Plasma treatment device and recipe registration method

Abstract

In the present invention, a control unit is provided with a storage unit, an input unit, a determination unit, and a registration unit. The storage unit stores a recipe for plasma treatment of a substrate. The input unit receives input of a recipe for plasma treatment of a substrate. The determination unit determines whether an upper electrode experiences an abnormal electrical discharge when plasma treatment is executed with the recipe received by the input unit. The registration unit rejects registration of the recipe in the storage unit if the determination unit determines that the upper electrode will experience an abnormal electrical discharge, and registers the recipe in the storage unit if the determination unit determines that the upper electrode will not experience an abnormal electrical discharge.

Description

Plasma processing apparatus and recipe registration method

[0001] The present disclosure relates to a plasma processing apparatus and a recipe registration method.

[0002] The following Patent Document 1 describes a chamber including: "a chamber body; a gas supply unit configured to supply gas to an internal space provided by the chamber body; a stage including a lower electrode and providing a workpiece placement area and disposed within the internal space, the center of the placement area being located on a central axis of the chamber body; an upper electrode disposed above the stage via the internal space; a power supply conductor connected to the upper electrode and extending upward from the upper electrode; a first high frequency power supply configured to generate a first high frequency wave and electrically connected to the upper electrode via the power supply conductor; a second high frequency power supply configured to generate a second high frequency wave having a frequency lower than the frequency of the first high frequency wave and electrically connected to the lower electrode; and an electromagnet disposed above the upper electrode, wherein the electromagnet generates a magnetic field distribution having a horizontal component at a position away from the central axis that is larger than the horizontal component on the central axis. and a grounded conductor extending above the chamber body to cover the upper electrode, the grounded conductor having a cylindrical first portion extending upward from the chamber body, a second portion spaced upward from the upper electrode and extending from the first portion toward the central axis, the second portion providing, together with the first portion, a first space above the upper electrode, and a cylindrical third portion located closer to the central axis than the first portion and extending upward from the second portion, providing therein a second space continuous with the first space, the power supply conductor extending upward through the first space and the second space. The plasma processing apparatus disclosed in this publication states:

[0003] Japanese Patent Application Laid-Open No. 2019-061848

[0004] The present disclosure provides a technique for suppressing the occurrence of problems.

[0005] A plasma processing apparatus according to one aspect of the present disclosure includes a chamber, a substrate support, an upper electrode, an electromagnet, a plasma source, a bias source, a DC power supply, and a control unit. The substrate support is disposed inside the chamber and supports a substrate. The upper electrode faces a substrate support surface of the substrate support. The electromagnet is disposed at the top of the chamber. The plasma source is configured to generate plasma inside the chamber. The bias source is configured to generate a bias inside the chamber. The DC power supply is electrically coupled to the upper electrode. The control unit controls plasma processing on the substrate. The control unit includes a memory unit, an input unit, a determination unit, and a registration unit. The memory unit stores a recipe for plasma processing on the substrate. The input unit accepts input of the recipe for plasma processing on the substrate. The determination unit judges whether abnormal discharge occurs in the upper electrode when plasma processing is performed using the recipe accepted by the input unit. The registration unit refuses to register the recipe in the memory unit when the judgment unit judges that abnormal discharge will occur in the upper electrode, and registers the recipe in the memory unit when the judgment unit judges that abnormal discharge will not occur in the upper electrode.

[0006] According to the present disclosure, the occurrence of problems can be suppressed.

[0007] FIG. 1 is a diagram illustrating an example of the configuration of a capacitively coupled plasma processing apparatus. FIG. 2 is a block diagram illustrating a schematic configuration of a control unit according to an embodiment. FIG. 3A is a diagram illustrating an example of a verification result according to an embodiment. FIG. 3B is a diagram illustrating an example of a verification result according to an embodiment. FIG. 3C is a diagram illustrating an example of a verification result according to an embodiment. FIG. 3D is a diagram illustrating an example of a verification result according to an embodiment. FIG. 4 is a diagram illustrating an example of a coefficient of determination for each regression model according to an embodiment. FIG. 5A is a diagram illustrating an example of values ​​of regression coefficients according to an embodiment. FIG. 5B is a diagram illustrating an example of values ​​of regression coefficients according to an embodiment. FIG. 6 is a graph illustrating an example of regression coefficients of lasso regression according to an embodiment. FIG. 7 is a diagram illustrating an example of a verification result according to an embodiment. FIG. 8 is a diagram illustrating an example of a processing sequence of a generation method for generating a prediction formula according to an embodiment. FIG. 9 is a diagram illustrating an example of a processing sequence of a recipe registration method according to an embodiment.

[0008] Hereinafter, embodiments of the plasma processing apparatus and recipe registration method disclosed in the present application will be described in detail with reference to the drawings. However, the disclosed plasma processing apparatus and recipe registration method are not limited to the embodiments.

[0009] A plasma processing apparatus registers a recipe, which defines the processing conditions for the plasma processing, and generates plasma in a chamber according to the registered recipe to perform the plasma processing. The recipe contains many parameters that can be set as processing conditions. Depending on the combination of recipe parameters, the plasma processing may become unstable, resulting in problems such as part damage. Therefore, technology to prevent such problems is required.

[0010] [Embodiment] [Apparatus Configuration] An example of a plasma processing apparatus according to the present disclosure will be described. In the embodiment described below, a case where the plasma processing apparatus according to the present disclosure is used as a plasma processing system having a system configuration will be described as an example.

[0011] An example of the configuration of a plasma processing system will be described below. Fig. 1 is a diagram illustrating an example of the configuration of a capacitively coupled plasma processing apparatus. The plasma processing system corresponds to the plasma processing apparatus of the present disclosure.

[0012] The plasma processing system includes a capacitively coupled plasma processing apparatus 1 and a controller 100. The capacitively coupled plasma processing apparatus 1 includes a plasma processing chamber 10, a gas supply unit 20, a power supply 30, and an exhaust system 40. The plasma processing chamber 10 corresponds to the chamber of the present disclosure. The plasma processing apparatus 1 also includes a substrate support 11 and a gas inlet. The gas inlet is configured to introduce at least one process gas into the plasma processing chamber 10. The gas inlet includes a showerhead 13. The substrate support 11 is disposed within the plasma processing chamber 10. The showerhead 13 is disposed above the substrate support 11. In one embodiment, the showerhead 13 forms at least a portion of the ceiling of the plasma processing chamber 10. The plasma processing chamber 10 has a plasma processing space 10s defined by the showerhead 13, a sidewall 10a of the plasma processing chamber 10, and the substrate support 11. The plasma processing chamber 10 has at least one gas inlet for supplying at least one processing gas to the plasma processing space 10s and at least one gas outlet for exhausting gas from the plasma processing space 10s. The plasma processing chamber 10 is grounded. The showerhead 13 and the substrate support 11 are electrically insulated from the housing of the plasma processing chamber 10.

[0013] The substrate support 11 includes a main body 111 and a ring assembly 112. The main body 111 has a central region 111a for supporting a substrate W and an annular region 111b for supporting the ring assembly 112. A wafer is an example of a substrate W. The annular region 111b of the main body 111 surrounds the central region 111a of the main body 111 in a plan view. The substrate W is disposed on the central region 111a of the main body 111, and the ring assembly 112 is disposed on the annular region 111b of the main body 111 so as to surround the substrate W on the central region 111a of the main body 111. Therefore, the central region 111a is also called a substrate support surface for supporting the substrate W, and the annular region 111b is also called a ring support surface for supporting the ring assembly 112.

