpp of the LF voltage when the LF frequency is 400 kHz is much greater than the Vpp of the HF voltage when the HF frequency is 100 MHz. Therefore, the electrode potential is basically determined by the LF voltage, and oscillates within the width (amplitude) of the Vpp of the HF voltage.

[0060] The thickness of the sheath above the electrode is also basically determined by the LF voltage. When the LF voltage is negative, the electrode potential becomes more negative than when the LF voltage is positive, due to the so-called self-bias voltage Vdc. When the electrode potential is positive relative to ground potential, it becomes close to the plasma potential, allowing some high-speed electrons to flow into the electrode, and when it is negative relative to ground potential, ions flow in.

[0061] Because the electrode is floating from ground by a blocking capacitor (a matching box in one embodiment), electrons that flow into the electrode do not flow to ground. Therefore, electrons flow into the electrode and accumulate during periods (half cycles) when the electrode surface is at a positive potential relative to the plasma. However, the accumulated electrons cause the electrode surface to become negatively charged, generating a negative bias relative to the plasma. This negative bias causes ions to flow onto the electrode surface. This forms a sheath on the electrode surface.

[0062] Ultimately, the surface of the electrode approaches the plasma potential, and the DC component of the electrode potential when the electrons flowing in at that time are in balance with the ions steadily flowing in due to the negative bias is the self-bias voltage Vdc.

[0063] 3 schematically shows the electrode potential corresponding to the phase of the LF, the plasma potential corresponding to the phase of the LF, the sheath thickness, and the impedance Z. The plasma potential is slightly higher than the highest potential inside the processing vessel 10. Therefore, when the electrode potential is positive, the plasma potential is slightly higher than the electrode potential, and when the electrode potential is negative, the plasma potential is slightly higher than the potential of the wall of the processing vessel 10, which is set to zero.

[0064] When the self-bias voltage Vdc makes the electrode potential negative, the sheath thickness is proportional to the voltage, so when the electrode potential is negative, a large voltage is applied to the electrode, and the sheath thickness becomes thicker. On the other hand, when the electrode potential is positive, a smaller voltage is applied to the electrode than when the electrode potential is negative, so the sheath thickness becomes thinner.

[0065] In one embodiment, LF power and HF power are applied to the mounting table 16 (lower electrode), so the electrode potential shown in FIG. 4 is the potential of the lower electrode. The thickness of the sheath on the mounting table 16 varies depending on the LF phase, with some times being almost flat and others being thick. Therefore, if we consider the sheath as a capacitor, when the sheath is thin, the capacitance of the capacitor increases, and the impedance Z of the sheath decreases relative to impedance Z = 1 / ωC. That is, when the electrode potential is positive, the sheath is thin, so the impedance Z is low and nearly constant. On the other hand, when the electrode potential is negative, the sheath is thick, so the impedance Z is high and varies greatly. Furthermore, the impedance Z is largely determined by the LF voltage. For these reasons, impedance matching of HF power is difficult. In particular, when the electrode potential is negative, i.e., when the LF voltage is negative, the impedance is high and varies greatly, making impedance matching of HF power difficult.

[0066] In response to such fluctuations in impedance Z, matching device 88, which performs impedance matching for HF power, can track up to a maximum frequency of about 1 Hz due to motor operation, but at frequencies higher than this, it becomes difficult to track and match, and matching ends up being performed at one timing of the impedance that changes moment by moment depending on the LF phase. In this state, matching device 88 cannot match at phases other than the one timing at which it matched, so the reflected wave power of IMD is large.

[0067] Therefore, in one embodiment, as shown in Figures 4(a) and 4(b), the HF power is controlled to be on or high when the electrode potential is positive, and the HF power is controlled to be off or low when the electrode potential is negative.

[0068] In one embodiment, when the electrode potential is positive, the impedance Z is almost constant, so supplying HF power at this timing makes it easy to achieve matching. Therefore, the HF power is controlled to ON or High at this timing. On the other hand, when the electrode potential is negative, the impedance is high and changes greatly, so supplying HF power at this timing makes it difficult to achieve matching. Therefore, the supply of HF power is controlled to OFF or Low at this timing. This makes it possible to reduce the occurrence of IMD.

[0069] As shown in Figure 4(b), when the HF power is controlled to High or Low, the HF power is not turned off but kept low when the electrode potential is negative, which makes it possible to suppress a decrease in plasma density compared to when the HF power is controlled to On or Off. Also, by making the HF power applied when the electrode potential is negative smaller than the HF power applied when the electrode potential is positive, it is possible to suppress the occurrence of IMD.

[0070] However, the control method of turning on or high the HF power by coinciding with the timing when the electrode potential is positive is merely an example and is not limited thereto. The HF power may be controlled to be on or high when at least a portion of the phase of the reference electrical state is positive. The HF power may also be controlled to be on or high when at least a portion of the phase of the reference electrical state is negative. That is, the HF power (source power) may have a first state and a second state smaller than the first state, and the period of the first state may include a timing when the phase of the reference electrical state peaks. In this case, the peak may be a positive peak or a negative peak. The period of the first state may include a timing when at least a portion of the phase of the reference electrical state is positive. The period of the first state may include a timing when at least a portion of the phase of the reference electrical state is negative. The HF power may be applied not only as a rectangular wave coinciding with the timing when the phase of the reference electrical state is positive, but also as a substantially rectangular wave including at least either a slow-up on the rising edge or a slow-down on the falling edge. Furthermore, the HF power may be applied at least either at a timing shifted a predetermined time later or a predetermined time earlier than the timing when the phase of the reference electrical state is positive.

[0071] The following is an example of the use of a control method in which the HF power is shifted by a predetermined time from the timing when the phase of the reference electrical state is positive. If HF power is applied only when the phase of the reference electrical state is positive, the ion energy will be reduced. Depending on the type of etching, a process with higher ion energy may be desired. In that case, HF power is applied until the timing when the reference electrical state of the LF changes from positive to negative and the desired amount of ion energy is obtained. This makes it possible to achieve a process with high ion energy.

[0072] Alternatively, the duration of the HF power supply may be adjusted so that the HF power is shortened or lengthened by a predetermined time based on the timing when the phase of the reference electrical state is positive. For example, the HF power may be supplied not only at the timing when the phase of the reference electrical state is positive, but also for a predetermined time before and after that timing.

