High-frequency integrated network and generator

High-bandwidth sensors in plasma processing systems quickly detect and correct plasma non-uniformity by analyzing RF surface waves, addressing the challenge of non-uniform etching rates and improving semiconductor manufacturing yield.

JP7713947B2Active Publication Date: 2025-07-28COMET TECHNOLOGIES USA INC
View PDF 6 Cites 0 Cited by

Patent Information

Application Number
JP2022543647
Authority / Receiving Office
JP · JP
Patent Type
Patents
Current Assignee / Owner
Priority Date
2021-01-19
Filing Date
2021-01-20
Publication Date
2025-07-28
Estimated Expiration
2041-01-20

Smart Images

  • Figure 0007713947000001
    Figure 0007713947000001
  • Figure 0007713947000002
    Figure 0007713947000002
  • Figure 0007713947000003
    Figure 0007713947000003
Patent Text Reader

Abstract

1. A method for providing data of a radio frequency pulse in a radio frequency plasma processing system, the method comprising: measuring an electrical parameter in a matching network of the radio frequency plasma processing system; determining an attribute of the measurement of the electrical parameter; defining a first statistic of the attribute of the measurement of the electrical parameter; defining a second statistic based on the first statistic for at least one of a phase and a process; providing the first statistic and the second statistic to a user; and storing the first statistic and the second statistic in the matching network.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] Cross - Reference to Related Applications This application claims priority to U.S. Provisional Application No. 62 / 963,444, filed on January 10, 2020, and U.S. Non - Provisional Patent Application No. 17 / 145,202, filed on January 8, 2021, the contents of which are hereby incorporated by reference.

Summary of the Invention

Problems to be Solved by the Invention

[0002] High - frequency (RF) plasma processes are widely used in semiconductor manufacturing for etching various types of films, depositing thin films at low to intermediate processing temperatures, and performing surface treatment and cleaning. A characteristic of such processes is the use of plasma, i.e., a partially ionized gas used to generate neutral species and ions from precursors in a reaction chamber, supply the energy of ion bombardment, and / or perform other operations. Controlling the plasma density during such processes is challenging, and the non - uniformity of the plasma in the reaction chamber affects the uniformity and yield of wafer processing of the integrated circuits or other devices being manufactured.

Means for Solving the Problems

[0003] Non-uniform plasma density in the reaction chamber can cause non-uniform etching rates or specific characteristics across the substrate. In certain systems, monitoring the uniformity of plasma density in the reaction chamber is performed using a probe. Such a probe may be exposed to a plasma environment that depends on the coating and may use active electronics to infer the plasma density. Such systems may take several milliseconds or more to respond to changes in the plasma. Emission spectroscopy may also be used to determine the profile of plasma density in the reaction chamber, but such systems require multiple lines of sight through the plasma and may use complex analysis to infer non-uniformities. None of these techniques have sufficient sensitivity and speed to effectively solve the problem of non-uniformities, and furthermore, they may be costly to implement.

Brief Description of the Drawings

[0004] The present disclosure is best understood from the following detailed description when read in conjunction with the accompanying drawings. It is emphasized that the various features are not drawn to scale in accordance with standard practice in the art. In fact, the dimensions of the various features can be arbitrarily increased or decreased to clarify the discussion.

[0005]

FIG. 1

[0006]

FIG. 2

[0007]

FIG. 3

[0008]

FIG. 4

[0009]

FIG. 5

[0010]

FIG. 6

[0011]

FIG. 7

[0012]

FIG. 8

[0013]

FIG. 9

[0014]

FIG. 10

[0015]

FIG. 11

[0016]

FIG. 12

[0017]

FIG. 13

[0018]

FIG. 14

FIG. 15

FIG. 16

[0019]

FIG. 17

[0020]

FIG. 18

[0021]

FIG. 19

[0022]

FIG. 20

DETAILED DESCRIPTION OF THE INVENTION

[0023] Disclosed are exemplary embodiments of the subject matter claimed below. To clarify, not all features of an actual implementation are described herein. It is understood that in the development of any such actual implementation, numerous implementation-specific decisions may be made to achieve the developer's specific goals, such as compliance with system-related and business-related constraints that vary from one implementation to another. Further, it is understood that such development efforts, even if complex and time-consuming, are routine for those of ordinary skill in the art having the benefit of this disclosure.

[0024] Further, as used herein, the article “a” is intended to have its ordinary meaning in patent technology, i.e., “one or more.” Here, the term “about,” when applied to a value, generally means within the tolerance of the device used to generate the value, and in some examples, means ±10%, ±5%, or ±1% unless specifically specified otherwise. Further, herein, the term “substantially” means, for example, an amount having a range of more than half, almost all, all, or from about 51% to about 100%. Further, the examples herein are intended only for illustration and are presented for discussion and not for limitation.

[0025] Turning to FIG. 1, a side schematic view of an RF plasma processing system 100 according to an embodiment of the present disclosure is shown. The RF plasma processing system 100 includes a first RF generator 105 and a second RF generator 110, a first impedance matching network 115, a second impedance matching network 120, a sheath 125, a plasma feeding device such as a showerhead 130 or an equivalent power supply element such as an electrode, and a pedestal 135. As used herein, the plasma feeding device may refer to any device that introduces power to generate a plasma, and may include, for example, the showerhead 130 and / or other types of electrodes, as well as antennas and the like.

[0026] The RF plasma processing system 100 may have one or more RF generators 105, 110 that supply power to the reaction chamber 140 via one or more impedance matching networks 115, 120. The RF power flows from the first RF generator 105 through the impedance matching network 115 to the plasma in the reaction chamber 140, flows to the showerhead 130 or the sidewalls, flows to electrodes other than the showerhead 130, or flows to an induction antenna (not shown) that electromagnetically supplies power to the plasma. Thereafter, the power flows from the plasma to ground and / or the pedestal 135 and / or the second impedance matching network 120. Generally, the first impedance matching network 115 compensates for variations in the load impedance within the reaction chamber 140 by adjusting reactive components within the first impedance matching network 115, such as a variable capacitor, so that the coupling impedance between the showerhead 130 and the first impedance matching network 115 equals the output impedance of the first RF generator 105, e.g., 50 Ω. Additionally, the reflected wave power can be corrected by adjusting the frequency within a range of approximately ±10% of the RF power. As used herein, the term “about” acknowledges that satisfactory results can be obtained even if some inaccuracy with respect to the range or value is experienced in practice. Such inaccuracies can result, for example, from calibration losses or degradation or drift during operation. However, in these situations, the stated range or value is a nominal target value for the operating conditions during use.

[0027] In certain examples, the first RF generator 105 may supply power at an RF frequency between about 400 KHz and 150 MHz, while the second RF generator 110 connected to the pedestal 135 may supply power at an RF frequency lower than the frequency of the first RF generator 105. However, in certain implementations, the second RF generator 110 may not supply power at an RF frequency lower than the frequency of the first RF generator 105. Typically, the frequencies of the first RF generator 105 and the second RF generator 110 are such that the first RF generator 105 is at an RF frequency that is neither an integer multiple nor an integer fraction of the frequency of the second RF generator 110.

[0028] The impedance matching networks 115, 120 are designed to adjust their internal reactive elements so that the load impedance matches the source impedance. Generally, low reflected wave power is considered positive, but embodiments of the present disclosure ensure that the supplied power is maintained within the reaction chamber 140, that the power is reflected towards the first RF generator 105 and the second RF generator 110, and that even when the reflected wave power is relatively high, the associated impedance matching networks 115, 120 monitor the forward wave power to the reaction chamber 140 and the reflected wave power from the reaction chamber 140 and can adjust variable reactor elements, such as vacuum variable capacitors, by use of a motor drive system. In certain embodiments, an electronically controlled capacitor, such as a PIN diode for an electronically variable capacitor, may be used. The impedance matching networks 115, 120 may include circuitry for measuring the phase and magnitude of the signals to determine the levels of forward wave power and reflected wave power from the intended load. Thus, embodiments of the present disclosure can be effective even when there is a large amount of reflected wave power. If there is a significant amount of reflected wave power at the primary frequency, the capacitor varies until the reflected wave power is minimized, e.g., to less than about 5 W and / or less than about 1% for a period, or, in certain embodiments, to less than 1 W. Generally, harmonic frequency signals, including reflected wave power at harmonic frequencies, are not measured.

[0029] The RF plasma processing system 100 has many advantages, but historically, it has been a challenge to maintain control of the plasma density through a multi-step process. Design tolerances with a density range of the same order of magnitude as the nominal value, for example, on the order of 1% non-uniformity, remain a challenge. Also, with miniaturization of about 3 nm or less and thick film formation of about 10 nm or less, it is necessary to strictly control the uniformity of the plasma and neutral species to the 1% level or even lower to achieve an optimal integrated circuit (IC) yield per wafer. A non-uniform plasma density or an average density that deviates beyond the desired range from the desired value in the reaction chamber is caused by slow changes in the chamber, changes in the RF circuit, or rapid growth (on the order of less than 1 millisecond) of parasitic or secondary plasmas, which can thereby result in non-uniformity of the nanoscale features across the processed wafer due to non-uniformity of the etching rate.

[0030] Even a 1% difference in etching rate across the wafer can potentially cause yield problems in state-of-the-art processes and often requires a significant amount of time to complete wafer processing to see yield losses. Therefore, in order to avoid irreversible deviations on the wafer from the desired feature profile, there is a need to quickly and accurately detect non-uniform plasma density or plasma density that deviates from the desired range in the reaction chamber at a time period that may need to be less than about 1 millisecond.

[0031] One skilled in the art understands that electromagnetic (EM) surface waves may propagate along the surface within the RF power plasma in the reaction chamber 140. These surface waves have energy that can be evaluated at both the fundamental RF drive frequency and RF harmonics. The average power and power distribution of the harmonics are factors that affect plasma density and non-uniformity. Here, the harmonic profile is defined as the spectrum of surface waves having frequencies that are integer multiples of the fundamental drive frequency of the RF plasma-based reaction chamber 140. For example, when 2 MHz of RF drive power is provided to the reaction chamber 140, the injected power generates surface waves at that frequency that propagate along the interface between the plasma and the surface of the inner reaction chamber 140. By adjusting the frequency by about ±10% of the RF power, the reflected wave power can be corrected. Harmonic surface waves at integer multiple frequencies may also be generated. For example, a 2 MHz electromagnetic wave may generate surface waves at 4 MHz, 6 MHz, or 8 MHz. Odd and even harmonics (second, third, fourth, fifth, etc.) may appear, and in some examples, the odd harmonics may be dominant.