[0014] In one embodiment, the main body 111 includes a base 1110 and an electrostatic chuck 1111. The base 1110 includes a conductive member. The conductive member of the base 1110 can function as a lower electrode. The electrostatic chuck 1111 is disposed on the base 1110. The electrostatic chuck 1111 includes a ceramic member 1111a and an electrostatic electrode 1111b disposed within the ceramic member 1111a. The ceramic member 1111a has a central region 111a. In one embodiment, the ceramic member 1111a also has an annular region 111b. Note that the annular region 111b may also be provided by another member surrounding the electrostatic chuck 1111, such as an annular electrostatic chuck or an annular insulating member. In this case, the ring assembly 112 may be disposed on the annular electrostatic chuck or the annular insulating member, or may be disposed on both the electrostatic chuck 1111 and the annular insulating member. At least one RF / DC electrode coupled to an RF (Radio Frequency) power supply 31 and / or a DC (Direct Current) power supply 32 (described later) may be disposed within the ceramic member 1111a. In this case, the at least one RF / DC electrode functions as a lower electrode. When a bias RF signal and / or a DC signal (described later) is supplied to the at least one RF / DC electrode, the RF / DC electrode is also referred to as a bias electrode. Note that the conductive member of the base 1110 and the at least one RF / DC electrode may function as multiple lower electrodes. Alternatively, the electrostatic electrode 1111b may function as a lower electrode. Therefore, the substrate support 11 includes at least one lower electrode. In this embodiment, the substrate support 11 includes lower electrodes at least below the central region 111a and below the annular region 111b. For example, the electrostatic chuck 1111 includes a central bias electrode 1111c and an annular bias electrode 1111d as lower electrodes. The central bias electrode 1111c is provided inside the central region 111a of the ceramic member 1111a. The central bias electrode 1111c is provided below the central bias electrode 1111c inside the central region 111a of the ceramic member 1111a. The annular bias electrode 1111d is provided annularly inside the annular region 111b of the ceramic member 1111a.It should be noted that another member surrounding the electrostatic chuck 1111, such as the annular electrostatic chuck or an annular insulating member, may have the annular bias electrode 1111d.

[0015] The ring assembly 112 includes one or more annular members. In one embodiment, the one or more annular members include one or more edge rings and at least one cover ring. The edge rings are formed of a conductive or insulating material, and the cover rings are formed of an insulating material.

[0016] The substrate support 11 may also include a temperature adjustment module configured to adjust at least one of the electrostatic chuck 1111, the ring assembly 112, and the substrate to a target temperature. The temperature adjustment module may include a heater, a heat transfer medium, a flow passage 1110a, or a combination thereof. A heat transfer fluid such as brine or a gas flows through the flow passage 1110a. In one embodiment, the flow passage 1110a is formed in the base 1110, and one or more heaters are disposed in the ceramic member 1111a of the electrostatic chuck 1111. The substrate support 11 may also include a heat transfer gas supply configured to supply a heat transfer gas to a gap between the backside of the substrate W and the central region 111a.

[0017] The shower head 13 is configured to introduce at least one processing gas from the gas supply unit 20 into the plasma processing space 10s. The shower head 13 has a main head unit 50 and a side head unit 51. The main head unit 50 is formed in a disk shape and is disposed above the substrate support unit 11. The side head unit 51 is formed in a ring shape and is disposed so as to surround the periphery of the main head unit 50. A dielectric unit 58 is provided between the main head unit 50 and the side head unit 51 so as to surround the periphery of the main head unit 50. The dielectric unit 58 is formed of a dielectric material and electrically isolates the main head unit 50 from the side head unit 51. The main head unit 50 and the side head unit 51 are each grounded.

[0018] The main head unit 50 has an upper electrode 52, which forms an electrode plate, and a support unit 53. The upper electrode 52 is formed in a disk shape. The upper electrode 52 is preferably made of a low-resistance conductor or semiconductor with little Joule heat. Examples of suitable materials for the upper electrode 52 include silicon and SiC. The upper electrode 52 contains silicon. The upper electrode 52 is disposed on the plasma processing space 10s side and faces the substrate support unit 11. The support unit 53 is disposed above the upper electrode 52 and supports the upper electrode 52. The support unit 53 is made of a conductive material, for example, aluminum with an anodized surface.

[0019] The side head portion 51 has an upper electrode 54 forming an electrode plate and a support portion 55. The upper electrode 54 is formed in a ring shape. The upper electrode 54 is also preferably made of a low-resistance conductor or semiconductor with little Joule heat. Examples of suitable materials for the upper electrode 54 include silicon and SiC. The upper electrode 54 is made of a material containing silicon. The support portion 55 is disposed above the upper electrode 54 and supports the upper electrode 54. The support portion 55 is made of a conductive material, for example, aluminum whose surface is anodized. The upper electrode 52 and the upper electrode 54 correspond to the upper electrode in this disclosure. The upper electrode 52 corresponds to the central portion in this disclosure. The upper electrode 54 corresponds to the annular portion in this disclosure.

[0020] The main head section 50 and the side head section 51 are configured to introduce at least one processing gas from the gas supply section 20 into the plasma processing space 10s.

[0021] The support 53 has at least one gas supply port 53a. The upper electrode 52 has at least one gas diffusion chamber 52a and multiple gas inlets 52b. The gas supply port 53a is connected to the gas diffusion chamber 52a. The gas diffusion chamber 52a is formed in a disk shape inside the upper electrode 52. The multiple gas inlets 52b penetrate from the gas diffusion chamber 52a to the lower surface of the upper electrode 52. The processing gas supplied to the gas supply port 53a passes through the gas diffusion chamber 52a and is introduced into the plasma processing chamber 10 from the multiple gas inlets 52b.

[0022] The support portion 55 has at least one gas supply port 55a. The upper electrode 54 has at least one gas diffusion chamber 54a and multiple gas inlets 54b. The gas supply port 55a is connected to the gas diffusion chamber 54a. The gas diffusion chamber 54a is formed in a ring shape along the circumferential direction inside the upper electrode 54. The multiple gas inlets 54b penetrate from the gas diffusion chamber 54a to the lower surface of the upper electrode 54. The processing gas supplied to the gas supply port 55a passes through the gas diffusion chamber 54a and is introduced into the plasma processing chamber 10 from the multiple gas inlets 54b.

[0023] In addition to the shower head 13, the gas introduction part may include one or more side gas injectors (SGIs) attached to one or more openings formed in the side wall 10a.

[0024] The shower head 13 is configured to have a variable temperature. For example, the support part 53 has a flow path 53b formed therein. A heat transfer fluid such as brine or gas flows through the flow path 53b. The shower head 13 can control its temperature by controlling the temperature of the heat transfer fluid flowing through the flow path 53b.

[0025] The gas supply unit 20 may include at least one gas source 21 and at least one flow controller 22. In one embodiment, the gas supply unit 20 is configured to supply at least one process gas from a corresponding gas source 21 to the main head unit 50 and the side head unit 51 via a corresponding flow controller 22. Each flow controller 22 may include, for example, a mass flow controller or a pressure-controlled flow controller. Additionally, the gas supply unit 20 may include one or more flow modulation devices that modulate or pulse the flow rate of the at least one process gas.

[0026] The power supply 30 includes an RF power supply 31 coupled to the plasma processing chamber 10 via at least one impedance matching circuit. The RF power supply 31 is configured to supply at least one RF signal (RF power) to at least one lower electrode and / or at least one upper electrode. This generates a plasma from at least one process gas supplied to the plasma processing space 10s. Therefore, the RF power supply 31 can function as at least a part of a plasma generating unit configured to generate a plasma from one or more process gases in the plasma processing chamber 10. Furthermore, by supplying a bias RF signal to the at least one lower electrode, a bias potential is generated on the substrate W, thereby attracting ion components in the formed plasma to the substrate W.

[0027] In one embodiment, the RF power supply 31 is coupled to the central bias electrode 1111c via at least one impedance matching circuit and is configured to generate a source RF signal (source RF power) for plasma generation. The RF power supply 31 may also be coupled to at least one lower electrode and / or at least one upper electrode via at least one impedance matching circuit and configured to generate a source RF signal for plasma generation. In one embodiment, the source RF signal has a frequency in the range of 10 MHz to 150 MHz. In one embodiment, the RF power supply 31 may be configured to generate multiple source RF signals having different frequencies. The generated one or more source RF signals are supplied to at least one lower electrode and / or at least one upper electrode. The RF power supply 31 corresponds to the plasma source of the present disclosure.