[0073] HF power may be supplied when the phase of the reference electrical state is negative. However, when the phase of the reference electrical state is negative, the impedance is high and changes over time. Therefore, in this case, it is preferable to control the HF power to be turned on for a shorter duration when the phase of the reference electrical state is negative. For example, it is preferable to adjust the timing and duration of the application of HF power using a circuit with a gate function or delay function. The reflection intensity during one cycle of the reference electrical state may be measured in advance, and based on the measurement results, a circuit with an automatic adjustment function may be used to control the application of HF power at a timing when there is little reflection of LF power.

[0074] For example, as shown in FIG. 4(c), when the electrode potential is negative, the HF power may be turned on or high for a short time period that includes the time when the electrode self-bias Vdc is most negative, and the HF power may be turned off or low during other times. Alternatively, the reflected wave power may be detected in advance and, based on its magnitude, the HF power may be turned off or low during times when the reflected wave power is high and turned on or high during times when the reflected wave power is low. By applying HF power for a short time period that includes the time when the electrode potential is most negative, as shown in FIG. 4(c), strong ion implantation can be achieved in specific etching processes such as HARC (High Aspect Ratio Contact). This can improve the etching rate and the etching profile.

[0075] As described above, according to the control method for the plasma processing apparatus 1 of one embodiment, the HF power is controlled to be on or off or high or low in synchronization with the phase within one cycle of the reference electrical state. This reduces the occurrence of IMD. Furthermore, it is possible to control the ion energy and thereby control the quantity and quality of radicals and ions.

[0076] As shown in Figures 3 and 4(a) to 4(c), the state in which the HF power is controlled to ON or High is an example of the first state, and the state in which the HF power is controlled to OFF or Low is an example of the second state.

[0077] In one embodiment, a method for controlling a plasma processing apparatus 1 includes a first control step of alternately applying a first state and a second state in synchronization with the phase of one cycle of a reference electrical state. The second state only needs to be lower than the first state, and the power in the second state may be 0 or may be a value other than 0 that is lower than the first state.

[0078] [Example of effect] Next, an example of the effect of controlling the HF power on / off or high / low in synchronization with the phase within one cycle of the reference electrical state will be described with reference to Figs. 5 to 7. The graph in Fig. 5 shows an example of the relationship between the LF phase, plasma density Ne, and absolute value of self-bias |Vdc| according to one embodiment. Figs. 6 and 7 show an example of reflected wave power according to one embodiment.

[0079] The graph in Figure 5 shows the results of measurements taken when HF power was applied periodically with a phase change for a time width of approximately 40% of one cycle of the reference electrical state. The vertical axis on the left of the graph shows the plasma density Ne (cm -3 ), and the right vertical axis shows the absolute value of the self-bias |Vdc| (V). When HF power and LF power are superimposed and applied to the lower electrode of the plasma processing apparatus 1, the sheath of the lower electrode fluctuates with the LF period, resulting in a change in impedance Z, which in turn fluctuates the plasma density Ne and the self-bias Vdc.

[0080] When the HF power is turned on when the electrode potential is positive and turned off when the electrode potential is negative (see the upper left diagram in Figure 5), the plasma density Ne is high, and the plasma generation efficiency can be increased, as shown in region a in the lower graph of Figure 5. Furthermore, in region a, the absolute value of the self-bias |Vdc| is low, and the occurrence of IMD can be effectively suppressed.

[0081] Furthermore, if the HF power is turned off when the electrode potential is positive and then turned on for a short period of time that includes the time when the electrode potential is most negative (see the upper right diagram in Figure 5), the plasma density Ne will be medium to high, and the plasma generation efficiency will be medium or higher, as shown in region b in the lower graph in Figure 5. This is because when the electrode potential is negative, a large voltage is applied to the electrode, the sheath becomes thicker, the HF electric field when the HF power is turned on decreases, and the plasma generation efficiency decreases.

[0082] Furthermore, in region b, the absolute value of the self-bias |Vdc| is high, which makes it possible to monochromatize the ion energy, that is, to attract ions with uniform ion energy to the wafer W. Particularly in processes with high aspect ratios, monochromatized high-energy ions can be attracted to the wafer W. At this time, the occurrence of IMD increases, but by applying HF power for a short period of time when the potential of the lower electrode is most negative, the occurrence of IMD can be reduced overall compared to when HF power is applied constantly.

[0083] As described above, according to the plasma processing apparatus 1 of one embodiment, the occurrence of IMD can be reduced by controlling the HF power to ON or High based on the timing when the electrode potential is positive, for example. Also, to address the issue of reduced plasma generation efficiency due to the thickening of the sheath when the electrode potential is negative, the plasma generation efficiency can be increased by applying the HF power when the electrode potential is positive.

[0084] Moreover, monochromatic high-energy ions can be attracted to the wafer W by applying HF power for a short time at the deepest timing when the electrode potential is negative.

[0085] For example, the upper and lower graphs on the left side of Figure 6 and the waveforms on the screens (a) and (b) on the right side of Figure 6 show an example of the detection results by the reflection detector 111 of the control unit 200 and the display results on the oscilloscope 112. The graph on the lower left shows Vpp and Vdc for one LF cycle. The deeper Vdc is negative, the thicker the sheath becomes, and the lower the plasma generation efficiency when HF power is applied. The graph on the upper side shows the HF forward wave power (Pf) and reflected wave power (Pr) versus Vpp and Vdc for one LF cycle.

[0086] An example of the display of oscilloscope 112 in Figure 6(a) shows a waveform A of the LF forward wave power measured when the LF phase is 180° shown in area c, and an amplitude B of the high frequency power on the wafer (i.e., the sum of the LF power and the HF power). The waveform of the HF forward wave power is shown as C, and the waveform of the HF reflected wave power is shown as D. An example of the display of oscilloscope 112 in Figure 6(b) shows a waveform A of the LF forward wave power measured when the LF phase is 0° (=360°) shown in area d, and an amplitude B of the high frequency power on the wafer, the waveform C of the HF forward wave power, and the waveform D of the HF reflected wave power.

[0087] According to this, the reflected wave power in region d is smaller than that in region c. Therefore, it can be seen that IMD can be suppressed by alternately applying HF power between a first state (e.g., an on or high state) and a second state (e.g., an off or low state) using a signal synchronized with the high-frequency period of the bias power so that the LF phase includes 0°, or by synchronizing with the phase within one period of the reference electrical state measured by the bias power supply system. For example, as described above, by switching the HF power to the first state when the electrode potential is positive and to the second state when the electrode potential is negative, control can be performed according to the absolute value |Vdc| of the self-bias, thereby suppressing IMD and improving plasma generation efficiency. Furthermore, by controlling the HF power between the first and second states at any timing according to the electrode potential, high-energy ions can be attracted to the wafer W by utilizing the region where the plasma density Ne is high and the region where the absolute value |Vdc| of the self-bias is large. In this case, applying HF power in a pulsed manner can reduce the overall occurrence of IMD.