[0032] Aspects of the present disclosure may provide sensor locations for the reaction chamber 140 and its components that enable the detection and analysis of RF surface waves to determine the amplitude and phase of the fundamental and harmonic waves at multiple points within or adjacent to the reaction chamber 140. The wave may be detected by sensing the RF voltage or RF current at the fundamental and harmonic frequencies on the surface of the chamber components. In some embodiments, the sensor for voltage has a pickup configured on or proximate to the surface of an electrode, pedestal base, chamber wall, or strap, and a conductive wire that transmits a signal from the pickup to a connector or cable. The current sensor may have a conductive element that includes one or more loops or partial loops or linear conductors, and one end of the conductive element may be at a reference potential that may be a local electrical ground.

[0033] A plurality of sensors, e.g., two or more sensors, may be placed at different angles about the symmetry axis of the chamber on specific chamber components described in detail below to measure surface voltages or currents associated with such surface waves. Here, the measured angle about the symmetry axis from a reference point of the chamber is defined as the azimuth angle. In some embodiments, such sensors may be placed at substantially the same distance from the symmetry axis of the chamber.

[0034] Sensors may be attached at various locations on or with respect to the reaction chamber and / or its components. For example, sensors may be attached to the surface of electrodes such as pedestal 135 and / or showerhead 130. Sensors may be attached to the pedestal of any electrode either inside or outside the vacuum. Sensors may be attached inside or outside one or more metal walls and wall regions containing dielectric material of the reaction chamber 140 within the chamber, or to an antenna that may be used to inductively power the plasma. Sensors may be placed on or in proximity to a passive antenna that may be used to detect electromagnetic waves in the vicinity of the plasma boundary, or on or in proximity to a plurality of conductive buses or straps that connect the first impedance matching network 115 or the second impedance matching network 120 to electrodes such as pedestal 135 and / or showerhead 130, antennas, or other components that transmit power to the plasma within the reaction chamber 140. Sensors may be connected to electrical ground. Thereby, the sensors can acquire signals as they propagate on the surfaces of the various components of the RF plasma processing system 100.

[0035] The spectrum of RF harmonics is generated at the electrode-plasma interface, e.g., sheath 125 of FIG. 1, and the waves propagate in all directions such that both the amplitude and phase of all wave components vary with the position of the electrode or support. Such waves propagate along the inner surface of the metal wall adjacent to the plasma and through any dielectric wall that may be adjacent to the plasma. The amplitude and phase of such waves vary in response to changes in the plasma, e.g., plasma density and non-uniformity, and the response time is on the order of a few microseconds or less. Further, the frequency distribution and phase distribution of the RF harmonic surface waves propagating across the electrode-plasma interface determine the frequency and phase distribution of the harmonic surface waves propagating across the electrode-based surface, the surface connected to the electrode or plasma-wall interface, or the wall, towards the impedance matching networks 115, 120. The amplitude and phase of the fundamental and harmonic signals at different sensor positions enable the determination of the azimuthally symmetric and asymmetric parts of the total EM wave field for each frequency.

[0036] In the case of an inductive plasma, signals from the plasma, e.g., fundamental and harmonic waves, may propagate back to the antenna and then to the impedance matching network that powers the antenna. The frequency and phase distribution of the fundamental and harmonic RF waves may be monitored on a microsecond or faster time scale using sensors attached to such a surface and compared in terms of specified ranges and phase relationships as an indicator of plasma asymmetry or changes in plasma density or conductivity. To generate the amplitude and phase values of each frequency component at each sensor position, the signals from such sensors may be transmitted via a cable or the like to a detector that analyzes the frequency components of the signal.

[0037] In a particular implementation, the amplitude and phase of the detected RF harmonic components may be determined rapidly by a circuit (detector) within the signal analysis section, which may be a separate metal box or chassis, or may be within or connected to impedance matching networks 115, 120. Such amplitude and phase may be used to determine a state including the radial distribution and asymmetry of the plasma by applying algorithms and plasma non-uniformity calibration. The signal from the sensor may be Fourier analyzed by a dedicated circuit (detector) at a speed sufficient to perform a virtually continuous spectral analysis, updated as frequently as possible, and may result in a high-speed data stream. For example, for a 13.56 MHz plasma output, it takes 50 microseconds or less to process 512 cycles by Fourier analysis, and the state of the plasma can be updated at a rate of 10 KHz for a pulsed plasma when each element of the pulse is generated at 5 KHz.

[0038] The results of the dedicated Fourier analysis of the fundamental wave and harmonics may be stored in another storage medium that can be read and / or written by an analysis processor associated with the signal analysis section. Either the stored results or the real-time signal may be sent to a high-speed calculation processor to determine the asymmetry parameters for each of the fundamental wave and harmonics. To very quickly recognize the "plasma defect" state, the asymmetry parameters may be compared using an algorithm (which may also be stored in another storage medium or a different storage medium) with values pre-stored in another storage medium (or a different storage medium). The analysis processor may then send appropriate commands, for example, to continue the process under current conditions or to make the necessary changes in the process conditions, to the first RF generator 105 and the second RF generator 110, more than two RF generators in a particular implementation, and the impedance matching network associated with these generators when appropriate. In a particular implementation, three, four or more RF generators may be used. The first RF generator 105 and the second RF generator 110 may then continue, stop, change the supplied power, change the frequency by about ±10% of the RF power to correct the reflected wave power, or any other appropriate method - for example, respond to a transition to a reduced power mode or a pulse mode, or perform specific corrective actions to avoid improper wafer processing during a plasma failure or other unacceptable situation, such as commands for an alarm trigger, power cut-off, etc.

[0039] The position of the sensor for detecting surface waves (electric and magnetic fields) and characterizing the surface waves may, in some embodiments, be the peripheral surface of the pedestal 135 (bare or covered by a dielectric) outside the area covered by the wafer. For example, when the reaction chamber 140 processes a circular wafer with a radius of 150 mm, the pedestal mount sensor may be placed at a radius greater than 150 mm from the center of the wafer, which may, in some cases, be under an annular peripheral dielectric for controlling edge effects. The sensor may be additionally or alternatively placed on the surface or periphery of the showerhead 130 facing the wafer or on the base of the pedestal 135 or the surface of the base of the showerhead 130, regardless of whether these locations are within or outside the vacuum process environment. The sensor may be placed in various other locations described in detail below and may be monitored continuously or periodically to provide process plasma uniformity.

[0040] Using sensors outside the vacuum process environment, such as one or more of the impedance matching networks 115, 120, sensors on the base of the pedestal 135 and / or straps or buses connecting the base to the showerhead 130, may obviate the need to transmit signals through vacuum feedthroughs or transmission cables within the vacuum vessel of the reaction chamber 140. Sensors in such locations may monitor the fundamental and harmonic EM waves substantially continuously. Thereby, the RF plasma processing system 100 may be able to continuously provide plasma density uniformity and determine, within a very short time, whether a fault condition has occurred or whether appropriate wafer or substrate processing should continue.

[0041] In certain exemplary embodiments, the present disclosure can provide an apparatus and method for detecting deviations of plasma from a required "process window" within an RF plasma processing system 100. The RF plasma processing system 100 may include a reaction chamber 140, which may include a showerhead 130 for injecting a reaction gas into the reaction chamber 140, and the reaction chamber 140 may include a wafer support pedestal 135. However, in other implementations, the showerhead 130 may not inject gas into the reaction chamber 140. In some embodiments, the showerhead 130 may be provided with a plurality of sensors having their centers mounted near the approximate axis of symmetry of the reaction chamber 140 and arranged in a selected orientation about the axis of symmetry. Additionally or alternatively, such sensors may be disposed on a surface facing the wafer in a peripheral region of the showerhead 130 for detecting and measuring EM surface waves propagating while the wafer is being processed.

[0042] Further, in some embodiments, there may be a plurality of sensors mounted on an outer surface of the wafer support pedestal 135 outside the region occupied by the wafer for detecting both the amplitude and phase of RF harmonics and fundamental surface waves. Such sensors may be exposed to the plasma or may be covered by a dielectric cover or a dielectric and metal cover. Additionally or alternatively, the sensors may be located at the periphery of the pedestal 135 base, inside or outside the vacuum chamber, and / or below the plane defined by the wafer. In some embodiments, the sensors may be disposed on the pedestal base for detecting surface electromagnetic waves propagating towards or away from the wafer support region of the pedestal and surface electromagnetic waves propagating on the surface of the pedestal base. In certain embodiments, the sensors may be mounted near (e.g., less than 10 cm) the wafer plane.

[0043] Alternatively, the sensor may be attached to a portion of the pedestal 135 that is made of metal or other conductive material and is located outside the vacuum region of the reaction chamber 140 under atmospheric conditions. The sensor located outside the vacuum region may be attached to the region of the pedestal 135 at a radius that is at least 50% of the maximum radius of the pedestal 135 or exceeds 75% of the maximum radius of the pedestal 135 from the pedestal symmetry axis. Such a sensor may be positioned near, within a few centimeters in some embodiments, a vacuum seal for the support pedestal 135, e.g., an O-ring. In some embodiments, the total radial and axial propagation distance from the edge of the wafer to the sensor may be less than about 25 cm, or less than about 15 cm, or about 10 cm in some embodiments. The specific location and orientation of the sensor according to embodiments of the present disclosure are discussed in detail below.

[0044] Turning to FIG. 2, a schematic side view of a plasma chamber having high impedance sensors attached at various positions of the electrodes in accordance with an embodiment of the present disclosure is shown. Each of the two components that function as electrodes, namely, the pedestal 235 and the showerhead 230 or other equivalent power supply element, may use separate RF generators 205 or 210 and impedance matching networks 215 and 220. Alternatively, the electrodes may have a plurality of generators and matching networks that supply power thereto. The arrow 245 along the surface of the pedestal 235 indicates the inner radial direction of the flow of RF current and power from the lower (bias) RF generator 210 electrically connected to the pedestal 235 via the impedance matching network 220. The generated electric field contributes to the formation of a plasma (not shown) between the electrodes and to the external radial backflow of current and power indicated by the arrow 250 along the lower surface of the showerhead 235 or other power element, and ultimately contributes to the selective ground circuit of the impedance matching network 215 for the showerhead 230 or other power element.

[0045] In certain embodiments, the reaction chamber 240 having RF power from the first RF generator 205, the second RF generator 210, and the impedance matching networks 215, 220 may have sensors 255 at the periphery of a pedestal 235 that may be covered by a dielectric 260. A communication line 265 may transmit signals from each of the sensors 255, which may be substantially equidistant from the pedestal symmetry axis in some embodiments, to a Fourier analysis circuit (not shown) that calculates the amplitudes and phases of both the fundamental frequency surface wave and the harmonic frequency surface wave acquired by each sensor 255.

[0046] In some implementations, the Fourier analysis circuit may calculate the magnitudes and phases of the fundamental and higher order harmonics of a periodic electromagnetic surface waveform. The resulting magnitudes and their phases, known as a series of Fourier series, arise from the relationship between a function in the time domain and a function in the frequency domain.