[0028] The power supply 30 may also include a DC power supply 32 coupled to the plasma processing chamber 10. The DC power supply 32 includes a first DC generator 32a and a second DC generator 32b. In one embodiment, the first DC generator 32a is connected to the at least one lower electrode and configured to generate a first DC signal. The generated first DC signal is applied to the at least one lower electrode. In one embodiment, the second DC generator 32b is connected to the at least one upper electrode and configured to generate a second DC signal. The generated second DC signal is applied to the at least one upper electrode.

[0029] In various embodiments, at least one of the first and second DC signals may be pulsed, where a sequence of voltage pulses is applied to the at least one lower electrode and / or the at least one upper electrode, and the voltage pulses may have a pulse waveform that is rectangular, trapezoidal, triangular, or a combination thereof.

[0030] In one embodiment, the first DC signal has a frequency in the range of 100 kHz to 60 MHz. The first DC generator 32a supplies voltage pulses of a predetermined period as the first DC signal to the central bias electrode 1111c and the annular bias electrode 1111d. In this embodiment, the first DC generator 32a is capable of individually changing the voltages of the first DC signals applied to the central bias electrode 1111c and the annular bias electrode 1111d. The first DC generator 32a is configured to supply a pulsed first DC signal (first pulsed DC signal) to the central bias electrode 1111c and a pulsed first DC signal (second pulsed DC signal) to the annular bias electrode 1111d. The voltage pulses may have positive polarity or negative polarity. Furthermore, the sequence of voltage pulses may include one or more positive voltage pulses and one or more negative voltage pulses within one period. The first DC generating section 32a corresponds to the bias source of the present disclosure.

[0031] In one embodiment, the second DC generating unit 32b is connected to at least the upper electrode 52 of the main head unit 50 and the upper electrode 54 of the side head unit 51, and is configured to generate a second DC signal. The generated second DC signal is applied to at least the upper electrode 52 of the main head unit 50 and the upper electrode 54 of the side head unit 51. The second DC generating unit 32b corresponds to a DC power supply of the present disclosure.

[0032] The first and second DC generating units 32 a and 32 b may be provided in addition to the RF power supply 31 .

[0033] The plasma processing apparatus 1 includes an electromagnet 70. The electromagnet 70 is disposed on the upper surface of the plasma processing chamber 10. The electromagnet 70 is formed in an annular shape. The electromagnet 70 includes a yoke 71 and a coil 72. The coil 72 is also formed in an annular shape. The yoke 71 is made of a soft magnetic material and is formed in an annular shape so as to surround the periphery of the coil 72. The electromagnet 70 is disposed so that the central axis of the coil 72 is aligned with the center of the central region 111a of the substrate support 11. Note that, although the example of FIG. 1 illustrates a case in which one annular electromagnet 70 is disposed on the upper surface of the plasma processing chamber 10, this is not limiting. The plasma processing apparatus 1 may also be configured such that multiple electromagnets 70 of different diameters are concentrically disposed on the upper surface of the plasma processing chamber 10.

[0034] A power supply 74 is connected to the coil 72 of the electromagnet 70 via a wire 73. The power supply 74 is configured to supply a current to the electromagnet 70. When a current flows through the coil 72, the electromagnet 70 generates a magnetic field.

[0035] During plasma processing, the plasma processing apparatus 1 supplies a source RF signal for plasma generation from the RF power supply 31 to the upper electrode of the shower head 13, the lower electrode of the base 1110 constituting the substrate support unit 11, or the lower electrode provided on the electrostatic chuck 1111. Furthermore, during plasma processing, the plasma processing apparatus 1 applies a pulsed first DC signal from the first DC generator 32a to the lower electrode of the base 1110. In this embodiment, during plasma processing, the plasma processing apparatus 1 supplies a source RF signal for plasma generation from the RF power supply 31 to the central bias electrode 1111c of the electrostatic chuck 1111. Furthermore, in this embodiment, during plasma processing, the plasma processing apparatus 1 supplies a first pulsed DC signal from the first DC generator 32a to the central bias electrode 1111c of the electrostatic chuck 1111. In addition, in this embodiment, the plasma processing apparatus 1 supplies a second pulse DC signal from the first DC generating unit 32a to the annular bias electrode 1111d of the electrostatic chuck 1111 during plasma processing.

[0036] Furthermore, during plasma processing, the plasma processing apparatus 1 applies a second DC signal from the second DC generating unit 32b to the upper electrode of the shower head 13. In this embodiment, during plasma processing, the plasma processing apparatus 1 applies the second DC signal from the second DC generating unit 32b to the upper electrodes 52, 54 of the shower head 13.

[0037] The exhaust system 40 may be connected to, for example, a gas exhaust port 10e provided at the bottom of the plasma processing chamber 10. The exhaust system 40 may include a pressure regulating valve and a vacuum pump. The pressure in the plasma processing space 10s is regulated by the pressure regulating valve. The vacuum pump may include a turbomolecular pump, a dry pump, or a combination thereof.

[0038] The operation of the plasma processing apparatus 1 configured as described above is controlled comprehensively by a control unit 100. The control unit 100 is, for example, a computer, and controls each unit of the plasma processing apparatus 1. The control unit 100 controls the plasma processing apparatus 1 to perform various processes described in this disclosure.

[0039] [Configuration of control unit 100] Next, the control unit 100 will be described. Fig. 2 is a block diagram showing a schematic configuration of the control unit 100 according to the embodiment. The control unit 100 includes an external I / F (interface) 101, a user I / F 102, a storage unit 103, and a process controller 104.

[0040] The external I / F 101 is capable of communicating with each part of the plasma processing apparatus 1 and inputs and outputs various data.

[0041] The user I / F 102 includes a keyboard through which a process manager inputs commands to manage the plasma processing apparatus 1, a display that visualizes and displays the operating status of the plasma processing apparatus 1, and the like. The user I / F 102 accepts various inputs. For example, the user I / F 102 inputs a recipe, which is a processing condition for plasma processing. The recipe includes many parameters that can be set as processing conditions. For example, the recipe includes the pressure in the plasma processing chamber 10 and the gas conditions supplied from the gas supply unit 20. The gas conditions include, for example, the type and flow rate of gas supplied from the gas supply unit 20 to the main head unit 50, and the type and flow rate of gas supplied from the gas supply unit 20 to the side head unit 51. The recipe also includes, for example, an ion energy control parameter and an electron density control parameter. The ion energy control parameters include, for example, the power of the source RF signal, the voltage of the first pulsed DC signal, the voltage of the second pulsed DC signal, and the voltage of the second DC signal. Examples of the electron density control parameters include the pressure inside the plasma processing chamber 10 and the value of the current flowing from the power supply 74 to the coil 72 of the electromagnet 70. In addition, parameters that can be set in the recipe include the temperature of a heat transfer fluid, such as a coolant, flowing through the flow path 1110a of the base 1110 and the temperature of a heat transfer fluid, such as a coolant, flowing through the flow path 1110a of the base 1110. The user I / F 102 corresponds to the input unit in the present disclosure.

[0042] The storage unit 103 stores a control program (software) and various programs for implementing various processes executed by the plasma processing apparatus 1 under the control of the process controller 104. The storage unit 103 also stores various data used by the programs executed by the process controller 104. For example, the storage unit 103 stores recipe data 103a, teacher data 103b, and prediction model data 103c. The programs and data may be stored on a computer-readable computer recording medium (e.g., a hard disk, an optical disk such as a DVD, a flexible disk, a semiconductor memory, etc.). The programs and data may also be transmitted from other devices, for example, via a dedicated line, for online use.

[0043] The recipe data 103a is data that stores a recipe indicating processing conditions for plasma processing. The teacher data 103b and the prediction model data 103c will be described in detail later.