[0088] FIG. 7 shows an example of LF Vpp, LF |Vdc|, HF forward wave power (Pf), and HF reflected wave power (Pr). According to this, the HF reflected wave power (Pr) changes by a maximum of approximately five times (approximately 10 W to approximately 50 W) in the phase of one LF voltage cycle. From the above, it can be seen that controlling the HF power in synchronization with the phase within one cycle of the reference electrical state can potentially reduce IMD to approximately one-fifth. It can also be seen that controlling the HF power in synchronization with the phase within one cycle of the reference electrical state can change the LF Vpp and LF |Vdc| within a range of up to approximately 1.6 times their minimum values.

[0089] [Variation] Next, control methods according to Modifications 1 to 4 of an embodiment will be described with reference to Figs. 8 to 11. Fig. 8 is a diagram for explaining a control method according to Modification 1 of an embodiment. Fig. 9 is a diagram for explaining a control method according to Modification 2 of an embodiment. Fig. 10 is a diagram for explaining a control method according to Modification 3 of an embodiment. Fig. 11 is a diagram for explaining a control method according to Modification 4 of an embodiment.

[0090] (Variation 1) In the embodiment described above, if it is desired to pulse-modulate the HF power in synchronization with the phase within one cycle of the reference electrical state (see HF AM modulation in FIG. 3), an HF power supply that pulse-modulates at the same frequency as the LF frequency is required, which may increase costs.

[0091] 8(a), in the plasma processing apparatus 1 according to the first modification, an additional circuit 250 constituting a bypass line is attached to the power supply line or the lower electrode connected to the first high frequency power supply 48 and the second high frequency power supply 90. In the additional circuit 250, a coil 252 and a variable capacitor 251 are connected in series to a power supply rod connected to the lower electrode, and the variable capacitor 251 is connected to the processing chamber 10 and is grounded.

[0092] The additional circuit 250 increases the ratio of the impedance on the processing vessel 10 side to the load impedance on the plasma side. This reduces the significant change in the impedance Z, which is the sum of the impedances on the additional circuit 250 and the processing vessel 10 side, even when the impedance changes compared to when the additional circuit 250 is not present. For example, as shown in FIG. 8(b), the additional circuit 250 reduces the change in the summed impedance Z, thereby further suppressing IMD when HF power is applied in phase with one cycle of the reference electrical state. Furthermore, simply installing the additional circuit 250 allows for a simple and inexpensive IMD suppression mechanism. Note that the additional circuit 250 is preferably inserted on the second RF power source 90 side after branching from the power feed rod, because it reduces the influence of the HF power on the LF power. Installing a filter between the first RF power source 48 and the second RF power source 90 further reduces the influence of the HF power on the LF power, thereby reducing the fluctuation in the summed impedance Z and further suppressing IMD. The additional circuit 250 may include at least one of a coil, a capacitor, and a diode.

[0093] (Variation 2) As shown in FIG. 9( a), in the plasma processing apparatus 1 according to the second modification, an impedance change circuit 300 is attached to the power supply line or the lower electrode connected to the first and second high-frequency power supplies 48 and 90. The impedance change circuit 300 functions to change the impedance so that the combined impedance of the load impedance on the plasma side and the impedance of the impedance change circuit 300 remains constant. Alternatively, the impedance change circuit 300 changes the impedance according to the phase of the LF signal so as to suppress changes in the impedance seen from the matching unit 88. This suppresses the reflected wave power and reduces the occurrence of IMD. The impedance change circuit 300 changes the impedance within one period of the reference electrical state according to the phase (or impedance) of the LF signal, the LF Vdc, the reflected wave power, etc., thereby suppressing IMD.

[0094] An example of the impedance variation circuit 300 is a configuration in which capacitors are arranged in an array and an electronic switch switches the connections of the capacitors. The control unit 200 controls the electronic switch to vary the impedance of the impedance variation circuit 300.

[0095] 9(b), the control unit 200 switches the connection of the capacitor of the impedance change circuit 300 so as to reduce the change in the combined impedance Z of the load impedance on the plasma side and the impedance of the impedance change circuit 300. This improves impedance matching and further suppresses the occurrence of IMD when HF power is applied in synchronization with the phase within one cycle of the reference electrical state.

[0096] The impedance variation circuit 300 may be inserted inside the matching box 88 and integrated with the matching box 88. It is preferable to insert the impedance variation circuit 300 on the second high frequency power supply 90 side after the power feed rod branches, because this makes the HF power less susceptible to the influence of the LF power. If a filter is provided between the first high frequency power supply 48 and the second high frequency power supply 90, the HF power will be even less susceptible to the influence of the LF power, and the fluctuations in the total impedance Z can be reduced, further suppressing the occurrence of IMD.

[0097] (Variation 3) In Modification 3, as shown in FIG. 10(a), an electromagnet 350 is provided above the processing vessel 10. The location of the electromagnet 350 is not limited to the location shown in FIG. 10(a) and may be any part of the processing vessel 10, such as inside the processing vessel 10. The control unit 200 controls the strength of the electromagnet 350 according to the phase (or impedance) of the reference electrical state, the phase of the LF, the electrode potential to which bias power is applied, the LF Vdc, or the HF reflected wave power, etc., to change the characteristics of the magnetic field. For example, as shown in FIG. 10(b), the magnetic field is strengthened when the LF Vdc is negative, which thickens the sheath, and weakened or eliminated when the LF Vdc is positive, which thins the sheath, thereby reducing the change in impedance Z. This further suppresses the occurrence of IMD. The electromagnet 350 may be a multi-pole electromagnet or a fixed magnet, and is an example of a magnetic field generating unit that generates a magnetic field. Control by the electromagnet 350 shown in Modification 3 may be used in combination with control by the additional circuit 250 of Modification 1 or the impedance changing circuit 300 of Modification 2.

[0098] (Variation 4) When the thickness of the sheath changes, the apparent capacitance changes, and the HF resonance frequency changes. The matching device 88 functions to match the sum of all L and C components of the inductance (e.g., the power feed rod) and conductance (e.g., the sheath) in the processing vessel 10, so that they resonate at the HF frequency.