[0047] Furthermore, some embodiments of the disclosed matching network 220 may have a signal analysis section 275 or appendage of the matching network 220 that is separated from the RF power processing and impedance matching circuits or components of the matching network 220 and is decoupled from the RF. The signal analysis section 275 may have one or more Fourier analysis circuits (detectors) for analyzing sensor signals and providing the digital amplitude and phase of the RF fundamental and harmonics. The signal analysis section 275 may have high-speed digital logic or a computing processor for analyzing the relative magnitude and phase of the signals at the harmonic frequencies and deriving quantitative parameters that characterize the relative magnitudes and relative phases of the axisymmetric and non-axisymmetric harmonic components at each frequency. Further, in some embodiments, the disclosed matching network 220 may be connected via a very high-speed network not only to the second RF generator 210 but also to a controller (not shown) for the reaction chamber 240 or the RF plasma processing system 200 in which the sensor 255 is disposed. In some embodiments, the disclosed enhanced impedance matching network 220 may be capable of transmitting commands to the first RF generator 205 and transmitting the calculated parameters to the process chamber controller and / or the tool control system.

[0048] Furthermore, another first RF generator 205 and an impedance matching network 215 may be electrically coupled to other electrodes, which may be the showerhead 230 of the reaction chamber 240. In one embodiment, the first RF generator 205 may operate at a frequency different from that of the second RF generator 210, and that frequency may not be an integer multiple of the frequency of the second RF generator 210.

[0049] Similarly, the impedance matching network 215 may monitor the reflected wave power from the electrodes and the processing chamber 240 and make adjustments if there is significant reflected wave power from the electrodes. In some embodiments, the second RF generator 210 may be a 400KHz RF generator, a 2MHz RF generator, a 13.56MHz RF generator, etc., while the first RF generator 205 may operate at a somewhat higher frequency. In some embodiments, the first RF generator 205 may operate at a frequency greater than 25MHz, such as 60MHz, 100MHz or higher.

[0050] In one embodiment, the main function of the first RF generator 205 is to generate reactive species such as fluorine, chlorine and their compositions and accelerate the ions from the generated plasma to collide with the wafer placed on the pedestal 235, so as to supply power to the reaction chamber 240 to generate plasma between another power source such as the shower head 230 or the electrode and the pedestal 235.

[0051] A set of sensors 280 having a bandwidth greater than about 10 times the frequency of the highest frequency RF generator connected to the electrodes may be disposed on the lower electrode surface, i.e., the upper electrode facing the pedestal 235, i.e., the shower head 230. In some embodiments, each of these may have an impedance greater than about 100Ω, and in some embodiments, an impedance greater than 500Ω. The sensor 280 may be a voltage sensor or a current sensor, or may combine both capabilities in a single package. For example, the current sensor may have one or more segments of wire that may be covered by an electrostatic shield.

[0052] In some embodiments, sensor 280 has an electrical connection to the Fourier analysis circuit in the signal analysis section 285 of impedance matching network 215. The Fourier analysis circuit may output the amplitudes and phases of different frequency components from each of sensors 280 and may compare the amplitudes and phases of the frequency components of other sensors 280 and / or with a reference level stored in the memory. The analysis of the signals in some embodiments may employ artificial intelligence (AI) including a learning algorithm that may use pattern recognition of amplitude or phase or both, or a neural network, or a conventional digital algorithm process of the signals from sensors 280.

[0053] Signal processing by a Fourier analysis circuit to find fundamental wave component signals and harmonic component signals of both amplitude and phase may be performed in less than about 10 microseconds for each of the sensor signals and, in a preferred embodiment, in less than 1 microsecond. The separate signal analysis section 285 of impedance matching network 215 may incorporate at least one computing or logic processor having substantial computing power by an ultra-fast (<1 ns cycle time) circuit employing ultra-fast logic ICs. In some embodiments, the processor of signal analysis section 285 is programmable such that a dedicated algorithm or analysis software of a computing “platform” provided by a provider or user of processing chamber 240 to impedance matching network 215 can be provided or implemented.

[0054] In some embodiments, the software program for calculating parameters from signal amplitude and phase and the separate logic algorithm for determining the impact on the process uniformity of deviations from acceptable plasma conditions may reside in a removable "plugin" component having data storage and connected to the signal processing section. This software or logic calculates the degree of deviation of the RF electromagnetic surface wave spectrum from the nominal operating conditions or characteristics of proper operating conditions. Based on this, the processor associated with the controller can "decide" on a corrective action or end of the process within about 1 millisecond before the wafer is mishandled. In some embodiments, a quantitative determination regarding the expected impact of the deviation on the process uniformity or other characteristics may be made within about 500 microseconds from occurrence so that the corrective action can be initiated within 1 millisecond. Further, measures may be taken to avoid loss of wafer or substrate yield in order to minimize or eliminate damage to the wafer or substrate being processed in the reaction chamber 240 at that point in time.

[0055] In some embodiments, the evaluation and / or determination performed in the signal analysis section 285 of the impedance matching network 215 may be performed by an ultra-high speed computing or analysis system using algorithms residing in a plugin storage and / or removable data processing device. In yet another embodiment, the evaluation determination performed in the signal analysis section 285 may be performed using an analog or neural net type processor. Such determination may further use a decision algorithm that may reside in a removable storage device or processing device. Thereafter, an instruction for a corrective action may be promptly transmitted from section 275 of the impedance matching network 215 to the RF generator 205 via a high-speed data line to temporarily interrupt, change, or terminate the power or RF frequency to the plasma. This ensures that the factory manager can promptly take or plan for a corrective action for its processing chamber 240 and RF plasma processing system 200.

[0056] Also, FIG. 2 shows a set of sensors 290, which are configured on the outer surface of the pedestal 295 of the showerhead 230 outside the vacuum region within the reaction chamber 240 under atmospheric conditions. In some embodiments, additional sensors 296 may be attached to the pedestal base 297 and connected by a high-speed signal cable to the disclosed signal processing section 275 of the impedance matching network 220 in the same manner as the sensors 290. Sensors 296 located outside the vacuum environment of the reaction chamber 240 are significantly less expensive since they do not require a vacuum feed and are not difficult to integrate into the information and processing network.

[0057] Sensors 255 may be arranged in some configurations to sense the voltage and / or current on the surface of the pedestal 235, and may be covered by the dielectric cover 260 and protected from the plasma. Sensors of this type and location are in proximity to the wafer and / or substrate and can thus have an advantage in sensitivity when detecting specific modes of EM surface waves that exhibit plasma asymmetry, an important type of plasma non-uniformity. These in-chamber sensors 255 may use a communication line passing through the vacuum wall via a feedthrough or a wireless communication link operating at optical frequencies or low frequencies in some embodiments.

[0058] Generally, the phase and amplitude pattern of the EM surface wave at each frequency on the surfaces of the showerhead 230 and the pedestal 235 may be determined by analyzing the signals from any of the group of voltage, current, phase, or combined sensors 255, 280, 290, and 296. Generally, an EM surface wave at a given frequency generates voltage and current signals that have a phase relationship with signals at other frequencies. The magnitude of the voltage at each frequency and each point is the sum of the voltages from all the waves at that frequency generated from all the points on the entire electrode surface. In the case of an axially symmetric electrode surface where power is symmetrically supplied and the plasma is axially symmetric, the axially symmetric surface wave mode results from the superposition of waves from all parts of the electrode and other surfaces of the reaction chamber 240. Generally, a fully symmetric plasma in a symmetric chamber with a symmetric electrode centered on the axis of symmetry of the chamber mainly has symmetric lines of equal phase and amplitude in the form of a circle centered on the center of the center of the pedestal 235.

[0059] Turning to FIG. 3, a cross-sectional view of a dual-plate electrode assembly having a broadband sensor that supplies a voltage signal to the periphery of the electrode and to electrical ground via an electrical connector having a low shunt capacitance according to an embodiment of the present disclosure is shown. In some embodiments, an electrode, such as showerhead 330, may have two conductive plates 331, 332 that are centered and configured substantially parallel to each other and have a shape substantially the same as the substrate or wafer. The surface of the first plate 331 facing away from the second plate 332 may be exposed to a vacuum environment and plasma. The first plate 331 is spaced apart from the second plate 332 by a distance that is the length of the dielectric standoff support 333. The first plate 331 may have an embedded sensor 334 whose pack or pickup is a conductive material and whose surface is on a plane substantially the same as the surface of the first plate 331 facing away from the second plate 332.

[0060] In some embodiments, sensor 334 may be attached to a first plate 331 surrounded by a dielectric 336 having a low dielectric constant, such as quartz or other suitable material. In some embodiments, dielectric 336 may have a dielectric constant of less than 5, and in some embodiments, the dielectric constant may be less than 2 for inorganic materials such as silica-based aerogels. Sensor 334 may have a high bandwidth extending from 100 KHz to at least 10 times the highest drive frequency connected to the chamber, the high bandwidth may be 300 MHz or more, and it may be possible to detect surface voltage, surface current, or both. The sensitivity of sensor 334 in some embodiments may vary by less than 30% over the frequency range of the harmonics of the main fundamental RF frequency used in the reaction chamber. In some embodiments, at least one lead wire 337 from each sensor is connected to an inner conductor 338 of a vacuum electrical signal feedthrough 339 having a base 341 attached to an electrically grounded second plate 215. In some embodiments, the lead wires from each sensor may be directly connected to a circuit board in a similar position to 332 having one detection circuit for each ground plane and each sensor to determine the amplitude and phase for each frequency component.

[0061] The internal conductor 338 of the feed-through 339 is attached to the base 341 of the feed-through 339 mounted on the grounded second plate 332 such that the total shunt capacitance from the sensor 334 + lead 337 + feed-through 339 to ground is less than 5 pf, and in some embodiments less than 3 pf. It may have a small shunt capacitance, for example, 5 picofarads (pf, less than 2 pf in some embodiments), with respect to the base 341 of the feed-through 339. In some embodiments, the output from the base 341 mounted on the grounded second plate 332 may be connected to an attenuator (not shown). In some embodiments, the attenuator may have an electrical resistor with a resistance value greater than about 100 Ω. A shunt resistor with respect to ground 405 may be present in parallel with the electrical resistor 404. The resistance of the shunt resistor may be, for example, 50 Ω, or alternatively, equal to the impedance of the cable connecting the attenuator to the controller of the communication network or plasma chamber. When a detector is placed instead of a connector as shown in FIG. 3, the signal output from the detector, which is the voltage or current amplitude and phase at each frequency of the sensor, may be transmitted to an analysis processor in a section of the matching network.