[0044] The process controller 104 includes a processor such as a CPU (Central Processing Unit) and an MPU (Micro Processing Unit) and controls each component of the plasma processing apparatus 1. The process controller 104 has an internal memory for storing programs and data, reads a control program stored in the storage unit 103, and executes the processing of the read control program. The process controller 104 functions as various processing units when the control program runs. For example, the process controller 104 has the functions of a plasma control unit 104a, a generation unit 104b, a determination unit 104c, and a registration unit 104d. Note that this embodiment will be described assuming that the process controller 104 has the functions of the plasma control unit 104a, the generation unit 104b, the determination unit 104c, and the registration unit 104d. However, the functions of the plasma control unit 104a, the generation unit 104b, the determination unit 104c, and the registration unit 104d may be distributed among multiple controllers.

[0045] The plasma control unit 104a controls each part of the plasma processing apparatus 1 in accordance with the recipe stored in the recipe data 103a, and performs plasma processing on the substrate W.

[0046] A recipe includes many parameters that can be set as processing conditions. Depending on the combination of recipe parameters, the plasma may transition to an unstable region, resulting in problems such as part damage. For example, in the plasma processing apparatus 1, if the heat input from the plasma to the shower head 13 becomes large, problems such as part damage may occur in the shower head 13. For example, in the shower head 13, abnormal discharge may occur in a region where the heat input exceeds a certain threshold, making the process unstable. As a result of the abnormal discharge, thermal stress may cause part cracks in the side head 51. For example, cracks may occur in the upper electrode 54 of the side head 51. However, it was unclear what combination of parameters would cause the plasma to transition to an unstable region.

[0047] Here, an example of the results of verifying the amount of heat input from the plasma to the shower head 13 and the presence or absence of cracks in the side head portion 51 when plasma processing is performed with different recipe parameters will be described. Figures 3A to 3D are diagrams illustrating an example of the verification results according to the embodiment.

[0048] 3A shows a graph in which the horizontal axis represents the voltage Va of the first pulse DC signal applied to the central bias electrode 1111c of the recipe, and the vertical axis represents the heat input q from the plasma to the shower head 13. The voltage increases toward the right on the horizontal axis, with the voltages increasing in the order Va1 to Va3. The heat input q increases toward the top of the vertical axis.

[0049] The amount of heat input q to the shower head 13 was calculated from the temperature rise of the heat transfer fluid that flowed through the flow path 53b of the support part 53 of the main head part 50. For example, if the temperature rise of the heat transfer fluid when it passes through the flow path 53b is defined as the temperature rise Tup, the specific heat of the heat transfer fluid is defined as the specific heat c, and the flow rate of the heat transfer fluid passing through the flow path 53b per unit time is defined as the flow rate V, the amount of heat input q to the shower head 13 can be calculated from the following formula (1):

[0050] q = c・V・Tup (1)

[0051] In Figure 3A, the voltage Va of the first pulse DC signal in the recipe is set to Va1 to Va3, and the heat input q is indicated by dots when conditions other than the recipe voltage Va are changed. Also, Figure 3A uses different dot patterns to indicate "Normal" and "Abnormal" to indicate whether or not part cracking has occurred in the side head portion 51. "Normal" indicates a normal case in which no part cracking has occurred in the side head portion 51. "Abnormal" indicates an abnormal case in which part cracking has occurred in the side head portion 51.

[0052] 3B shows a graph in which the horizontal axis represents the voltage Vd of the second DC signal of the recipe, and the vertical axis represents the heat input q from the plasma to the shower head 13. The voltage decreases toward the right on the horizontal axis, and the voltages Vd1 to Vd3 have increasingly negative polarities in this order. The heat input increases toward the top of the vertical axis.

[0053] In Figure 3B, the voltage Vd of the second DC signal in the recipe is set to Vd1 to Vd3, and the heat input q is indicated by dots when conditions other than the recipe voltage Vd are changed. Also, Figure 3B uses different dot patterns to indicate "Normal" and "Abnormal" to indicate whether or not part cracking has occurred in the side head portion 51. "Normal" indicates a normal case in which no part cracking has occurred in the side head portion 51. "Abnormal" indicates an abnormal case in which part cracking has occurred in the side head portion 51.

[0054] 3C shows a graph in which the horizontal axis represents the side gas flow rate SG of the recipe and the vertical axis represents the heat input q from the plasma to the shower head 13. The side gas flow rate SG is the total flow rate of gas supplied from the gas supply unit 20 to the side head unit 51. The flow rate increases toward the right on the horizontal axis, with the flow rates increasing in the order of SG1 to SG6. The heat input increases toward the top of the vertical axis.

[0055] In Figure 3C, the side gas flow rate SG in the recipe is set to SG1 to SG6, and the heat input q is indicated by dots when conditions other than the recipe flow rate SG are changed. Also, Figure 3C uses different dot patterns to indicate "Normal" and "Abnormal" to indicate whether or not part cracking has occurred in the side head section 51. "Normal" indicates a normal case in which no part cracking has occurred in the side head section 51. "Abnormal" indicates an abnormal case in which part cracking has occurred in the side head section 51.

[0056] FIG. 3D shows a graph with the spatial magnetic field integral value B on the horizontal axis and the heat input q from the plasma to the showerhead 13 on the vertical axis. The spatial magnetic field integral value B is the integral value of the magnetic flux density generated in the plasma processing chamber 10 by the electromagnet 70. In this embodiment, the spatial magnetic field integral value is the integral value of the magnetic flux density in the space above the central region 111a in the plasma processing chamber 10 where the substrate W is placed. A magnetic field is generated in the plasma processing chamber 10 by passing a current through the coil 72 of the electromagnet 70. The magnetic flux density of the magnetic field generated in the plasma processing chamber 10 correlates with the current flowing through the coil 72. Therefore, the spatial magnetic field integral value B correlates with the current flowing through the coil 72 and increases as the current value flowing through the coil 72 increases. The current value to be passed through the coil 72 of the electromagnet 70 is set in the recipe. The spatial magnetic field integral value B can be calculated from the current value to be passed through the coil 72 set in the recipe. On the horizontal axis, the spatial magnetic field integral value increases in the order of B1 to B7 toward the right, and the current value flowing through the coil 72 also increases. On the vertical axis, the amount of heat input increases toward the top.

[0057] 3D shows the heat input q with dots when the spatial magnetic field integral value B is set to B1 to B7 by changing the value of the current flowing through the coil 72 in the recipe and when conditions other than the value of the current flowing through the coil 72 in the recipe are changed. Also, FIG. 3D shows whether or not part cracking has occurred in the side head portion 51 by changing the dot pattern to "Normal" and "Abnormal." "Normal" indicates a normal case in which no part cracking has occurred in the side head portion 51. "Abnormal" indicates an abnormal case in which part cracking has occurred in the side head portion 51.

[0058] 3A to 3D, the heat input threshold Tq, which roughly determines whether part cracking has occurred in the side head portion 51, is indicated by a dotted line. As shown in FIGS. 3A to 3D, the voltage Va of the first pulse DC signal, the voltage Vd of the second DC signal, the side gas flow rate SG, and the spatial magnetic field integral value B each tend to roughly correlate with the occurrence of part cracking even when used as a single parameter. However, the occurrence of part cracking cannot be accurately determined using only one of the voltage Va of the first pulse DC signal, the voltage Vd of the second DC signal, the side gas flow rate SG, and the spatial magnetic field integral value B. Furthermore, it is difficult to determine the heat input q using only one parameter.

[0059] Therefore, a prediction formula for predicting the heat input q is generated from parameters of various processing conditions included in the recipe by performing regression analysis. Several examples of the prediction formula will be described below.