[0099] Therefore, if the sheath thickness changes, the C component changes, and so the reflected wave power increases unless the matching box 88 is re-matched in response to the change in the C component due to the change in the sheath thickness. However, since it takes about one second for the matching box 88 to move the variable capacitor, it may not be able to follow the change in the sheath thickness and may not be able to match accurately.

[0100] Therefore, in Modification 4, the control unit 200 changes the HF frequency by the amount of change in the C component due to the change in the sheath thickness. That is, based on the supply frequency equation f(supply frequency) ∝ 1 / √LC, the HF frequency f is changed in accordance with the change in the C component due to the sheath thickness.

[0101] For example, if the capacitance of the sheath above the electrode is assumed to be C, and the capacitance C changes by a factor of four in response to a change in sheath thickness, the HF frequency will change by a factor of approximately two. This allows for a roughly matched state in response to changes in sheath thickness.

[0102] Furthermore, when the capacitance C changes tenfold in response to a change in the sheath thickness, the HF frequency changes approximately 3.3 times. This allows for a state in which the frequency is roughly matched to the change in the sheath thickness. That is, in Modification 4, as shown in FIG. 11(a), the HF frequency is changed based on the above-described resonance frequency equation so as to match the change in the sheath thickness in response to a change in one cycle of the LF voltage. This allows for a state in which the frequency is roughly matched to the change in the sheath thickness, reduces the power of the reflected HF wave, and suppresses the occurrence of IMD. In Modification 4, the second high-frequency power supply 90 uses a variable-frequency power supply capable of changing the HF frequency. The control shown in Modification 4 may be used in combination with at least one of the additional circuit 250 of Modification 1, the impedance changing circuit 300 of Modification 2, or the electromagnet 350 of Modification 3.

[0103] Furthermore, in all of the above-described embodiments and modified examples, the shift time and delay width may be adjusted by a circuit having a gate function or a delay function based on any of the original signals and measurement signals, such as the LF phase, electrode potential, power supply system potential, Vdc, electrode sheath thickness, plasma light emission, and HF power reflection intensity.

[0104] Furthermore, instead of controlling the timing of applying HF power in synchronization with the phase within one cycle of the LF voltage, as shown in FIG. 11(b), it is also possible to apply pulsed power (hereinafter also referred to as "LF pulse") corresponding to the peak of the LF voltage, and control the timing of applying HF power in response to the LF pulse. That is, for example, an LF pulse corresponding to a 400 kHz LF may be applied on and off, and the HF power may be controlled in pulsed form (HF pulse) in response to this. It is also possible to apply LF pulsed power corresponding to the peak of the phase of the reference electrical state, and control the timing of applying HF power in response to the LF pulse.

[0105] As explained above, the intensity of IMD varies depending on the LF power. Therefore, according to the control method of the plasma processing apparatus 1 according to the above embodiment and each of the above modifications, the occurrence of IMD can be reduced by selecting a timing when the HF reflected wave power is low and applying the HF power at that timing. By reducing the occurrence of IMD, the stability of the process and the plasma processing apparatus 1 can be improved, and the cost of the apparatus can be reduced. Furthermore, it is possible to control the plasma density, self-bias Vdc, etc.

[0106] However, if the time for applying HF power is reduced, there is a concern that the absolute amount of HF power will decrease, resulting in a lower plasma density Ne. Therefore, LF power and HF power may be applied at each of two timings, including when the reference electrical state reaches its peak twice within one cycle. In addition, the method for controlling the application of HF power can be freely changed. Note that the application of HF power is not limited to the lower electrode, and may also be the upper electrode.

[0107] [Control method] As described above, a control method for a parallel plate plasma processing apparatus 1 according to one embodiment includes a step of supplying bias power to a lower electrode on which a wafer W is placed, and a step of supplying source power having a frequency higher than the bias power to a plasma processing space by applying the source power to the lower electrode or the upper electrode. In this control method, the source power has a first state and a second state, and includes a first control step of alternately applying the first state and the second state in synchronization with a signal synchronized with the high frequency cycle of the bias power, or with a phase within one cycle of a reference electrical state indicating any of a voltage, a current, or an electromagnetic field measured in a bias power supply system.

[0108] The above control method can also be performed by a plasma processing apparatus other than a parallel-plate type plasma processing apparatus. The control method for a plasma processing apparatus other than a parallel-plate type plasma processing apparatus includes the steps of supplying bias power to a lower electrode and supplying source power having a frequency higher than that of the bias power to a plasma processing space. The control method also includes a first control step in which the source power has a first state and a second state, and the first state and the second state are alternately applied in synchronization with the phase within one period of the reference electrical state.

[0109] [Variations 5-1 to 5-4] Next, control methods for the plasma processing apparatus 1 according to Modifications 5-1 to 5-4 of an embodiment will be described. In Modifications 5-1 to 5-4, control is performed to intermittently stop the source power and / or bias power. Figures 13A to 13D are timing charts illustrating the control methods according to Modifications 5-1 to 5-4 of an embodiment.

[0110] 13A includes, in addition to the first control step, a second control step in which the source power is intermittently stopped at an LF voltage at a cycle independent of the cycle of the reference electrical state, as an example. The first control step and the second control step are repeatedly executed.

[0111] In Modification 5-1, the LF voltage is applied at the same cycle in the first control step and the second control step, while the source power alternates between the first state and the second state at least once in the first control step, and is intermittently stopped between the first control steps in the second control step.

[0112] In the first and second control steps, the LF frequency may be, for example, 0.1 Hz to 100 Hz. The source power duty ratio (=fourth state / (third state+fourth state)) may be in the range of 1% to 90%.

[0113] The state of the source power synchronized with the period of the reference electrical state in the first control step is an example of a third state. The state of the source power independent of the period of the reference electrical state in the second control step is an example of a fourth state different from the third state.

[0114] 13B includes a third control step in which the bias power is intermittently stopped at a cycle independent of the cycle of the HF voltage or current, in addition to the first control step as in the first control step of the first modification. The state of the bias power in the third control step is an example of the fourth state.

[0115] In Modification 5-2, the first control step and the third control step are repeatedly executed. In Modification 5-2, the source power in the third control step alternates between the first state and the second state in the same cycle as the first control step.