[0062] Each sensor 334 may measure the voltage or current amplitude of the coupled electromagnetic surface wave mode having the fundamental frequency and harmonic frequencies of all RF generators supplying power to the plasma. The range of the fundamental frequency and harmonic frequencies is from about 10 KHz to about 500 MHz or more to the same extent. In other embodiments, the sensor may measure the voltage at fundamental and harmonic frequencies in the range from about 100 KHz to about 1 GHz.

[0063] FIG. 4 is a cross-sectional view of a pedestal in which a broadband voltage sensor according to an embodiment of the present disclosure is embedded. The voltage sensor 401 may be attached to an electrode such as the pedestal 400. In some embodiments, the sensor 401 may be connected to an electrical ground 406 via a resistor. The tip or pack of the sensor 401 may have a lead wire 402 surrounded by a dielectric 403 (which may optionally be air or vacuum). In some embodiments, the lead wire 402 from the sensor 401 may pass through an attenuator such as a resistor(s) 404 having a shunt resistor 405 which in some embodiments is about 50 Ω and may be connected to the electrical ground 406. Such a resistor 404 may be non-inductive and may have a resistance in the range between about 100 Ω and about 100,000 Ω. In some embodiments, the resistance may be between about 500 Ω and about 10,000 Ω. The resistor 405 may be non-inductive.

[0064] Furthermore, the dielectric 403 is generally non-magnetic and has a low loss tangent which is less than about 0.01 in some embodiments or less than about 0.001 in other embodiments. The shunt capacitance between the tip of the sensor 401 and the lead wire 402 to the ground electrode is less than about 5 pf, or in some embodiments less than about 2 pf, such that the reactance between the sensor 401 and the electrode of the pedestal 400 is greater than about 100 Ω at 300 MHz. The purpose of such a low shunt capacitance is to minimize the absorption of wave energy and to reduce the loading of surface waves by the sensor 401 such that waves propagate as they would in the absence of the sensor 401. Under such conditions, the detected surface potential does not differ significantly from the potential at the electrode without such a sensor 401.

[0065] Turning to FIG. 5, a schematic side view of a pedestal having associated RF and control components according to an embodiment of the present disclosure is shown. The power supply circuit of pedestal 501 includes an RF power generator 405 and an impedance matching network 506. High-speed signal lines, such as cables 511, 512, carry signals from sensors 502, 503 to a section that may be within or attached to impedance matching network 506 in some embodiments. The high-speed line 513 of the data network captures information from the impedance matching network 506 to the (one or more) controllers 514 of the reaction chamber or generator or to a tool or factory (not shown). Sensors 502, 503 are mounted on or near base 504 of pedestal 501, which may be inside or outside the vacuum region of the reaction chamber.

[0066] In some embodiments, there may be a signal analysis section associated with impedance matching network 506, such as a fault detection section 510. Signal analysis section 510 may be electrically and / or RF isolated from specific components such as the vacuum capacitor and high voltage electronics of impedance matching network 506. Signal analysis section 510 receives signals from sensors 502, 503 via cables 511, 112. Signal analysis section 510 directs the signals to an internal circuit that may include electronic components such as transistors and passive components, also referred to as detectors, from each of sensors 502, 503. In an alternative embodiment where the amplitude and phase are found directly adjacent to the sensor for each frequency component, the signals arriving at the signal analysis section may be the amplitude and phase for each frequency component rather than the unprocessed signals.

[0067] Each detector (not shown) within the partition 510 may perform RF spectral analysis of the signals from one sensor 502, 503 or from a group of sensors analyzed in parallel. The analysis may include temporally averaging the signals of the group of sensors or the signals of one or more sensors 502, 503 for noise reduction. In some embodiments, there may be outputs from each detector for the signals acquired by each sensor 502, 503, for example, the amplitude and phase for each frequency component of the fundamental wave and harmonics. The outputs from each detector may be input to an analog-to-digital converter for each harmonic signal, resulting in digitized values for both the measured amplitude and phase of each harmonic.

[0068] These digital amplitude and phase values for each frequency component and each sensor may be input to a high-speed digital processor in the signal analysis section associated with the disclosed impedance matching network with little delay, for example, in less than 10 microseconds. The digital processor may analyze information on both the amplitude and phase of the fundamental wave and each harmonic from the sensor and determine the relative magnitudes of various surface wave modes, including axisymmetric and non-axisymmetric modes, for both the fundamental wave and harmonics. There may be various non-axisymmetric modes for each frequency component, one or more of which may be indicators of plasma non-uniformity.

[0069] In some embodiments, such non-axisymmetric modes may be rapidly identified by algorithms present in the plugin. A reference database correlating the size of the non-axisymmetric mode with the plasma non-uniformity % may also be present in this plugin or a removable processor. The digital processor may calculate the rate of change of the amplitude of the wave mode and the rate of increase of the amplitude of one or more wave modes to determine the likelihood of a near-future malfunction. One indicator of the size of the non-axisymmetric mode at a given frequency may be the phase difference of the surface wave at a given frequency at various sensor positions having the same radial distance from the center of the circular electrode symmetrically disposed in the axisymmetric chamber. Alternatively, a second indicator of the non-axisymmetric mode may be the difference between the amplitudes of the surface wave at a given frequency at various sensor positions having the same radial distance from the center of the circular electrode symmetrically disposed in the axisymmetric chamber.

[0070] The matching network 506 having the insulating compartment 510 including a multi-channel detector system (not shown) can perform simultaneous Fourier analysis, digitization, and recording of the voltage amplitude and phase of the EM waves propagating at various positions of the pedestal 501. Due to the inherent noise, each of the determined voltage amplitudes and phases may be averaged over a short time interval as needed, averaged over the group of sensors 502, 503 to make a determination of the relative magnitude, or averaged temporally over a relatively large number of pulses.

[0071] Other power supply elements such as a showerhead, pedestal, or electrode equipped with a group or array of sensors may be used as a test system to generate data characterizing and recording the relationship between the spectrum and spatial pattern of the EM wave modes during the RF process and the various non-uniformities of the plasma density. In some embodiments, these data may be analyzed offline by an engineer to characterize and classify the behavior of the plasma and stored in a database that may be placed in a plugin storage device that may be connected to the matching network compartment or other controller or monitoring system.

[0072] The relationship between the amplitude and phase pattern characteristics of non-axisymmetric EM modes and axisymmetric EM modes and the non-uniformity of the process and plasma or deviations from proper conditions may be stored in a plugin that connects to the disclosed signal analysis section of the matching network. In an implementation where an RF plasma processing system may be used as a production tool, plasma and process non-uniformities may be rapidly detected when the operation of the chamber is thereby being monitored. For example, a sensor of the disclosed type shown in FIG. 4 configured as shown in FIG. 2 may be incorporated into the RF plasma system as shown in FIG. 1.

[0073] To determine whether a process plasma may have experienced a plasma fault condition, an analysis processor of a signal analysis section associated with an impedance matching network may calculate a parameter based in part on the magnitude of a non-axisymmetric EM mode for each of a predefined series of harmonics of the drive frequency of several electrodes or antennas. The processor in some embodiments may compare these parameters to a reference range in a database. Such a reference database may be present in a plugin connected to a signal analysis section that may be a section of the impedance matching network or a section associated therewith.

[0074] The database may store parameters characterizing various plasma conditions to assist in determining how severe a deviation of the plasma from an acceptable "process window" is. In some embodiments, the analysis may include a comparison of the phases of each harmonic from all sensors or groups thereof at a given distance from the center of the electrode. Such a change in phase for a sensor or group of sensors for any orientation may be a measure of the asymmetry of the generation and / or propagation of that harmonic mode and thus may be a measure of the asymmetry and non-uniformity of the plasma. In some embodiments, the analysis may include calculating the difference in amplitude between sensors or groups of sensors at a given distance from the axis of symmetry. Such a change in amplitude for adjacent sensors or groups of sensors over a range of orientations may be a measure of the asymmetry of the generation and / or propagation of that harmonic mode and thus may be a measure of the asymmetry and non-uniformity of the plasma.

[0075] A quantitative measure of asymmetry, a parameter, for each of a set of harmonics may be stored in a plug-in unit and transmitted to the chamber and tool controller via a data network. Further, trends and rates of increase in the parameters may be calculated as part of a process for determining whether a fault condition occurs and compared to reference values and criteria in the database. In some embodiments, when such a fault condition occurs, algorithms and criteria that may be stored in the plug-in may be executed in a processor present in the section to determine a course of corrective or preventive action. Such an action may be transmitted quickly to the RF generator and / or chamber and / or tool controller.

[0076] In some embodiments, all such databases of parameters, algorithms, criteria, and specifications for comparing parameters, rates of change of parameters, and rates of increase of parameters may be in a data storage device or removable processor that may be connected to ports that may be input / output ports of the signal analysis section. Analysis of surface wave modes based on signals from sensors and parameters obtained therefrom may be performed very quickly by a processor, and any fault declarations and improvement measure instructions may be sent to the RF generator and reported to the controller for the chamber or system via the network within 5 milliseconds of occurrence. In some embodiments, the fault condition and specified improvement measure instructions may be sent to the generator within 1 millisecond.

[0077] In some embodiments, deviations of many types of plasmas from the desired plasma uniformity may be detected quickly enough so that the tool or chamber controller can take measures to correct the plasma fault condition before the wafer or substrate is mishandled. In some situations, the specified improvement measure may be in the form of RF power, for example, the continuous wave (CW) or pulse is changed in a short time or the power is completely turned off in a short time, or the processing of the current wafer is stopped and the wafer is saved for later processing or disposal, or the reaction chamber may be powered off for maintenance. Therefore, the reflected wave power may be changed by adjusting the frequency within a range of about ±10% of the RF power.

[0078] In certain embodiments, upon detection of a plasma fault condition, the disclosed signal analysis section related to the matching network may command appropriate corrective actions to be performed by the RF generator and / or in some embodiments by the matching network. For example, the RF process generator may initiate an end process to end the processing of the wafer in response to signals measured by sensors of the showerhead and / or pedestal. In a specific embodiment, the frequency may be adjusted within a range of about one-tenth to 10%, i.e., increased or decreased. Alternatively, the power, e.g., the pulse power mechanism, may be interrupted by the RF plasma processing deposition system to stop or pulse the plasma such that the secondary plasma stops or significantly decreases. In some embodiments, after a very short interruption, the specified remedial measure may provide that the processing can then continue. In a particular embodiment, the remedial measure may be determined, for example, through machine learning based on yield data or other wafer diagnostics and / or a programmed remedial program.

[0079] Turning to FIG. 6, a top view of the propagation of an axisymmetric surface wave across the entire pedestal where the plasma in the reaction chamber according to an embodiment of the present disclosure is axisymmetric is shown. In FIG. 6, circle 601 is a curve with constant phase and amplitude for the fundamental frequency component and harmonic frequency components of the axisymmetric surface wave mode. The circle is concentric with the electrode. These modes are very dominant when the chamber, electrode, and plasma are all axisymmetric and coaxial. The propagation vector 602 of the surface wave at any frequency is radial. The wave propagates towards the center and away from the center, and as the wave propagates, such a wave injects power into the plasma.