[0060] First, the first prediction formula will be described. The first prediction formula predicts the heat input q from the voltage Va of the first pulse DC signal, the voltage Vd of the second DC signal, the side gas flow rate SG, and the spatial magnetic field integral value B. For example, the heat input q is used as the objective variable. Furthermore, the voltage Va of the first pulse DC signal, the voltage Vd of the second DC signal, the side gas flow rate SG, and the spatial magnetic field integral value B are used as explanatory variables. In generating the first prediction formula, a regression formula is used to model the relationship between the objective variable and the explanatory variables. For example, the regression formula is set to the following formula (2), which adds the linear term, square term, and cross term of the voltage Va, the voltage Vd, the side gas flow rate SG, and the spatial magnetic field integral value B.

[0061] q = a １ ×Va+a ２ ×Vd+a ３ ×SG+a ４ ×B + a ５ ×Va ２ +a ６ ×Vd ２ +a ７ ×SG ２ +a ８ ×B ２ +a ９ ×Va ×Vd+a １０ ×Va×SG+a１１ ×Va×B +a １２ ×Vd×SG+a １３ ×Vd××B+a １４ ×SG×B ... (2)

[0062] Here, a １ ~a １４ are the regression coefficients, respectively.

[0063] The regression coefficient is a １ ~a ４ The term is a linear term obtained by adding Va, Vd, SG, and B. The regression coefficient is a ５ ~a ８ The term is a squared term obtained by squaring Va, Vd, SG, and B and adding them together. The regression coefficient is a ９ ~a １４ The term is a cross term obtained by multiplying Va, Vd, SG, and B together and then adding them together.

[0064] In generating the first prediction formula, teacher data is prepared that stores at least the values ​​of the voltage Va, the voltage Vd, the side gas flow rate SG, the spatial magnetic field integral value B, and the heat input q when plasma processing is performed using various recipes including the recipes used in the verification of Figures 3A to 3D. Then, in generating the first prediction formula, the teacher data is used to obtain the regression coefficient a for equation (2) by regression analysis. １ ~a １４ Calculate.

[0065] There are various regression models for regression analysis. Therefore, for equation (2), the regression coefficient a is calculated using various regression models using training data. １ ~a １４ In the training data, the voltage Va, voltage Vd, flow rate SG, spatial magnetic field integral value B, and heat input q were each standardized (average 0, variance 1) as preprocessing. Then, for each regression model, the calculated regression coefficient a １ ~a １４The coefficient of determination R was calculated for equation (2) using the following regression models: ordinary least squares regression (OLS), ridge regression, lasso regression (LASSO), elastic net regression, principal component regression (PCR), partial least squares regression (PLS), decision tree (DT), random forest regression, k-nearest neighbor algorithm (k-NN), support vector regression (SVR), and Gaussian process regression (GPR).

[0066] 4 is a diagram showing an example of the coefficient of determination R2 for each regression model according to the embodiment. In FIG. 4, the coefficient of determination R2 (R2 score) for each regression model in Equation (2) for the heat input q from the plasma to the shower head 13 is shown as CP heat input Test. Also, in FIG. 4, the coefficient of determination R2 calculated using a similar method for the heat input q from the plasma to the main body 111 is shown as Lower heat input Test.

[0067] 4, the least squares method (OLS) has the highest accuracy for the heat input q from the plasma to the shower head 13. On the other hand, the lasso regression (LASSO) has the highest accuracy for the heat input q from the plasma to the main body 111.

[0068] 5A and 5B are graphs showing the regression coefficient a １ ~a １４ 5A is a diagram showing an example of the value of the regression coefficient a in Equation (2) using the ordinary least squares method (OLS). １ ~a １４ FIG. 5B shows an example of the regression coefficient a of Equation (2) calculated using LASSO. １ ~a １４5A and 5B show an example of the regression coefficient a １ ~a １４ Corresponding to the regression coefficient a １ ~a １４ The term is shown.

[0069] In the ordinary least squares (OLS) method, the regression coefficient a of the term Va １ Although is a negative value, it is unlikely that it will be negative. Ordinary least squares (OLS) has high regression accuracy but is difficult to interpret physically. In this embodiment, lasso regression, which is easy to explain physically, is adopted, and negative regression coefficients are set to 0.

[0070] FIG. 6 shows the regression coefficient a of the Lasso regression according to the embodiment. １ ~a １４ 6 is a graph showing an example of the regression coefficient a １ ~a １４ The heat input q from the plasma to the shower head 13 is expressed as a regression coefficient a ５ , a ８ , a １３ It can be seen that the value of the term (Va) is relatively large, and the voltage Va of the first pulse DC signal, the voltage Vd of the second DC signal, and the spatial magnetic field integral value B are dominant.

[0071] In generating the first prediction formula, the regression coefficient a of the Lasso regression １ ~a １４ The regression coefficient a of the Lasso regression is generated as a prediction formula. １ ~a １４ The prediction formula of formula (2) to which the above formula is applied has a large coefficient of determination R2 of 0.9 or more, and can determine the heat input q with sufficiently high accuracy.

[0072] Next, the second prediction formula will be described. The second prediction formula predicts the heat input q from the voltage Va of the first pulsed DC signal, the voltage Vd of the second DC signal, the side gas flow rate SG, the spatial magnetic field integral value B, the power PW of the source RF signal, the voltage Vb of the second pulsed DC signal, the main gas flow rate MG, the chamber pressure P, and the fluid temperature BT. The main gas flow rate MG is the total flow rate of gas supplied from the gas supply unit 20 to the main head unit 50. The chamber pressure P is the pressure within the plasma processing chamber 10. The fluid temperature BT is the temperature of the heat transfer fluid flowing through the flow path 1110a of the base 1110. For example, the heat input q is used as the objective variable. Furthermore, the voltage Va of the first pulsed DC signal, the voltage Vd of the second DC signal, the side gas flow rate SG, the spatial magnetic field integral value B, the power PW of the source RF signal, the voltage Vb of the second pulsed DC signal, the main gas flow rate MG, the chamber pressure P, and the fluid temperature BT are used as explanatory variables. In generating the second prediction equation, the relationship between the objective variable and the explanatory variables is also modeled using a regression equation. For example, the regression equation may be the following equation (3), which adds the linear terms, square terms, and cross terms of the voltage Va, the voltage Vd, the side gas flow rate SG, the spatial magnetic field integral value B, the power PW, the voltage Vb, the main gas flow rate MG, the chamber pressure P, and the fluid temperature BT.

[0073] q = b １ ×Va+b ２ ×Vd+b ３ ×SG+b ４ ×B+b ５ ×PW + b ６ ×Vb+b ７ ×MG+b ８ ×P+b ９ ×BT+a １０ ×Va ２ +a １１ ×Vd ２ +a １２ ×SG ２ +a １３ ×B ２ +b １４ ×P.W. ２ +b １５ ×Vb ２ +b １６ ×MG ２ +b １７ ×P ２ +b １８ ×BT ２+b １９ ×Va ×Vd+b ２０ ×Va×SG+b ２１ ×Va×B +b ２２ ×Va ×PW+b ２３ ×Va ×Vb + b ２４ ×Va×MG +b ２５ ×Va ×P+b ２６ ×Va×BT+b ２７ ×Vd×SG +b ２８ ×Vd×B+b ２９ ×Vd×PW+b ３０ ×Vd×Vb +b ３１ ×Vd×MG+b ３２ ×Vd×P+b ３３ ×Vd×BT +b ３４ ×SG×B+b ３５ x SG x PW + b ３６ ×SG ×Vb +b ３７ ×SG ×MG + b ３８ ×SG×P+b ３９ ×SG×BT +b ４０ ×B×PW+b ４１ ×B×Vb+b ４２ ×B×MG +b ４３ ×B×P+b ４４ ×B×BT+b ４５ ×PW×Vb +b ４６ ×PW×MG+b ４７ ×PW ×P+b ４８ ×PW×BT +b ４９ ×Vb×MG+b ５０ ×Vb ×P+b ５１ ×Vb×BT +b ５２ ×MG ×P+b ５３ ×MG×BT+b ５４ ×P×BT ... (3)

[0074] Here, b １ ~b ５４ are the regression coefficients, respectively.