[0116] In the first control step, the LF frequency may be, for example, 0.1 Hz to 100 Hz. The duty ratio of the bias power (=fourth state / (third state+fourth state)) may be within the range of 1% to 90%.

[0117] 13C, in addition to the first control step that is the same as in Modification 5-1, the control method according to Modification 5-3 also includes source power control in the second control step of Modification 5-1 and bias power control in the third control step of Modification 5-2. In other words, the state in Modification 5-3 in which both the source power and bias power are intermittently stopped is an example of the fourth state.

[0118] The period for intermittently stopping the bias power may be synchronized with the period for intermittently stopping the source power. In this case, the periods for intermittently stopping the source power and the bias power may be the same as shown in FIG. 13C, or the source power may be delayed relative to the bias power, or the source power may be delayed relative to the bias power, as shown in FIG. 13D. The period for stopping the source power may be longer or shorter than the period for stopping the bias power.

[0119] [Effects of the control methods according to Modifications 5-1 to 5-4] As described above, the control methods according to Modifications 5-1 to 5-4 can control the quality and quantity of radicals and ions. Specifically, when HF is turned off, most of the ions in the plasma disappear, but radicals have a long lifespan and therefore exist for a certain period of time. Therefore, for example, while HF is turned off, the radicals can be uniformly diffused. Furthermore, while HF is controlled to be off or low, the ratio of ions to radicals in the plasma can be changed. This allows the quantity of radicals and ions to be controlled.

[0120] As the gas dissociates, radicals are generated according to the progress of dissociation. For example, C4F8 gas is converted into C4F8 → C4F7 * →····→CF2, and different radicals (C4F7 * Parameters that promote dissociation include ion energy and reaction time. Therefore, by controlling the timing and duration of application of bias power and source power, the ion energy and / or reaction time can be controlled, and the quality of radicals and ions can be controlled by promoting the generation of radicals suitable for the process.

[0121] Furthermore, while the bias power is off, the ion energy decreases, preventing etching, and by-products deposited at the bottom of holes, etc., can be removed outside the holes and deposited on the mask. Furthermore, while the bias power is off, radicals can be attached to the pattern surface on the wafer W. This allows the radicals attached to the mask to protect the mask and improve the mask selectivity. This accelerates etching, increases the etching rate, and improves the etching profile.

[0122] Although the above describes an example of the effect of intermittently stopping the source power, the effect is not limited to this. For example, plasma can sometimes be generated using bias power, and the same effect can sometimes be obtained when intermittently stopping the bias power. In other words, by intermittently stopping the bias power, the quality and quantity of radicals and ions can be controlled. This increases the etching rate and improves the etching profile.

[0123] 13A to 13D, in the third state, the source power is turned on when the LF Vdc is deeply negative, but this is not limiting and the source power may be turned on when the LF Vdc is positive or at other times. Furthermore, instead of cyclically turning the source power on and off, it may be controlled to be periodically high and low.

[0124] [Variation 6] Next, a control method according to a sixth modification of the embodiment will be described with reference to Fig. 14. Fig. 14 is a timing chart showing the control method according to the sixth modification of the embodiment.

[0125] For example, in the control method according to the sixth modification, an LF pulse is applied to the mounting table 16 as shown in Fig. 14. The positive value of the LF pulse coincides with the positive peak of the LF voltage, and the negative value of the LF pulse coincides with the negative peak of the LF voltage.

[0126] In this case, in the control method according to Modification 6, the first and second HF states are applied alternately in synchronization with the phase within one cycle of the LF pulse, which also makes it possible to control the quantity and quality of radicals and ions.

[0127] Specifically, the source power may be controlled to be off or low during part or all of the positive period of the LF pulse, and to be on or high during part or all of the negative period of the LF pulse. This facilitates control because the LF pulse is binarized and the source power is binarized and controlled accordingly. While FIG. 14 illustrates controlling the HF state shown in FIG. 13A in response to an LF pulse obtained by pulsing the LF voltage of FIG. 13A, this is not limiting. For example, the HF states shown in FIGS. 13B to 13D may be controlled in response to an LF pulse obtained by pulsing the LF voltage of FIG. 13B to 13D.

[0128] [Modifications 7-1 to 7-4] Next, control methods according to Modifications 7-1 to 7-4 of an embodiment will be described with reference to Figs. 15A to 15D. Fig. 15A is a timing chart showing a control method according to Modification 7-1 of an embodiment. Fig. 15B is a timing chart showing a control method according to Modification 7-2 of an embodiment. Fig. 15C is a timing chart showing a control method according to Modification 7-3 of an embodiment. Fig. 15D is a timing chart showing a control method according to Modification 7-4 of an embodiment.

[0129] In the control methods according to Modifications 7-1 and 7-2 shown in FIGS. 15A and 15B, in a first control step, a first state and a second state of source power are alternately applied in synchronization with the phase within one cycle of a reference electrical state, such as an LF voltage or an electrode potential. In Modification 7-1, the first state of source power has two or more states in a stepped manner, synchronized with the timing of the negative electrode potential. In Modification 7-2, the first state of source power smoothly has two or more states, synchronized with the timing of the negative electrode potential. However, the first state of source power may also be synchronized with the timing of the positive electrode potential.

[0130] The control methods according to Modifications 7-3 and 7-4 shown in FIGS. 15C and 15D include a second control step in addition to a first control step, in which the first and second states of source power are alternately applied in synchronization with the phase within one cycle of a reference electrical state, e.g., an LF voltage. In Modification 7-3, the first state of source power has two or more states in a stepped manner synchronized with the positive timing of the electrode potential. In Modification 7-4, the first state of source power smoothly has two or more states synchronized with the positive timing of the electrode potential. However, the first state of source power may also be synchronized with the negative timing of the electrode potential.

[0131] In Modifications 7-1 to 7-4, the source power in the first state is controlled to a plurality of values, thereby enabling more accurate control of the quantity and quality of radicals and ions. Note that in addition to the first control step shown in Figures 15C and 15D, a third control step shown in Figure 13B may be provided instead of the second control step shown in Figures 15C and 15D, or the second and third control steps shown in Figures 13C and 13D may be provided.