[0080] Turning to FIG. 7, there is shown a top view of the lateral electromagnetic surface wave propagation across the entire electrode according to an embodiment of the present disclosure. In FIG. 7, the lines 701 - 704 of constant phase and equal amplitude for a particular single non - axisymmetric mode are substantially straight and parallel regardless of the fundamental frequency or its harmonics. Such surface waves may be detected by sensors disposed on the pedestal or showerhead of an RF plasma deposition apparatus. This mode may be referred to as "lateral" in that, as seen by the propagation vectors 705, 707, the propagation direction is from one side to the other across the entire electrode surface or from the central plane to both the left and right sides. There may also be other non - axisymmetric modes in which the lines of constant phase become curves having a center of curvature displaced from the center of the electrode. The measured values of the detectors at each frequency can be decomposed into a sum of the axisymmetric mode and the (often few) non - axisymmetric modes that reflect the main inhomogeneities of the plasma. Typically, the decomposition enables the identification of lateral mode components and / or one main "off - center" or displaced radial mode, either of which is characteristic of the plasma inhomogeneity configuration. The correlation between the plasma inhomogeneity configuration and a particular non - axisymmetric mode is made prior to the production process as part of the construction of a database that may exist in a plug - in unit or elsewhere.

[0081] Turning to FIG. 8, there is shown a top view of one exemplary azimuth sensor arrangement for a reaction chamber according to an embodiment of the present disclosure. In this embodiment, a plurality of sensors 800 may be azimuthally disposed around one or more components of the reaction chamber and / or on the reaction chamber itself. As briefly described above, in this embodiment, a plurality of sensors 800, which may be four, may be disposed at various angles about the chamber symmetry axis 805 on particular chamber components such as the showerhead and / or pedestal to measure the surface voltage or current associated with the surface wave. In this case, the intervals are 90 degrees, although in some embodiments, the azimuth intervals may be irregular.

[0082] Sensor 800 may include a passive sensor 800 that acquires a changing electric potential or magnetic field. Sensor 800 may be arranged in various orientations to detect EM waves having various types of propagation modes with respect to the chamber symmetry axis 805. Sensor 800 may be arranged at equidistant positions around the chamber symmetry axis 805 and / or components of the reaction chamber or the reaction chamber itself. Similarly, sensor 800 may be arranged diametrically opposite to each other such that the distance between sensor 800 and the symmetry axis is substantially the same. For example, the distance between sensors 800-1 and 800-2 is substantially the same as the distance between 800-3 and 800-4. Similarly, each sensor 800 is arranged at the same distance from the chamber symmetry axis 805. Examples of the spacing and location of sensor 800 will be discussed in more detail below.

[0083] As shown in the figure, the sensors 800 are arranged at diametrically opposite positions. For example, sensor 800-1 is diametrically opposite to sensor 800-3, while sensor 800-2 is diametrically opposite to sensor 800-4. The sensors 800 may find differences in waveforms on different sides of the non-axisymmetric plasma, and thus the reaction chamber and / or its components, and as described above, provide a notification so that corrective or preventive measures can be taken when such waveform differences occur. For example, if sensors 800-1 and 800-4 detect and report a difference in waveforms from diametrically opposite positions, such a difference provides an indication that the harmonics are out of phase or have different amplitudes, thereby indicating the presence of plasma non-uniformity and asymmetry. Such waveform differences occur when there are differences in the relative phase or amplitude of one or more harmonics of the signals acquired by detectors diametrically opposite.

[0084] In certain embodiments, as shown in FIG. 8, four sensors 800 may be used. However, in other embodiments, a different number of sensors 800, such as 6, 8, 12, 14, 16, 18, 20 or more sensors 800, may be used. In some embodiments, the azimuth angles between the sensors may not be equal, and nevertheless, the same characteristics of the non-azimuth-symmetric plasma mode can be observed by the sensors. In certain embodiments, it may be beneficial to have a number of sensors 800 between 6 and 12. As the number of sensors 800 increases, more data can be collected, thereby enhancing the discrimination ability against noise and the sensitivity for the recognition of non-uniformity. However, increasing the number of sensors 800 may slow down data processing, and thereby, improvements and preventive measures may occur even more slowly. One skilled in the art understands that it is possible to optimize the RF plasma process by balancing the number of sensors 800 and the desired level of data granularity. Therefore, it is beneficial to increase the number of sensors 800 as the computing power increases and the speed at which data can be processed increases. In certain embodiments, specific sensors 800 may be selectively turned off and on, thereby enabling the controller to access specific desired data. For example, in a system having a number of 8 sensors, four of the sensors may be selected and turned off, thereby reducing the amount of data generated. In other embodiments, additional sensors may be added or removed from operation, thereby changing the amount of data generated.

[0085] The sensor 800 may include various types of sensors, both circular and other geometric shapes. In certain embodiments, the sensor 800 may be circular with an area between about 0.1 cm 2 and about 10 cm 2 The sensor 800 may further include a surface insulator layer or coating for protecting the sensor 800 from the plasma or reactive species in the reaction chamber, and may include any other coatings and layers such as a Faraday shield for a current sensor, an aluminum coating, etc.

[0086] Turning to FIG. 9, a vertical cross-sectional view of a sensor azimuthally attached to a reaction chamber according to an embodiment of the present disclosure is shown. In this embodiment, the reaction chamber 940 has a symmetry axis 905 that extends longitudinally from the center of the showerhead 930 through the pedestal 935. In other embodiments, the symmetry axis 905 may extend longitudinally from the center of another electrode, such as an antenna. A plurality of sensors 900 may be arranged azimuthally around and inside the reaction chamber 940 and around or in relation to specific components such as the showerhead 930 and / or the pedestal 935 at various positions. Since FIG. 9 is a cross-section, only two sensors 900 are shown for each location, but as discussed in detail with respect to FIG. 8, more sensors 900 may be used during the implementation of the RF plasma monitoring process.

[0087] In a particular embodiment, the sensor 900-1 may be arranged around the edge or periphery of the showerhead 930. In such an embodiment, the sensor 900-1 may be arranged at least partially or completely embedded within the showerhead 900-1, and the outer surface of the sensor 900-1 may be coated with an insulating layer, thereby protecting the sensor 900-1 from the environment within the reaction chamber 940. In such an embodiment, two or more sensors 900-1 may be arranged azimuthally around the edge of the showerhead 930, preferably four or more sensors, thereby enabling the detection of non-uniformities and asymmetries in the RF plasma treatment.

[0088] In other embodiments, sensor 900-2 may be disposed along the edge of pedestal 935 within the vacuum of reaction chamber 940. As described above with respect to sensor 900-1, sensor 900-2 may be partially or fully embedded in pedestal 935, and sensor 900-2 may or may not include an insulating layer disposed on its outer surface. Further, in some embodiments, sensor 900-2 may have a dielectric protection portion that covers sensor 900-2. In addition to sensor 900-2 disposed around pedestal 935 within the vacuum, other sensors 900-3 and 900-4 may be disposed around pedestal 935 outside the vacuum of reaction chamber 940. Such sensors 900-3 and 900-4 may be disposed on a metal surface along pedestal 935 and / or its base portion. Sensor 900 may be disposed on pedestal 935 or other support structure associated therewith.

[0089] In yet another embodiment, sensor 900-5 may be disposed on the sidewall of reaction chamber 940 and / or incorporated in other ways. In such embodiments where the wall is dielectric, sensor 900-5 may be disposed on chamber outer wall 915 outside of reaction chamber 940 or incorporated into the sidewall such that sensor 900-5 is within the vacuum of reaction chamber 940. In the case of a metal wall, the sensor needs to expose a pickup on the inner surface of the wall so that it can detect the EM field inside the chamber. Other sensors 900-6 may be disposed in viewing window 920 disposed along outer chamber wall 915. In such embodiments, the viewing window sensor 900-6 may be located outside the vacuum of reaction chamber 940 or inside reaction chamber 940.

[0090] In yet another embodiment, the sensor 900-7 may be disposed, for example, in a dielectric located around the showerhead 930, while in other embodiments, the sensor 900-7 may be disposed in a dielectric located around the pedestal 935. While the specific location of the sensor 900 is discussed herein, the sensor 900 may be disposed at various other locations within and around the reaction chamber 940. For example, the sensor 900 may be disposed inside or outside the dielectric wall near the antenna or other components. The sensor 900 may be disposed at various other locations inside the metal wall of the reaction chamber 940.

[0091] In certain embodiments, a combination of sensors 900-1 to 900-7 may be used to more accurately monitor the RF plasma process. For example, the sensor 900-1 around the edge of the showerhead 930 may be combined with the sensor 900-2 around the edge of the pedestal 935. Similarly, a combination of the sensor 900-5 outside the reaction chamber 940 may be combined with the sensors 900-1 / 900-2 located within the reaction chamber 940. In yet another embodiment, combinations of positions of the sensor 900 in 3, 4, 5, 6, 7 or more variations may be used to further optimize the monitoring of the RF plasma process.

[0092] Turning to FIG. 10, a vertical cross-section of a model reaction chamber according to an embodiment of the present disclosure is shown. In this embodiment, exemplary positions of sensors 1000 disposed at a plurality of azimuth angles around a lower electrode, which is the pedestal 1035 in this example, are shown. Similar to the sensor 1000 described above with respect to FIG. 9, FIG. 10 shows sensors 1000 disposed at various positions. The sensor 1000-1 is disposed around the outer edge of the pedestal 1035. The sensor azimuth position shown as 1000-2 is disposed around within the reaction chamber 1040, and the sensor azimuth position 1000-3 is disposed around outside the reaction chamber 1040 adjacent to the viewing window.

[0093] In this embodiment, 12 sensors 1000 are shown at each position, but in other embodiments, a smaller number or a larger number of sensors 1000 may be used. Also, in addition to the positions of the sensors 1000 shown explicitly, positions of other sensors 1000 may be used to further enhance the RF plasma treatment.

[0094] Turning to FIG. 11, a vertical schematic cross-sectional view of a reaction chamber according to an embodiment of the present disclosure is shown. In this embodiment, the sensor 1100 is shown to be disposed around the antenna of the inductively coupled plasma source 1105. Accordingly, the sensor 1100 can detect an RF current or voltage from a plasma source disposed within the reaction chamber 1140.

[0095] Turning to FIG. 12, a partial cross-section of an RF plasma treatment system according to an embodiment of the present disclosure is shown. In this embodiment, the RF plasma treatment system 1200 has a pedestal 1235. The pedestal 1235 has sensors 1240 disposed along the upper outer edge of the pedestal 1235. As described above, the sensors 1240 may be disposed on the upper outer edge, may be embedded within the pedestal 1235, or may be alternately disposed around any outer edge, either inside or outside the vacuum of the reaction chamber.