[0075] The regression coefficient is b １ ~b ９ The term is a linear term obtained by adding Va, Vd, SG, B, PW, Vb, MG, P, and BT. The regression coefficient is b １０ ~b１８ The term is a squared term obtained by squaring and adding Va, Vd, SG, B, PW, Vb, MG, P, and BT. The regression coefficient is b １９ ~b ５４ The term is a cross term obtained by multiplying Va, Vd, SG, B, PW, Vb, MG, P, and BT with each other and then adding them together.

[0076] In generating the second prediction formula, training data is prepared that stores values ​​of the voltage Va, voltage Vd, side gas flow rate SG, spatial magnetic field integral value B, power PW, voltage Vb, main gas flow rate MG, chamber pressure P, fluid temperature BT, and heat input q when plasma processing is performed using various recipes. Then, in generating the second prediction formula, the training data is used to calculate the regression coefficient b １ ~b ５４ In the teacher data, the voltage Va, voltage Vd, side gas flow rate SG, spatial magnetic field integral value B, power PW, voltage Vb, main gas flow rate MG, chamber pressure P, fluid temperature BT, and heat input q were each standardized (average 0, variance 1) as preprocessing.

[0077] In generating the second prediction formula, the regression coefficient b of the Lasso regression １ ~b ５４ The regression coefficient b of the Lasso regression is generated as a prediction formula. １ ~b ５４ The prediction formula of formula (3) to which the above formula is applied has a large coefficient of determination R2 of 0.9 or more, and can determine the heat input q with sufficiently high accuracy.

[0078] Here, an example of the results of a comparison between the heat input q from the plasma to the shower head 13 when plasma processing is performed with different recipe parameters and the heat input q predicted from the recipe using the prediction formula of Equation (3) to which the regression coefficients of Lasso regression are applied will be described. Figure 7 is a diagram illustrating an example of the verification results according to the embodiment.

[0079] In FIG. 7, the heat input q from the plasma to the shower head 13 is shown on the horizontal axis as the heat input Aq, and the regression coefficient b of the Lasso regression is shown on the horizontal axis as the heat input Aq. １ ~b ５４The graph shows the heat input q predicted by the prediction formula of Equation (3) applied to the heat input Pq as the vertical axis. The heat input Aq increases toward the right on the horizontal axis. The heat input Pq increases toward the top on the vertical axis.

[0080] 7, for each recipe, dots are shown at positions corresponding to the heat input Aq and heat input Pq of the recipe. The dots for each recipe are distributed in a manner close to a direct proportion with a slope of 1, because the heat input Pq predicted by the prediction formula is close to the heat input Aq. This shows that the prediction formula (3) can be regressed with high accuracy.

[0081] FIG. 7 also uses different dot patterns to indicate whether or not part cracking has occurred in the side head 51, labeling it "Normal" and "Abnormal." "Normal" indicates a normal case in which no part cracking has occurred in the side head 51. "Abnormal" indicates an abnormal case in which part cracking has occurred in the side head 51. In FIG. 7, the threshold value Tq, which indicates the heat input amount used to determine whether or not part cracking has occurred in the side head 51, is shown by dotted lines on the vertical and horizontal axes. In FIG. 7, the threshold value Tq is set to 5 kW. As can be seen from FIG. 7, if the heat input amount from the plasma to the shower head 13 is equal to or less than the threshold value Tq, part cracking will not occur in the side head 51. The threshold value Tq may be set to a value within a range in which part cracking will not occur. For example, the threshold value Tq may be set to any value between 4 kW and 5 kW.

[0082] Returning to Fig. 2, the storage unit 103 stores teacher data 103b used to generate the prediction formulas. For example, the storage unit 103 stores teacher data prepared when generating the first and second prediction formulas as the teacher data 103b.

[0083] The generation unit 104b generates a prediction formula for predicting the heat input q. For example, the generation unit 104b performs regression analysis using the teacher data 103b to generate a prediction formula for predicting the heat input q from parameters of various processing conditions included in the recipe. For example, the generation unit 104b performs lasso regression using the teacher data 103b to generate the prediction formula. For example, the generation unit 104b generates the first prediction formula or the second prediction formula described above, and generates the prediction formula of Equation (2) or Equation (3).

[0084] The generating unit 104b stores the generated prediction formula data in the storage unit 103 as prediction model data 103c.

[0085] The plasma processing apparatus 1 receives a registration of a recipe as a processing condition for plasma processing from the user I / F 102 .

[0086] When the determination unit 104c receives a recipe input via the user I / F 102, it determines whether abnormal discharge will occur in the shower head 13 when plasma processing is performed using the received recipe. For example, the determination unit 104c uses a prediction formula stored in the prediction model data 103c in the storage unit 103 to calculate the heat input q to the shower head 13 when plasma processing is performed using the received recipe. For example, when the prediction formula is Formula (2), the determination unit 104c calculates the heat input q from the voltage Va, voltage Vd, side gas flow rate SG, and spatial magnetic field integral value B of the received recipe. When the prediction formula is Formula (3), the determination unit 104c calculates the heat input q from the voltage Va, voltage Vd, side gas flow rate SG, spatial magnetic field integral value B, power PW, voltage Vb, main gas flow rate MG, chamber pressure P, and fluid temperature BT of the received recipe. When the judgment unit 104c uses the formula (2) or (3) as the prediction formula, the judgment unit 104c performs preprocessing on the recipe values ​​used to calculate the heat input q in the same manner as the teacher data, and calculates the heat input q using the standardized values.

[0087] The determination unit 104c determines whether or not abnormal discharge will occur in the shower head 13 based on the calculated heat input q. The determination unit 104c compares the calculated heat input q with a threshold value Tq. If the heat input q is smaller than the threshold value Tq, the determination unit 104c determines that abnormal discharge will not occur in the shower head 13, and if the heat input q is equal to or greater than the threshold value Tq, the determination unit 104c determines that abnormal discharge will occur in the shower head 13.

[0088] When the determination unit 104c determines that no abnormal discharge occurs in the shower head 13, the registration unit 104d registers the recipe in the storage unit 103. For example, the registration unit 104d registers the inputted recipe in the recipe data 103a stored in the storage unit 103.

[0089] On the other hand, when the determination unit 104c determines that abnormal discharge will occur in the shower head 13, the registration unit 104d refuses to register the recipe in the storage unit 103. For example, the registration unit 104d displays a message refusing to register the recipe whose input has been accepted on the user I / F 102. For example, the registration unit 104d displays on the user I / F 102 the calculated heat input q and a message indicating that the recipe whose input has been accepted may cause problems such as part damage in the shower head 13.

[0090] [Processing Flow] The processing flow performed by the plasma processing system according to the embodiment will be described. First, the processing flow for generating a prediction formula will be described. Fig. 8 is a diagram illustrating an example of the processing order of the generation method for generating a prediction formula according to the embodiment.

[0091] The generating unit 104b generates a prediction formula for predicting the heat input q (step S10). For example, the generating unit 104b performs a regression analysis using the training data 103b to generate a prediction formula for predicting the heat input q from parameters of various processing conditions included in the recipe.

[0092] The generating unit 104b stores the generated prediction formula data as prediction model data 103c in the storage unit 103 (step S11), and ends the process.

[0093] Next, a process flow for registering a recipe will be described. Fig. 9 is a diagram illustrating an example of a process sequence for registering a recipe according to an embodiment. The process in Fig. 9 is executed when a recipe is input via the user I / F 102.

[0094] When the determination unit 104c receives a recipe input via the user I / F 102, it determines whether abnormal discharge will occur in the shower head 13 when plasma processing is performed using the received recipe. For example, the determination unit 104c uses a prediction formula stored in the prediction model data 103c in the storage unit 103 to calculate the heat input q to the shower head 13 when plasma processing is performed using the received recipe (step S20). The determination unit 104c determines whether abnormal discharge will occur in the shower head 13 based on the calculated heat input q (step S21). The determination unit 104c compares the calculated heat input q with a threshold Tq. If the heat input q is smaller than the threshold Tq, the determination unit 104c determines that abnormal discharge will not occur in the shower head 13. If the heat input q is equal to or greater than the threshold Tq, the determination unit 104c determines that abnormal discharge will occur in the shower head 13.