[0132] In a control method according to a third modification of the embodiment, the strength of the electromagnet 350 is controlled according to the phase (or impedance) of the reference electrical state, the phase of the LF, the electrode potential to which bias power is applied, the LF Vdc, or the reflected wave power of the HF. This reduces the change in impedance seen from the matching devices 46 and 88, thereby suppressing the occurrence of IMD. In addition, in a control method according to a fourth modification of the embodiment, the HF frequency is changed by the amount of change in the C component due to a change in the sheath thickness. That is, based on the supply frequency equation f(supply frequency) ∝ 1 / √LC, the HF frequency f is changed according to the change in the C component due to the sheath thickness. This achieves a roughly matched state according to the change in the sheath thickness, thereby reducing the reflected wave power of the HF and suppressing the occurrence of IMD. In the fourth modification, the second high-frequency power supply 90 uses a variable-frequency power supply capable of changing the HF frequency.

[0133] When the load impedance is constant, the variable frequency power supply continuously changes the frequency and controls it so that the reflected wave power of the source power is minimized. However, when the source power is controlled in synchronization with the phase within one cycle of the LF voltage or current, the HF load fluctuates significantly and periodically within one cycle of the LF voltage or current. Therefore, the second high frequency power supply 90 needs to change the frequency in accordance with the sheath thickness, which fluctuates significantly and periodically depending on the LF phase within one cycle, or more preferably, in accordance with the impedance corresponding to the sheath thickness.

[0134] For example, FIG. 16 shows an example of the HF reflected wave power (HF-Pr) according to an embodiment. The HF reflected wave power varies depending on the gas type and the LF phase. For example, FIG. 16(a) shows an example of the HF reflected wave power (see B) when argon gas is supplied into the processing vessel 10, 500 W of HF forward wave power (HF-Pf) is applied (see A), and 1000 W of LF power is applied. FIG. 16(b) shows an example of the HF reflected wave power (see B) when SF gas is supplied into the processing vessel 10, 500 W of HF forward wave power is applied (see A), and 1000 W of LF power is applied. Note that C indicates the potential of the wafer mounted on the mounting table 16. The wafer potential is approximately equal to the LF voltage Vpp when the LF frequency is, for example, 400 kHz, and the width (amplitude) of the HF voltage Vpp when the HF frequency is, for example, 100 MHz is added, resulting in oscillation between both potentials.

[0135] Looking at B in Figures 16(a) and (b), it can be seen that the HF reflected wave power output with respect to the phase within one LF cycle differs when argon gas is supplied and when SF6 gas is supplied.

[0136] It is not practical to change the frequency of the HF wave output from the second high-frequency power supply 90 to an optimal frequency in real time depending on the difference in the output of the HF reflected wave power, because it takes time for the second high-frequency power supply 90 to determine the optimal frequency. For example, a typical variable frequency power supply can measure the reflected wave power while shifting the frequency at a maximum of 1 kHz to 10 kHz. On the other hand, for example, for a 400 kHz LF, dividing one cycle into 10 parts results in 4 MHz. To further shift the wavelength within each division, it is necessary to change the wavelength by 40 MHz and synchronize with this to determine the amount of HF reflection and the direction of frequency change in real time. This process is not practical because the operating frequency of a typical variable frequency power supply is insufficient.

[0137] Therefore, in the control method according to Modifications 8 to 11 of the embodiment, when controlling the reflected HF wave power in synchronization with the phase within one cycle of the reference electrical state (the phase of the LF in these modifications), a frequency with less HF reflection is found in each phase obtained by dividing one LF cycle, and the reflected HF wave power is minimized. In Modifications 8 to 11, each phase of the LF refers to each phase obtained by dividing one LF cycle into at least 10 parts. However, the number of divisions into one LF cycle is not limited to this and may be any of 10 to 100 parts. The greater the number of divisions into the phase of one LF cycle, the higher the accuracy of control and the more the reflected HF wave power can be reduced. The control of Modifications 8 to 11 is executed by the processor 100.

[0138] [Variation 8] First, a control method according to Modification 8 of an embodiment will be described with reference to FIG. 17 . FIG. 17 is a timing chart illustrating the control method according to Modification 8 of an embodiment. In the control method according to Modification 8 of an embodiment, the HF frequency is changed in synchronization with each of the phases obtained by dividing one LF cycle into a plurality of parts. The reflected HF power at that time is monitored, and the HF frequency of the second high-frequency power supply 90 is controlled based on the monitoring results so that the reflected HF power in each phase is reduced. A sequence for finding a new HF frequency to be output from the second high-frequency power supply 90 is performed at predetermined time intervals, such as before or during the process, and the HF frequency to be controlled by the second high-frequency power supply 90 is determined based on the results of the sequence.

[0139] The horizontal axis in FIG. 17 represents time, the left vertical axis represents HF forward wave power (HF-Pf) and reflected wave power (HF-Pr), and the right vertical axis represents wafer potential.

[0140] 17 shows the HF reflected wave power (see B) when the frequency controlled by the second high-frequency power supply 90 in the first LF cycle (first cycle of C) is set to an initial frequency. The initial frequency is an arbitrary value, and is set to, for example, one fundamental frequency (e.g., 40 MHz).

[0141] In the second cycle of Fig. 17, the frequency controlled by the second high-frequency power supply 90 is changed from the initial frequency to another frequency within the second cycle of the LF (the second cycle of C). However, in the second cycle, the frequency is not increased or decreased for each phase corresponding to the number of divisions of the phase of one cycle of the LF, but is set to either increase or decrease, and the phase-dependent result of the reflection amount is obtained. In the example of the second cycle of Fig. 17, the frequency (second-cycle frequency) is increased, but it may also be decreased.

[0142] As a result, it can be seen that the second reflected wave power of HF, shown by solid line B in Figure 17, has some areas where reflection is reduced and some areas where reflection is increased depending on the phase, compared to the first reflected wave power of HF, shown by dashed line B. In Figure 17, there are time periods where the reflected wave power of HF is reduced when the LF is in positive phase and increased when the LF is in negative phase. However, the reflected wave power at this time is merely an example and is not limited to this.

[0143] In the third cycle of Fig. 17, the shift direction and shift amount of the frequency controlled by the second high frequency power supply 90 in the third cycle of LF are determined based on the increase or decrease in the reflected wave power of HF for each previous phase. The third frequency shown in Fig. 17 is an example of the determined shift direction and shift amount. Furthermore, the reflected wave power at this time is merely an example and is not limited to this.