[0096] The RF plasma treatment system 1200 also has a circuit 1245 connected to the sensors 1240 via a communication line 1250. When the sensors 1240 receive detected data from the RF plasma treatment system 1200, the data may be transmitted to the circuit 1245 for processing. Since the circuit 1245 is relatively close to the sensors 1240, the time required to transfer the detected data therebetween can be reduced. Accordingly, initial calculations regarding the electrical characteristics detected by the sensors 1240 may be performed more quickly and then transferred to other components 1255 of the RF plasma treatment system 1200. The other components 1255 may include, for example, an RF generator, an impedance matching network, a fault detection section, an operation controller for the reaction chamber, an operation controller for the tool, a plug-in device, a signal analysis section, or other components (s) connected to the RF plasma treatment system 1200.

[0097] Components of the plasma processing system 1200 or other components 1255 or still other components not shown can adjust the side of the RF plasma processing system 1200 to correct a fault detected by the sensor 1240 and at least partially processed within the circuit 1245. The circuit 1245 may be arranged in a pedestal 1235 outside the vacuum of the reaction chamber in a separated structure to protect the circuit 1245 from the state within the reaction chamber. In other embodiments, the circuit 1245 may be arranged in the base of the pedestal 1235 or other regions adjacent to the pedestal 1235.

[0098] Since FIG. 12 shows a cross section of components of the RF plasma processing system 1200, those skilled in the art will understand that the circuit 1245 may be arranged at substantially the same radius at various azimuth angles around the pedestal 1235. Thus, an independent circuit 1245 may be available for each sensor 1240, or the sensor 1240 may be connected to a centralized circuit 1245 arranged at one or more selected positions around and / or inside the pedestal 1235.

[0099] Turning to FIG. 13, a partial cross-sectional view of an inductively coupled RF plasma processing system according to an embodiment of the present disclosure is shown. The sensor 1340 is shown configured to be close to the inductive antenna 1330 and may be attached outside or inside a dielectric wall (not shown) adjacent to the antenna.

[0100] Turning to FIGS. 14, 15, and 16, diagrams of phases of a process for an RF plasma processing system matching network according to an embodiment of the present disclosure are shown. In FIG. 14, phase 1 of the process is shown, in which case the DC part of the generator provides a higher voltage, so the amplitude of the high frequency increases with time. Phase 1 can last, for example, about 1 millisecond according to the design function provided by the manufacturer. As shown, the forward voltage 1400 increases with each pulse, and similarly, the reflected voltage 1405 also increases with each pulse.

[0101] In FIG. 15, Phase 2 of the process is shown where the amplitude of the high-frequency voltage is constant but the matching is not adjusted. Thus, the amplitude of the forward voltage 1500 is constant as well as the amplitude of the reflected voltage 1505. In FIG. 16, Phase 3 of the process is shown where the matching is adjusted in this phase. In Phase 3, the amplitude of the forward voltage 1600 is constant. The amplitude of the reflected voltage 1605 is also constant but the amplitude of the reflected voltage 1605 is also lower. A fourth phase (not shown) may also occur in some embodiments where the processing parameters are changed, such as a change in the capacitor, and the end point is reached. During the fourth phase, the impedance may change and as a result, the voltage may change.

[0102] Turning to FIG. 17, a schematic diagram of the phases of a high-frequency processing system matching network according to an embodiment of the present disclosure is shown. As discussed in more detail above, the process can include four phases. During Phase 1 1700, the DC increases. During Phase 2 1705, the adjustment of the matching network is started but the matching network may not be fully adjusted. In Phase 3 1710, a steady state may occur where the reflected wave voltage is relatively low. In Phase 4 1715, an end time may be reached. This process may occur multiple times for a particular wafer.

[0103] Aspects of the present disclosure enable the derivation of statistics for individual wafers based on the dynamic changes of the processes represented above with respect to various phases of operation. Thus, statistics may be taken for one or more of the phases and may include signals associated with specific pulses. For example, the statistics may be prepared to measure high-frequency voltages at or about a sensor or at the output of a matching network, in or around a plasma processing system, and at components external to the matching network. The statistics may be prepared for phases regarding the high-frequency current and / or the degree between the high-frequency voltage and the high-frequency current. By completing the process for a number of wafers and aggregating the statistics, trends can be monitored. The trends may be used to determine the timing for issuing various warnings and / or intervention commands. Aspects of determining the statistics and generating warnings and intervention commands will be discussed in detail below.

[0104] Turning to FIG. 18, a graph of the voltage at the start of a high-frequency pulse of a high-frequency plasma processing system according to an embodiment of the present disclosure is shown. Before discussing the accumulation of statistical data and how the data is used, an example of how a pulse reaches a steady state is illustrated. In this figure, the x-axis represents time in microseconds, and the y-axis represents a unit measurement value proportional to the voltage. Thus, various measurements and calculations are performed, such as the purchase, the time to reach the maximum voltage, the time to reach the maximum steady-state voltage, and the steady-state voltage. Such measurements may be made for each pulse and / or a plurality of pulses, thereby enabling calculations for monitoring the average, standard deviation, slope, trend, etc. for specific values. A detailed description of the acquisition and use of such measurement values will be omitted.

[0105] Embodiments of the present disclosure can provide a method for providing data of high-frequency pulses in a high-frequency plasma processing system. The method may include measuring electrical parameters in a matching network of the high-frequency plasma processing system. Initially, the matching network may have a function of detecting the type of process occurring, and thus, the matching network knows the power, the preset position of the capacitor, the position the capacitor reaches when the reflected wave power is substantially zero, and the process identification value that may be provided by the user. Each of these known aspects may be referred to herein individually or collectively as matching network values.

[0106] The measurement of the electrical parameters may include measuring one or more aspects of the high-frequency plasma processing operation. Some of the measurements may include relatively slow variables that occur only every 10 milliseconds, such as the measurement of the position of the capacitor, while other measurements may occur relatively quickly, such as the measurement of current, voltage, and phase, as discussed in detail above. The relatively quick measurements may occur over a period of microseconds, such as less than 10 microseconds.

[0107] In operation, the method may further include determining attributes of the measurement of the electrical parameters. For each of the measured electrical parameters, identify specific features such as the time until the reflected wave power reaches a minimum value and / or the gradient of the envelope of the voltage measured in the plasma chamber or a component related to the plasma chamber and / or the time until the envelope reaches a steady state and the characteristics indicating how the voltage and / or current changed. The attributes may include one or more parameters such as characteristics of transient phenomena such as minimum value, maximum value, gradient, increase, or trend. The attributes may be determined over a set period or a predefined period based on the type of attribute specified.

[0108] In operation, the method may further comprise defining a first statistic of an attribute of a measurement of an electrical parameter. Examples of statistics may include an average, a standard deviation, a trend, etc. for a particular attribute. The statistic may thereby reflect an attribute of a particular electrical parameter over a particular period of time. The period for the statistic may be based on the type of measurement taken first. Thus, as described above, for a slow variable, the period may be longer than for a relatively fast variable.

[0109] In operation, the method may further comprise defining a second statistic based on a first statistic regarding at least one of a phase and a process. The second statistic may be represented as a collection of calculated or measured values defined by the first statistic aggregated for the phase or process. For example, the first statistics collected for phase 1 may be combined to collectively define the second statistic. A similar method of aggregating the second statistic may occur for a process, where the process may be a period of time, a type of operation, a matching network parameter, or another operational aspect of the process.

[0110] In operation, the method may further comprise providing the user with a first statistic and a second statistic. In certain embodiments, the providing may occur at a rate substantially the same as the user's data acquisition rate. Since the first statistic and the second statistic may be used individually or collectively to determine the operating conditions of the matching network or other components or aspects of the RF plasma processing system, providing the statistics to the user may enable the user to understand how the process is progressing. The statistics may be used by the matching network, components related to the matching network or the user to determine whether any changes to the process are beneficial. For example, the user may use the first statistic and the second statistic to determine that a situation is occurring that is damaging to the wafer or other aspects of the process. Thus, the user may be able to take improvement measures, stop the process, etc. to prevent the occurrence of an event. Further, the first statistic and the second statistic may be used to provide a warning or an intervention command to any of the user and / or the matching network and / or the high-frequency generator and / or the components of the plasma processing system. The start of the warning may notify the user and / or the matching network and / or the high-frequency generator and / or the components of the plasma processing system that a situation is occurring, while the intervention may take action to address the situation.

[0111] In operation, the method may further include storing a first statistic and a second statistic within the matching network. By storing the first statistic and the second statistic within the matching network, the storage location of the collected data may be available for use in other aspects of the plasma process. For example, the stored information may be used to determine the expected lifetime of a component such as a capacitor. The data may also be used to determine whether a particular action occurs by comparing one of the first statistic and the second statistic to an actual measurement of the electrical characteristic. In certain embodiments, the action may include a warning or intervention when the comparison of at least one of the first statistic and the second statistic to the actual measurement of the electrical characteristic falls within a defined criterion.

[0112] In yet another embodiment, the matching network design parameters may be adjusted based on at least one of the first statistic and the second statistic. The design parameters may include parameters that control the operation of the matching network, such as finding the minimum value of the reflected wave power or an intentional shift of an algorithm to optimize another parameter without reaching the exact minimum value. The intentional shift of the algorithm may be, for example, making the time to reach the minimum reflected wave power equal to a predetermined target value and / or the slope of the envelope of the voltage measured at the plasma chamber or a component related to the plasma chamber. Similarly, the matching network operation parameters may be adjusted based on at least one of the first statistic and the second statistic. In yet another embodiment, the first statistic and the second statistic may be grouped when they occur under a common process state that may be defined by identification information provided by a user, input power, preset position of a capacitor, and adjusted position of the capacitor. Thus, the statistics for a common process may be analyzed to determine the expected state within the RF plasma processing system.

[0113] In addition to the methods provided above, specific diagnostics may be performed in a matching network. The diagnostics may have the use of the data described above, or may have other components and devices that collect additional information. Examples of such components and methods are described in detail below in connection with FIG. 19.

[0114] Turning to FIG. 19, a schematic diagram of a matching network 1900 of an RF plasma processing system according to an embodiment of the present disclosure is shown. In this embodiment, an RF signal may be supplied to a matching network proximate to the output by a low power RF source 201. The RF signal may be used as a diagnostic signal to determine certain characteristics of the matching network, which characteristics are described in detail below.

[0115] The matching network may have a plurality of components including inductors, capacitors, sensors, etc. In the embodiment shown in FIG. 19, the matching network has attenuation elements 203 disposed on both the input side and the output side of the matching network. The attenuation element 203 may attenuate the signal, for example, by 40 decibels so that the plasma chamber at the output 1905 of the matching network or the generator on the input side of the matching network does not affect the measurements made in the matching network.