[0095] If the determining unit 104c determines that no abnormal discharge occurs in the shower head 13 (step S21: No), the registering unit 104d registers the recipe in the storage unit 103 (step S22), and ends the process.

[0096] On the other hand, if the determining unit 104c determines that abnormal discharge occurs in the shower head 13 (step S21: Yes), the registering unit 104d refuses to register the recipe in the storage unit 103 (step S23), and ends the process.

[0097] As a result, the plasma processing system according to the embodiment can prevent the registration of recipes that may cause abnormal discharge in the shower head 13, thereby preventing the occurrence of problems.

[0098] In this way, the plasma processing system according to the embodiment can prevent trouble by refusing to register a recipe if there is a risk of abnormal discharge occurring in the shower head 13 for the recipe that has been input.

[0099] In the above embodiment, the shower head 13 has been described as having a configuration in which one side head 51 is arranged concentrically around the main head 50. However, this is not limiting. The shower head 13 may have a configuration in which multiple side heads 51 are arranged concentrically around the main head 50. The shower head 13 may have multiple upper electrodes 54 arranged concentrically around the upper electrode 52. The shower head 13 may also have a configuration in which a dielectric portion is provided between the main head 50 and each of the multiple side heads 51 to electrically isolate the main head 50 from the multiple side heads 51.

[0100] As described above, the plasma processing system (plasma processing apparatus) according to the embodiment includes the plasma processing chamber 10, the substrate support 11, the upper electrodes (upper electrode 52 and upper electrode 54), the electromagnet 70, the RF power supply 31 (plasma source), the first DC generator 32a (bias source), the second DC generator 32b (DC power supply), and the controller 100. The substrate support 11 is disposed inside the plasma processing chamber 10 and supports a substrate W. The upper electrode faces a substrate support surface of the substrate support 11. The electromagnet 70 is disposed above the plasma processing chamber 10. The RF power supply 31 is configured to generate plasma inside the plasma processing chamber 10. The first DC generator 32a is configured to generate a bias inside the plasma processing chamber 10. The second DC generator 32b is electrically coupled to the upper electrode. The controller 100 controls the plasma processing on the substrate W. The control unit 100 includes a storage unit 103, a user I / F 102 (input unit), a determination unit 104c, and a registration unit 104d. The storage unit 103 stores a recipe for plasma processing on a substrate W. The plasma control unit 104a performs plasma processing on a substrate W in accordance with the recipe stored in the storage unit 103. The user I / F 102 accepts input of the recipe for plasma processing on a substrate W. The determination unit 104c determines whether abnormal discharge will occur in the upper electrode when plasma processing is performed using the recipe accepted by the user I / F 102. The registration unit 104d refuses to register the recipe in the storage unit 103 if the determination unit 104c determines that abnormal discharge will occur in the upper electrode, and registers the recipe in the storage unit 103 if the determination unit 104c determines that abnormal discharge will not occur in the upper electrode. This allows the plasma processing system according to the embodiment to prevent problems from occurring.

[0101] The upper electrode has a central portion (upper electrode 52) and one or more annular portions (upper electrode 54) arranged concentrically around the central portion. The determining unit 104c determines whether abnormal discharge occurs in the annular portion. This allows the plasma processing system according to the embodiment to prevent problems from occurring in the annular portion.

[0102] The upper electrode further includes a dielectric portion 58 that electrically isolates the adjacent central portion and one or more annular portions, thereby preventing the central portion from being affected by abnormal discharges occurring in the annular portions.

[0103] Furthermore, the upper electrode contains silicon, which allows the plasma processing system according to the embodiment to suppress heat generation in the upper electrode.

[0104] Furthermore, the determining unit 104c calculates the amount of heat input to the upper electrode when plasma processing is performed using a recipe received by the user I / F 102 using a predetermined calculation formula (e.g., formulas (2) and (3)), and determines that no abnormal discharge will occur in the upper electrode if the amount of heat input is smaller than a predetermined threshold Tq. This allows the plasma processing system according to the embodiment to prevent problems from occurring.

[0105] The formula is a formula for calculating the amount of heat input to the upper electrode from the set values ​​of the ion energy control parameter and the electron density control parameter in the recipe received by the user I / F 102. The set value of the ion energy control parameter is at least one of the set values ​​of the first DC generator 32a or the second DC generator 32b. The set value of the electron density control parameter is at least one of the set value of the RF power supply 31, the pressure of the plasma processing chamber 10, or the current value of the electromagnet 70. As a result, the formula can calculate the amount of heat input to the upper electrode.

[0106] Furthermore, the threshold value Tq is any value between 4 kW and 5 kW. This allows the plasma processing system according to the embodiment to suppress the occurrence of abnormal discharge at the upper electrode.

[0107] The calculation formula is obtained by regression analysis. The calculation formula is obtained by lasso regression using data that stores the heat input amount to the upper electrode when plasma processing is performed and the recipe for the plasma processing. This allows the calculation formula to accurately calculate the heat input amount to the upper electrode.

[0108] The first DC generator 32a applies a periodically pulsed voltage to the substrate support 11. The second DC generator 32b applies a DC voltage to the upper electrode. The calculation formula includes a linear term obtained by adding the voltage value of the pulsed voltage, the voltage value of the DC voltage applied to the upper electrode, the gas flow rate supplied into the plasma processing chamber 10, and the integral value of the magnetic field generated in the plasma processing chamber 10 by the electromagnet 70, a square term obtained by squaring and adding these terms, and a cross term obtained by multiplying and adding these terms. This allows the calculation formula to accurately calculate the amount of heat input to the upper electrode.

[0109] Although the embodiments have been described above, the disclosed embodiments should be considered to be illustrative in all respects and not restrictive. Indeed, the above-described embodiments may be embodied in various forms. Furthermore, the above-described embodiments may be omitted, substituted, or modified in various forms without departing from the scope and spirit of the claims.

[0110] For example, in the above embodiment, the plasma etching process is performed on a semiconductor wafer as the substrate W, but the present invention is not limited to this.

[0111] It should be noted that the disclosed embodiments are illustrative in all respects and should not be considered limiting. Indeed, the above-described embodiments may be embodied in various forms. Furthermore, the above-described embodiments may be omitted, substituted, or modified in various forms without departing from the scope and spirit of the appended claims.

[0112] In addition, the following supplementary notes are disclosed regarding the above-described embodiment.

[0113] a control unit that controls plasma processing on the substrate, wherein the control unit comprises: a memory unit that stores a recipe for plasma processing on the substrate; an input unit that accepts input of the recipe for plasma processing on the substrate; a determination unit that determines whether abnormal discharge will occur in the upper electrode when plasma processing is performed using the recipe accepted by the input unit; and a registration unit that refuses to register the recipe in the memory unit if the determination unit determines that abnormal discharge will occur in the upper electrode, and registers the recipe in the memory unit if the determination unit determines that abnormal discharge will not occur in the upper electrode.

[0114] (Supplementary Note 2) The plasma processing apparatus according to Supplementary Note 1, wherein the upper electrode has a center portion and one or more annular portions arranged concentrically outside the center portion, and the determination unit determines whether or not abnormal discharge occurs in the annular portions.

[0115] (Supplementary Note 3) The plasma processing apparatus according to Supplementary Note 2, wherein the upper electrode further includes a dielectric portion that electrically separates the central portion and one or more annular portions adjacent to each other.

[0116] (Supplementary Note 4) The plasma processing apparatus according to any one of Supplementary Notes 1 to 3, wherein the upper electrode contains silicon.

[0117] (Appendix 5) The plasma processing apparatus according to any one of Appendices 1 to 4, wherein the determination unit calculates the amount of heat input to the upper electrode when plasma processing is performed using a recipe received by the input unit using a predetermined formula, and determines that no abnormal discharge will occur in the upper electrode if the amount of heat input is smaller than a predetermined threshold value.