[0144] The frequency shift amount and initial shift direction (the direction of the third frequency arrow) controlled by the second high-frequency power supply 90 at one time may be determined based on past data. The frequency shift amount and initial shift direction controlled based on past data may be preset in a recipe and controlled based on the recipe. The past data may be the reflected wave power of the previous HF, the reflected wave power of the HF before the previous, or the reflected wave power of the previous and previous HFs. For example, the frequency shift direction and shift amount for each phase into which one LF cycle is divided may be determined based on the state of the previous reflected wave power so that the reflected wave power of the HF at each phase into which one LF cycle is divided is reduced. The shift direction and shift amount may be determined based on the state of the reflected wave power in the past, in addition to or instead of the state of the previous reflected wave power.

[0145] The frequency controlled by the second high frequency power supply 90 for the third time is shifted in the direction to reduce reflection, and the control timing at this time is, for example, if one LF cycle is divided into 10, the frequency is changed at a time interval obtained by dividing one LF cycle into 10.

[0146] From the fourth time onwards, the second high frequency power supply 90 oscillates the HF at an optimum frequency based on the data from the third time or data previously obtained before. The frequency controlled by the second high frequency power supply 90 is repeated within the allowable frequency range a predetermined number of times or until the reflected HF wave power reaches a predetermined amount, thereby narrowing down the frequencies with low reflected HF wave power in each phase of one LF cycle.

[0147] In a control method according to an eighth modification of an embodiment, the above-described sequence is performed at a specified timing. This makes it possible to minimize the power of the reflected HF wave, which varies depending on the phase of the LF and the gas type. Examples of the specified timing include a synchro pulse period, which is a time interval when one LF period is divided into n (n≧10), a time specified in a recipe, a predetermined time interval, etc.

[0148] [Variation 9] Next, a control method according to a ninth modification of an embodiment will be described with reference to Fig. 18. Fig. 18 is a timing chart illustrating the control method according to the ninth modification of an embodiment. In the control method according to the ninth modification of an embodiment, similar to the eighth modification, the second high frequency power supply 90 controls the HF frequency in synchronization with each phase within one cycle of the LF, and also controls the value of the source power output by the second high frequency power supply 90.

[0149] 18, in a time period around B1 where the reflected HF wave power is small, as shown by B, the second high frequency power supply 90 controls the HF frequency (see FIG. 17) and increases the output of the HF power (source power) as shown by D1. Conversely, in a time period B2 where the reflected HF wave power is large, the second high frequency power supply 90 changes the HF frequency and decreases the output of the source power as shown by D2.

[0150] In a control method according to a ninth modification of an embodiment, the reflected HF power, which varies depending on the phase of the LF and the gas type, can be minimized, and a decrease in plasma density can be suppressed by increasing the source power when the reflected HF power is low. Note that, when the LF has a positive phase, the source power may be controlled within a range from the first source power to the second source power, and when the LF has a negative phase, the source power may be controlled within a range from the third source power to the fourth source power. The range from the first source power to the second source power and the range from the third source power to the fourth source power may be different ranges or the same range, or one range may be included in the other range, or they may partially overlap.

[0151] [Variation 10] Next, a control method according to a tenth modification of an embodiment will be described with reference to Fig. 19. Fig. 19 is a timing chart for explaining the control method according to the tenth modification of an embodiment.

[0152] The horizontal axis in FIG. 19 shows an example of one LF cycle and one RF pulse cycle. One RF pulse cycle may be 0.1 to 100 kHz, or may be longer or shorter than this, but is set to a time period longer than one LF cycle. In a control method according to a tenth modification of an embodiment, the second high-frequency power supply 90 may control the HF frequency in accordance with each phase obtained by dividing one RF pulse cycle into a plurality of parts. The HF frequency and source power may be controlled in accordance with each phase obtained by dividing one RF pulse cycle into a plurality of parts.

[0153] In particular, immediately after the RF pulse is turned on and off, the plasma density Ne and the electrode potential change significantly, and the HF reflected wave power tends to change differently from that observed during the steady state of the RF pulse. Therefore, as shown in Figure 19, immediately after the time V when the RF pulse is turned on for each cycle, it takes time for the LF to rise, and the sheath thickness is thin (i.e., the sheath volume is large). Therefore, immediately after the time V when the RF pulse is turned on, the second high-frequency power supply 90 controls the HF frequency to be high (see E).

[0154] Also, in FIG. 19, the bias power is turned off during the latter half of the RF pulse (see C1). During this time, the HF reflected wave power is constant and close to 0, as shown by B3. That is, when the bias power is turned off and the source power is turned on, the sheath impedance is always constant because the bias power is off. As a result, the HF reflected wave power is constant. Therefore, while the bias power is off, a frequency is determined so that the HF reflected wave power is minimized, and the second high-frequency power supply 90 outputs source power of the determined frequency.

[0155] While the bias power is off, the source power may be controlled to be off or on. For example, as shown in C1 of FIG. 19, while the bias power is off, the HF frequency may be set to E1 and the source power may be controlled to High (or on) in the first half, and the HF frequency may be changed to E2 and the source power may be controlled to Low (or off) in the second half. Note that the periods for intermittently stopping the source power and the bias power may be the same, or the source power may be delayed relative to the bias power, or the source power may be delayed relative to the bias power. The stop time of the source power may be longer or shorter than the stop time of the bias power.

[0156] The control methods according to Modifications 8 to 10 are executed by the processor 100 of Fig. 2A, and a control signal for controlling the HF frequency and HF power is sent to the second high frequency power supply 90 via the signal generating circuit 102. The second high frequency power supply 90 changes the frequency and power of the HF to be output in accordance with the control signal.

[0157] [Variation 11] Next, a control method according to the eleventh modification of the embodiment will be described with reference to Fig. 20. Fig. 20 is a timing chart for explaining the control method according to the eleventh modification of the embodiment.

[0158] As explained in Modification 10, immediately after the RF pulse is turned on, it takes time for the LF to rise, and the sheath thickness is thin (i.e., the sheath capacity is large). Therefore, at the rise of the RF pulse, the plasma density Ne changes significantly, and the impedance fluctuates greatly.

[0159] Therefore, in the control method according to the eleventh modification, the second high-frequency power supply 90 oscillates a composite wave of multiple frequencies at the rising edge of the RF pulse in one LF cycle, i.e., at E3 in Fig. 20. The reflection detector 111 in Fig. 2A detects the power of the reflected HF wave for each of the multiple frequencies. The detected power of the reflected HF wave for each frequency is sent to the processor 100.

[0160] For example, suppose that the second high-frequency power supply 90 can amplify frequencies from 35 MHz to 45 MHz and oscillates composite waves of five frequencies: 41, 42, 43, 44, and 45 MHz. The reflection detector 111 detects the reflected wave power relative to the source power of each of the five frequencies and sends the results to the processor 100. The processor 100 selects the frequency with the smallest reflected wave power.