[0116] The matching network may have various sensors such as an input sensor 223 disposed on the input side of the matching network that may measure, for example, phase and magnitude. Other types of sensors may include one or more voltage sensors 212, impedance sensors 211, current sensors 213, etc. The matching network may have various other components such as variable impedance elements 221 such as capacitors and fixed impedance elements 222 such as inductors. Such components of the matching network are operationally defined below.

[0117] The low-power RF source 201 may be incorporated within the matching network or may be an add-on that plugs into a port (not shown) of the matching network. Thus, in certain embodiments, the add-on low-power RF source 201 enables the use of the same low-power RF source in multiple matching networks or allows for replacement to accommodate various operating constraints. The low-power source 201 may provide a wave having a known spectrum, such as a sine wave or an arbitrary waveform constructed in terms of frequency and phase.

[0118] The RF signal may include a spectrum of frequencies having amplitudes and phases defined relative to a single reference RF signal. The impedance at the output of the low-power RF source 201 may be measured by the sensor 211. In certain embodiments, the RF signal may pass through the blocking circuit 202. The blocking circuit 202 may block the process frequency of the matching network but not the RF signal.

[0119] The RF signal may pass through the main section of the circuit of the matching network but may not reach the output. Movement of the RF signal through the output may be prevented by the component 203. The component 203 may have a high impedance with respect to the RF signal while having a low impedance, such as less than 0.1 ohm (Ω), at the process frequency. Similarly, the RF signal may not pass through the input port of the matching network.

[0120] The integrated circuit may have one or more inductors 222 and capacitors 221. In certain embodiments, capacitor 221 may be a variable capacitor such as a variable vacuum capacitor and / or an electronic variable capacitor such as a pindioide switchable capacitor. In certain aspects of the integrated circuit, one or more sensors 212 may measure the voltage, current, and phase of the RF signal. In certain embodiments, sensor 212 may be a current sensor such as sensor 213 disposed at a location within the integrated circuit where the voltage is expected to be relatively low. In certain embodiments, sensors 212 and 213 may measure the voltage, current, and phase at the process frequency.

[0121] During operation, the RF signal may be monitored by sensors 212 and 213, and the monitored RF signal may be compared to a predefined mathematical model for the matching network. The predefined mathematical model may be incorporated into the matching network. The predefined mathematical model may include the actual values of the fixed and variable elements of the matching network. The difference in voltage measured by sensors 212 on both sides of a component may be proportional to the impedance of the component. When a discrepancy occurs between the measured values and the model during processing, the matching network may send a warning and / or intervention command to the user and / or a component of the matching network and / or the high frequency generator and / or the plasma processing system. The warning or intervention command may occur when a predefined limit or range of discrepancy occurs.

[0122] The warning may have the effect of sending a message to the user that an inconsistency has occurred, whereby the user can decide on the next step. For example, the user may choose to stop the process, or may determine that the inconsistency for stopping the process is minor. The user may change the operating parameters of the matching network in response to the change. The intervention command enables immediate action such as an automatic change of the operating parameters of the matching network or tells the user what action to take. In certain embodiments, the intervention command may automatically stop the operation of the RF plasma processing system.

[0123] In certain embodiments, if the current measured on either side of a component is not the same, a certain percentage of the current may be lost due to arc discharge. Further, if the voltage difference between two sides of a component does not correspond to the calculated voltage implied by the current and the impedance of the component, such a difference may indicate that the component is malfunctioning. For example, a high resistance may indicate that the component is aging and may require maintenance or replacement. In certain embodiments, the frequency of the RF signal may be varied over time to cause resonance at a predetermined location within the matching circuit for a given value of a particular component.

[0124] Accordingly, the systems and methods described above can provide a significantly accurate and relatively rapid diagnostic method for use in an RF plasma processing system matching network. For example, in certain embodiments, the diagnosis may occur over a period of microseconds. Thus, a user of such a matching network may have more knowledge of the functional dynamics of the matching network to ensure that the matching network functions according to the allowable manufacturing tolerances.

[0125] According to various embodiments, the above-described diagnostic method may occur when the plasma is on or when the plasma is off. When the diagnosis is performed while the plasma is on, a complete return path for the RF signal is provided. Such a diagnosis may be used to detect relatively rapid changes in impedance, such as an arc, by using a frequency that is not a power carrier. When the plasma is turned off and the diagnosis is performed, the measurement can be focused on the structures connected to the plasma reaction chamber and / or the matching network.

[0126] During operation, the above-described system and method may be used for the diagnosis of a matching network and / or an RF plasma processing system. Such a method may include supplying a diagnostic RF signal such as the above-described RF signal, and the diagnostic RF signal has a plurality of frequencies. The diagnostic RF signal may be provided to the matching network of the high-frequency plasma processing system.

[0127] In operation, the method may further include measuring the voltage, current, and phase of the diagnostic RF signal as a function of a plurality of variable capacitors of the matching network when the diagnostic RF signal propagates through the matching network. The measurement may be performed when the diagnostic RF signal propagates through the matching network circuit, and the matching network circuit may have, for example, sensors, capacitors, inductors, and other such components.

[0128] In operation, the method may further include blocking the processing high-frequency between the matching network circuit core and the source that supplies the diagnostic high-frequency signal. The processing radio signal may include a signal indicating the RF supplied to the reaction chamber during the plasma process, and the matching network circuit core may include any of the above-described matching network components and circuits including sensors, capacitors, inductors, etc.

[0129] In operation, the method may include blocking diagnostic high-frequency signals at the input and output of the matching network, thereby preventing the diagnostic RF signals from interfering with the operation of the RF plasma processing system.

[0130] In operation, the method may further include obtaining a plurality of resonance conditions from the diagnostic high-frequency signals and collecting at least one of the typical criteria of voltage, current, and phase and the statistics of voltage, current, and phase.

[0131] In operation, the method may further include comparing the voltage, current, and phase with at least one of the typical criteria of voltage, current, and phase and the statistics of voltage, current, and phase. Based on the comparison, the method may include generating at least one of a warning and an intervention based on comparing the voltage, current, and phase with at least one of the typical criteria of voltage, current, and phase and the statistics of voltage, current, and phase. Such warnings and interventions were described in detail above.

[0132] In certain embodiments, the method may include computationally obtaining at least one of a fixed component and a variable component of the matching. Obtaining such a component may thereby also enable a diagnosis to be performed to determine the function of such a component and may further be used to determine whether to require or suggest a particular warning or intervention.

[0133] FIG. 20 shows a computer processing device 1901 according to one or more examples of the present disclosure. The computer processing device 1901 may be used to implement aspects of the present disclosure, such as the methods and systems described above, having a controller or other processing device used in the implementation of the above-described embodiments. The computer processing apparatus 1901 may have one or more central processing units (a single CPU or multiple CPUs) 1906 disposed on one or more printed circuit boards (not shown). Each of the one or more CPUs 1906 may be a single-core processor (not shown independently) or a multi-core processor (not shown independently). A multi-core processor typically has a plurality of processor cores (not shown) disposed on the same physical die (not shown), or a plurality of processor cores (not shown) disposed on a plurality of dies (not shown) collectively disposed within the same machine package (not shown). The computer processing apparatus 1901 may have one or more core logic devices, such as a host bridge 1910 and an input / output (I / O) bridge 1915, for example.

[0134] The CPU 1906 may have an interface 1908 for the host bridge 1911, an interface 1918 for the system memory 1920, and an interface 1923 for one or more I / O devices such as, for example, a graphics processing unit (GFX) 725. The GFX 1925 may have one or more graphics processor cores (not shown separately) and an interface 1928 for the display 1930. In a particular example, the CPU 1906 may integrate the functions of the GFX 1925 and interface directly (not shown) with the display 1930. The host bridge 1911 may have an interface 1908 for the CPU 1906, an interface 1913 for the I / O bridge 1915 if, for example, the CPU 1906 does not have an interface 1918 for the system memory 1920, an interface 1916 for the system memory 1920 if, for example, the CPU 1906 does not have an integrated GFX 1925 or an interface 1923 for the GFX 1925, and an interface 1921 for the GFX 1925. Those skilled in the art will recognize that all or part of the CPU 1906 and the host bridge 1910 can be integrated to reduce the number of chips, motherboard footprint, thermal design power, and power consumption. The I / O bridge 1915 may have an interface 1913 for the host bridge 1910, one or more interfaces 1933 for one or more I / O expansion devices 1935, an interface 1938 for the keyboard 1940, an interface 1943 for the mouse 1945, and an interface 1948 for one or more local STORAGE device 7190 and an interface 1953 for one or more network interface devices 1955.

[0135] Each local STORAGE device 7190It may be a solid-state memory device, a solid-state memory device array, a hard disk drive, a hard disk drive array, or any other non-transitory computer-readable medium. Each network interface device 1955 may provide one or more network interfaces including, for example, Ethernet®, Fibre Channel, WiMAX, Wi-Fi, Bluetooth®, or any other network protocol suitable for facilitating network communication. The computer processing device 1901 may have one or more local STORAGE devices 7190 In addition to or instead of, it may have one or more network-connected storage devices 1960. The network-connected storage device 1960 may be a solid-state memory device, a solid-state memory device array, a hard disk drive, a hard disk drive array, or any other non-transitory computer-readable medium. The network-connected storage device 1960 may or may not be co-located with the computer processing device 1901 and may be accessible to the computer processing device 1901 via one or more network interfaces provided by one or more network interface devices 1955.

[0136] One skilled in the art will recognize that the computer processing device 1901 may have one or more application specific integrated circuits (ASICs) configured to more efficiently perform certain functions, such as hashing (not shown). The one or more ASICs may interface directly with the interfaces of the CPU 1905, host bridge 1910, or IO bridge 1915. Alternatively, a specific purpose computing system (not shown), sometimes referred to as a mining system, may be reduced to only the components necessary to perform desired functions, such as hashing, via one or more hashing ASICs to reduce the number of chips, motherboard footprint, thermal design power, and power consumption. Accordingly, one skilled in the art will recognize that one or more of the CPUs 705, host bridge 1910, IO bridge 1915, or ASICs, or various subsets, supersets, or combinations of their functions or features, may be integrated, in whole or in part, or distributed among various devices in a manner that may be modified based on an application, design, or form factor according to one or more embodiments. Accordingly, the description of the computer processing device 700 is merely exemplary and is not intended to limit the type, kind, or configuration of components that make up a computing system suitable for performing computing operations, including but not limited to, a hash function. Further, one skilled in the art will recognize that the computer processing device 1901, an application specific computing system (not shown), or a combination thereof, may be arranged in a stand-alone, desktop, server, or rack-mounted form factor.