[0118] (Supplementary Note 6) The plasma processing apparatus according to Supplementary Note 5, wherein the calculation formula is a formula for calculating a heat input amount to the upper electrode from respective set values ​​of an ion energy control parameter and an electron density control parameter in the recipe received by the input unit.

[0119] (Supplementary Note 7) The plasma processing apparatus according to Supplementary Note 6, wherein the set value of the ion energy control parameter is at least one of the set values ​​of the bias source or the DC power supply, and the set value of the electron density control parameter is at least one of the set value of the plasma source, the pressure of the chamber, or the current value of the electromagnet.

[0120] (Supplementary Note 8) The plasma processing apparatus according to any one of Supplementary Notes 5 to 7, wherein the threshold value is any value between 4 kW and 5 kW.

[0121] (Supplementary Note 9) The plasma processing apparatus according to any one of Supplementary Notes 5 to 8, wherein the calculation formula is obtained by regression analysis.

[0122] (Supplementary Note 10) The plasma processing apparatus according to any one of Supplementary Notes 5 to 9, wherein the calculation formula is obtained by Lasso regression using stored data on the heat input amount of the upper electrode when plasma processing is performed and a recipe for the plasma processing.

[0123] (Supplementary Note 11) The plasma processing apparatus according to any one of Supplementary Notes 5 to 10, wherein the bias source applies a periodically pulsed voltage to the substrate support portion, the DC power supply applies a DC voltage to the upper electrode, and the calculation formula includes a linear term obtained by adding a voltage value of the pulsed voltage, a voltage value of the DC voltage applied to the upper electrode, a gas flow rate supplied into the chamber, and an integral value of a magnetic field generated in the chamber by the electromagnet, a square term obtained by squaring each of these terms and adding them, and a cross term obtained by multiplying each of these terms and adding them together.

[0124] (Supplementary Note 12) A recipe registration method for a plasma processing apparatus comprising: a chamber; a substrate support part disposed inside the chamber and supporting a substrate; an upper electrode facing a substrate support surface of the substrate support part; an electromagnet disposed in an upper part of the chamber; a plasma source configured to generate plasma inside the chamber; a bias source configured to generate a bias inside the chamber; a DC power supply electrically coupled to the upper electrode; and a control part that controls plasma processing on the substrate, wherein the control part executes the following steps when receiving an input of a recipe for plasma processing on a substrate: (a) determining whether or not abnormal discharge will occur in the upper electrode disposed inside the chamber and facing the substrate support part when plasma processing is performed according to the received recipe; and (b) refusing to register the recipe in a storage part when it is determined that abnormal discharge will occur in the upper electrode, and registering the recipe in the storage part when it is determined that abnormal discharge will not occur in the upper electrode.

[0125] REFERENCE SIGNS LIST 1 Plasma processing apparatus 10 Plasma processing chamber 11 Substrate support 13 Shower head 20 Gas supply unit 21 Gas source 31 RF power supply 32 DC power supply 32a First DC generation unit 32b Second DC generation unit 40 Exhaust system 50 Main head unit 51 Side head unit 52, 54 Upper electrode 52a, 54a Gas diffusion chamber 52b, 54b Gas inlet 53, 55 Support unit 53a, 55a Gas supply port 53b Flow path 58 Dielectric unit 70 Electromagnet 71 Yoke 72 Coil 73 Wiring 100 Control unit 101 External I / F 102 User I / F 103 Memory unit 103a Recipe data 103b Teacher data 103c Prediction model data 104 Process controller 104a Plasma control unit 104b Generation unit 104c Determination unit 104d Registration unit 111 Main body unit 111a Central region 111b Annular region 112 Ring assembly 1110 Base 1110a Flow path 1111 Electrostatic chuck 1111a Ceramic member 1111b Electrostatic electrode 1111c Central bias electrode 1111d Annular bias electrode W Substrate

Claims

1. A plasma processing apparatus comprising: a chamber; a substrate support section disposed inside the chamber and supporting a substrate; an upper electrode facing a substrate support surface of the substrate support section; an electromagnet disposed in the upper part of the chamber; a plasma source configured to generate plasma inside the chamber; a bias source configured to generate a bias inside the chamber; a DC power supply electrically coupled to the upper electrode; and a control section that controls plasma processing on the substrate, wherein the control section comprises: a memory section that stores a recipe for plasma processing on the substrate; an input section that accepts input of the recipe for plasma processing on the substrate; a determination section that determines whether abnormal discharge will occur in the upper electrode when plasma processing is performed using the recipe accepted by the input section; and a registration section that refuses to register the recipe in the memory section if the determination section determines that abnormal discharge will occur in the upper electrode, and registers the recipe in the memory section if the determination section determines that abnormal discharge will not occur in the upper electrode.

2. The plasma processing apparatus according to claim 1, wherein the upper electrode has a center portion and one or more annular portions arranged concentrically around the center portion, and the determination portion determines whether or not abnormal discharge occurs in the annular portions.

3. The plasma processing apparatus according to claim 2, wherein the upper electrode further comprises a dielectric portion electrically isolating the central portion and one or more annular portions adjacent to each other.

4. The plasma processing apparatus according to any one of claims 1 to 3, wherein the upper electrode is made of silicon.

5. A plasma processing apparatus according to any one of claims 1 to 3, wherein the judgment unit uses a predetermined formula to calculate the amount of heat input to the upper electrode when plasma processing is performed using a recipe received by the input unit, and judges that no abnormal discharge will occur in the upper electrode if the amount of heat input is smaller than a predetermined threshold value.

6. The plasma processing apparatus according to claim 5, wherein the calculation formula is a formula for calculating the amount of heat input to the upper electrode from the respective set values ​​of an ion energy control parameter and an electron density control parameter in the recipe received by the input unit.

7. The plasma processing apparatus according to claim 6, wherein the set value of the ion energy control parameter is at least one of the set values ​​of the bias source or the DC power supply, and the set value of the electron density control parameter is at least one of the set value of the plasma source, the pressure of the chamber, or the current value of the electromagnet.

8. The plasma processing apparatus according to claim 5, wherein the threshold value is any value between 4 kW and 5 kW.

9. The plasma processing apparatus according to claim 5, wherein the calculation formula is obtained by regression analysis.

10. The plasma processing apparatus according to claim 5, wherein the formula is determined by Lasso regression using stored data on the amount of heat input to the upper electrode when plasma processing is performed and the recipe for the plasma processing.

11. The plasma processing apparatus according to claim 5, wherein the bias source applies a periodically pulsed voltage to the substrate support, the DC power supply applies a DC voltage to the upper electrode, and the calculation formula includes a linear term obtained by adding together a voltage value of the pulsed voltage, a voltage value of the DC voltage applied to the upper electrode, a gas flow rate supplied into the chamber, and an integral value of a magnetic field generated in the chamber by the electromagnet, a square term obtained by squaring and adding each of these, and a cross term obtained by multiplying and adding each of these.

12. A recipe registration method for a plasma processing apparatus comprising: a chamber; a substrate support section disposed inside the chamber and supporting a substrate; an upper electrode facing the substrate support surface of the substrate support section; an electromagnet disposed in the upper part of the chamber; a plasma source configured to generate plasma inside the chamber; a bias source configured to generate a bias inside the chamber; a DC power supply electrically coupled to the upper electrode; and a control section that controls plasma processing on the substrate, wherein the control section performs the following steps: (a) when receiving input of a recipe for plasma processing on a substrate, determines whether or not abnormal discharge will occur in the upper electrode disposed inside the chamber and facing the substrate support section when plasma processing is performed according to the received recipe; and (b) when determining that abnormal discharge will occur in the upper electrode, rejecting registration of the recipe in a memory section, and when determining that abnormal discharge will not occur in the upper electrode, registering the recipe in the memory section.

Images