[0161] For example, if the frequency with the lowest reflected wave power is 41 MHz, the frequency may be determined to be 41 MHz at the timing of the rise of the RF pulse in the next LF cycle, and may be used as the HF frequency output from the second high frequency power supply 90. Also, at E4 in Fig. 20, for example, source power may be output at five frequencies of 39, 40, 41, 42, and 43 MHz, with the 41 MHz frequency at which the previous reflected wave power was lowest being the center.

[0162] This allows the HF frequency output from the second high frequency power supply 90 to quickly reach the target frequency at which the reflected HF wave power is lowest. As a result, the source power output from the second high frequency power supply 90 can be brought to a frequency at which the reflected HF wave power is lowest more quickly, allowing plasma to be ignited more quickly.

[0163] When the processor 100 executes the control method according to the eleventh modification, a control signal for controlling the HF frequency is sent to the second high-frequency power supply 90 via the signal generating circuit 102 based on the HF reflected wave power corresponding to the multiple frequencies detected by the reflection detector 111 of FIG. 2A.

[0164] However, without being limited to this, the second high frequency power supply 90 may have the function of the processor 100. In this case, the HF reflected wave power corresponding to the multiple frequencies detected by the reflection detector 111 is sent directly from the reflection detector 111 to the second high frequency power supply 90.

[0165] In this case, the second high frequency power supply 90 can be realized as a variable frequency power supply having a control unit with the functions of the processor 100. That is, in this case, the control unit of the variable frequency power supply acquires the HF reflected wave power corresponding to each of a plurality of HF frequencies from the reflection detector 111, and selects the frequency with the least reflected wave power based on the acquired HF reflected wave power. Then, the control unit determines to output source power of the selected frequency from the variable frequency power supply. The variable frequency power supply changes the frequency of the source power to the determined frequency and outputs it at a predetermined power. This allows the second high frequency power supply 90 to control the HF frequency and source power to be output without using the processor 100 and the signal generating circuit 102. This allows the second high frequency power supply 90 to execute the control methods of Modifications 8 to 11 without using the processor 100.

[0166] The control method of Modification 11 uses a high frequency wave obtained by combining multiple frequencies, but a high frequency wave obtained by combining multiple frequencies can also be used in the control methods of Modifications 8 to 10. Furthermore, the mixing ratio of the high frequency waves of multiple frequencies can be freely changed or optimized in the control methods of Modifications 8 to 10.

[0167] The control methods of variants 8 to 11 described above provide a control method for a plasma processing apparatus having a first electrode on which a workpiece is placed, comprising the steps of supplying bias power to the first electrode and supplying source power having a frequency higher than that of the bias power to a plasma processing space, wherein the source power has a first state and a second state, and the control method includes a first control step of controlling the first state and the second state to two or more frequencies according to each phase when a phase within one period of a reference electrical state indicating either a voltage, current, or electromagnetic field measured in a power supply system of the bias power is divided into multiple parts.

[0168] [Variation 12] In Modification 12, the first state of the HF voltage is a pulsed voltage value that alternates between two or more voltage values. In the example of Fig. 21, the first state of the HF voltage alternates between a positive voltage value and a zero voltage value. However, this is not limited to this, and two or more voltage values ​​may be alternated, such as three voltage values.

[0169] [Variation 13] The bias power may be a sine wave or pulse waveform power, or may be a tailored waveform power. That is, the bias voltage or current may be a sine wave, an LF pulse waveform, or a tailored waveform as shown in FIG. 22. In the tailored waveform, the bias power may be modulated when the HF shown in FIG. 22 is in the second state, or when the HF is in the first state.

[0170] Similarly, when the first state of HF has two or more voltage values, the waveform of HF may be a tailored waveform shown in FIG. 22 in addition to the waveforms shown in FIGS. 15A to 15D and 21.

[0171] The plasma processing apparatus and control method according to the presently disclosed embodiment should be considered to be illustrative in all respects and not restrictive. The above-described embodiment can be modified and improved in various ways without departing from the spirit and scope of the appended claims. The features described in the above-described embodiments can be configured in other ways as long as they are not inconsistent. Furthermore, the features described in the above-described embodiments can be combined as long as they are not inconsistent.

[0172] The plasma processing apparatus according to the present disclosure can be applied to any type of plasma processing apparatus, including Capacitively Coupled Plasma (CCP), Inductively Coupled Plasma (ICP), Radial Line Slot Antenna (RLSA), Electron Cyclotron Resonance Plasma (ECR), and Helicon Wave Plasma (HWP).

[0173] For example, a control method for a plasma processing apparatus having a first electrode on which a workpiece is placed and a second electrode opposite the first electrode may include a step of supplying bias power to the first electrode and a step of supplying source power having a frequency higher than that of the bias power to a plasma processing space, the source power having a first state and a second state, and a first control step of alternately applying the first state and the second state in synchronization with a phase within one period of the reference electrical state.

[0174] The present invention may also include a control method for a plasma processing apparatus having a first electrode on which a workpiece is placed, the control method comprising the steps of: supplying bias power to the first electrode; and supplying source power having a frequency higher than that of the bias power to a plasma processing space, the source power having a first state and a second state, and including a first control step of alternately applying the first state and the second state in synchronization with a phase within one period of the reference electrical state.

[0175] The step of supplying source power having a frequency higher than that of the bias power to the plasma processing space may be performed by a plasma generation source for generating plasma supplying source power such as a microwave source, a high-frequency power source, or the like to the plasma processing space.

[0176] In this specification, a wafer W has been described as an example of an object to be processed, but the substrate is not limited to this and may be various substrates used in LCDs (Liquid Crystal Displays) and FPDs (Flat Panel Displays), CD substrates, printed circuit boards, etc. [Explanation of symbols]

[0177] 1...Plasma processing device 10...Processing container 16...Placement table (lower electrode) 34…Top electrode 47…Power supply rod 46…matching box 48...First high frequency power source 50...Variable DC power supply 66...Processing gas supply source 84...Exhaust system 88…matching box 89…Power supply rod 90...Second high frequency power supply 91...GND block 100...processor 102...Signal generating circuit 105,108...Directional coupler 111...Reflective detector 112...Oscilloscope 200...Control unit 250...Additional circuit 300...Impedance change circuit 350...Electromagnet.

Images