[0137] One skilled in the art will recognize that the computer processing device 1901 may be a cloud-based server, server, workstation, desktop, laptop, netbook, tablet, smartphone, mobile device, and / or any other type of computing system, according to one or more exemplary embodiments.

[0138] Embodiments of the present disclosure may be directed to a non - transitory computer - readable medium storing computer - executable instructions and accessible by one or more processors of a computer. The computer - readable medium may be any available medium that can be accessed by a computer. By way of example, such computer - readable media may comprise RAM, ROM, EEPROM, CD - ROM or other optical disk storage, magnetic disk storage or other magnetic storage devices, or any other medium that can be used to carry or store desired program code in the form of instructions or data structures and that can be accessed by a computer. As used herein, Disk and disc include compact disc (CD), laser disc, optical disc, digital versatile disc (DVD), floppy disc and Blu - ray disc, where Disk typically magnetically reproduces data and disc optically reproduces data using a laser.

[0139] Note also that the software - implemented aspects of the subject matter claimed are typically encoded on some form of program storage medium or implemented over some type of transmission medium. The program storage medium may be a non - transitory medium and may be magnetic (e.g., floppy disk or hard disk) or optical (e.g., compact disc read - only memory or CD ROM) and may be read - only or random access. Similarly, the transmission medium may be a twisted pair, coaxial cable, fiber optic or other suitable transmission medium known in the art. The subject matter claimed is not limited by these aspects of any given implementation.

[0140] In the foregoing description, for the sake of explanation, a specific nomenclature has been used to provide a complete understanding of the present disclosure. However, it will be apparent to those skilled in the art that specific details are not required to implement the systems and methods described herein. The foregoing description of specific embodiments is presented for purposes of illustration and description. They are not intended to be exhaustive or to limit the disclosure to the precise forms described. Obviously, many modifications and variations are possible in light of the above teachings. The examples are shown and described in order to best explain the principles of the disclosure and its practical applications, to thereby enable others skilled in the art to best utilize the disclosure and various embodiments with various modifications as are suited to the particular use contemplated. It is intended that the scope of the disclosure be defined by the following claims and their equivalents. The invention disclosed in this specification includes the following. [Aspect 1] A method for providing data of high-frequency pulses in a high-frequency plasma processing system, comprising: measuring electrical parameters in the matching network of the high-frequency plasma processing system; determining attributes of the measurement of the electrical parameters; defining a first statistic of the attributes of the measurement of the electrical parameters; defining a second statistic based on the first statistic related to at least one of phase and process; providing the first statistic and the second statistic to a user; storing the first statistic and the second statistic in the matching network; and a method comprising the steps of. [Aspect 2] The method according to Aspect 1, further comprising comparing at least one of the first statistic and the second statistic with an actual measured value of an electrical characteristic in the high-frequency plasma processing system. [Aspect 3] The method according to Aspect 2, further comprising warning the user when a comparison between the actual measured value of the electrical characteristic and at least one of the first statistic and the second statistic is within a defined criterion. [Aspect 4] The method according to Aspect 2, further comprising intervening when a comparison between the actual measured value of the electrical characteristic and at least one of the first statistic and the second statistic is within a defined criterion. [Aspect 5] The method according to Aspect 1, further comprising determining the lifetime of a capacitor in the matching network based on at least one of the first statistic and the second statistic. [Aspect 6] The method according to Aspect 1, further comprising adjusting matching network design parameters based on at least one of the first statistic and the second statistic. [Aspect 7] The method according to Aspect 1, further comprising grouping the first statistic and the second statistic under a common process condition defined by at least one of process identification, input power, preset position of a capacitor, and adjustment position of a capacitor provided by a user. [Aspect 8] The method according to aspect 1, wherein the determination of the attribute of the measurement of the electrical parameter is performed in less than about 10 milliseconds. [Aspect 9] The method according to aspect 1, further comprising adjusting the consistent network operation parameter based on at least one of the first statistic and the second statistic. [Aspect 10] The method according to aspect 1, wherein the providing is performed at a speed substantially the same as the data acquisition speed of the user. [Aspect 11] The method according to aspect 1, further comprising performing an intervention when a comparison between measured values before and after a component of the consistent network exceeds a predefined limit value. [Aspect 12] A high-frequency plasma processing system, comprising: a reaction chamber; a consistent network electrically connected to the reaction chamber; a sensor disposed within the consistent network for measuring an electrical parameter within the consistent network of the high-frequency plasma processing system; a controller connected to the sensor, determining an attribute of the measurement of the electrical parameter, defining a first statistic of the attribute of the measurement of the electrical parameter, defining a second statistic based on the first statistic related to at least one of phase and process, providing the first statistic and the second statistic to a user, and storing the first statistic and the second statistic within the consistent network; A high-frequency plasma processing system comprising the above. [Aspect 13] The high-frequency plasma processing system according to aspect 12, wherein the controller further compares at least one of the first statistic and the second statistic with an actual measured value of an electrical characteristic within the high-frequency plasma processing system. [Aspect 14] The high-frequency plasma processing system according to aspect 12, further comprising a second sensor disposed within the consistent network. [Aspect 15] The high-frequency plasma processing system according to aspect 12, wherein the sensor measures at least one of voltage, current, and phase at a processing frequency. [Aspect 16] The high-frequency plasma processing system according to aspect 12, wherein the controller performs an intervention when a comparison between measured values before and after a component of the consistent network exceeds a predefined limit value. [Aspect 17] The high-frequency plasma processing system according to aspect 12, wherein the first statistic and the second statistic are provided to the user at a speed substantially the same as the data acquisition speed of the user. [Aspect 18] A method for providing diagnostics in a high-frequency plasma processing system, comprising: supplying a diagnostic high-frequency signal having a plurality of frequencies to a matching network of the high-frequency plasma processing system; measuring a voltage, a current, and a phase of the diagnostic high-frequency signal as a function of a plurality of variable capacitors of the matching network when the diagnostic high-frequency signal propagates through the matching network; blocking a processing high-frequency between a matching network circuit core and a source supplying the diagnostic high-frequency signal; blocking the diagnostic high-frequency signal at an input part and an output part of the matching network; acquiring a plurality of resonance conditions from the diagnostic high-frequency signal; collecting at least one of a typical reference of the voltage, the current, and the phase and a statistic of the voltage, the current, and the phase; comparing the voltage, the current, and the phase with at least one of a typical reference of the voltage, the current, and the phase and a statistic of the voltage, the current, and the phase; generating at least one of a warning and an intervention based on comparing the voltage, the current, and the phase with at least one of a typical reference of the voltage, the current, and the phase and a statistic of the voltage, the current, and the phase; A method comprising the above steps. [Aspect 19] The method according to aspect 18, further comprising obtaining at least one of a fixed component and a variable component of the matching. [Aspect 20] The method according to aspect 18, further comprising performing an intervention when a comparison between measured values before and after a component of the matching network exceeds a predefined limit value.

Claims

Claim 1 A method for providing data of high-frequency pulses in a high-frequency plasma processing system, comprising: measuring electrical parameters by a sensor disposed in a matching network of the high-frequency plasma processing system during plasma processing performed on each of a plurality of substrates in a reaction chamber of the high-frequency plasma processing system; determining an attribute of the measurement of the electrical parameters; defining a first statistic of the attribute of the measurement of the electrical parameters collected for at least one of a process of the matching network of the high-frequency plasma processing system and a phase included in the process; defining a second statistic by combining the first statistic; providing the first statistic and the second statistic to a user; storing the first statistic and the second statistic in the matching network; A method comprising the above steps. Claim 2 The method according to claim 1, further comprising comparing at least one of the first statistic and the second statistic with an actual measurement value of an electrical characteristic in the high-frequency plasma processing system. Claim 3 The method according to claim 2, further comprising warning the user when a comparison between at least one of the first statistic and the second statistic and the actual measurement value of the electrical characteristic is within a defined criterion. Claim 4 The method according to claim 2, further comprising intervening when a comparison between at least one of the first statistic and the second statistic and the actual measurement value of the electrical characteristic is within a defined criterion. Claim 5 The method according to claim 1, further comprising determining a lifetime of a capacitor in the matching network based on at least one of the first statistic and the second statistic. Claim 6 The method according to claim 1, further comprising adjusting matching network design parameters based on at least one of the first statistic and the second statistic. Claim 7 The method according to claim 1, further comprising grouping the first statistic and the second statistic under a common process condition defined by at least one of a process identification, an input power, a preset position of a capacitor, and an adjustment position of a capacitor provided by a user. Claim 8 The method according to claim 1, wherein the determination of the attribute of the measurement of the electrical parameter is performed in less than about 10 milliseconds.

9. The method according to claim 1, further comprising adjusting the network operation parameter for consistency based on at least one of the first statistic and the second statistic.

10. The method according to claim 1, wherein the providing is performed at a speed substantially the same as the data acquisition speed of the user.

11. The method according to claim 1, further comprising intervening when a comparison between measured values before and after components of the network for consistency exceeds a predefined limit value.

12. A high-frequency plasma processing system, comprising: a reaction chamber; a network for consistency electrically connected to the reaction chamber; a sensor disposed within the network for consistency, the sensor measuring an electrical parameter during plasma processing performed on each of a plurality of substrates within the reaction chamber of the high-frequency plasma processing system; a controller connected to the sensor, determining an attribute of the measurement of the electrical parameter, defining a first statistic of the attribute of the measurement of the electrical parameter collected for at least one of a process of the network for consistency of the high-frequency plasma processing system and a phase included in the process, defining a second statistic by combining the first statistic, providing the first statistic and the second statistic to a user, and storing the first statistic and the second statistic within the network for consistency; A high-frequency plasma processing system comprising the above.

13. The high-frequency plasma processing system according to claim 12, wherein the controller further compares at least one of the first statistic and the second statistic with an actually measured value of an electrical characteristic within the high-frequency plasma processing system.

14. The high-frequency plasma processing system according to claim 12, further comprising a second sensor disposed within the network for consistency.

15. The high-frequency plasma processing system according to claim 12, wherein the sensor measures at least one of voltage, current, and phase at a processing frequency.

16. The high-frequency plasma processing system according to claim 12, wherein the controller intervenes when a comparison between measured values before and after a component of the matching network exceeds a predefined limit value.

17. The high-frequency plasma processing system according to claim 12, wherein the first statistic and the second statistic are provided to the user at a speed substantially the same as the user's data acquisition speed.

Citation Information

Patent Citations

  • Method for monitoring plasma treatment device, and plasma treatment device

    JP2003197609A

  • Impedance matching apparatus

    JP2006166412A

  • System-level power delivery to plasma processing loads

    JP2014508378A

  • Impedance matching system and method of operating the same

    US10269540B1

  • Computation of Statistics for Statistical Data Decimation

    US20140167613A1