Device for rapidly sensing RF signals from an RF plasma processing apparatus

Abstract

To provide non-invasive sensing of an emitted RF spectrum.SOLUTION: A sensing device for monitoring electromagnetic radiation emitted from a plasma processing system includes at least two of (i) a first probe for detecting a time-varying RF electric field, (ii) a second probe for detecting a time-varying RF magnetic field, and (iii) an optical probe for detecting modulated light emission, and further includes a signal processing unit configured to receive a signal from each of the probes and to monitor electromagnetic radiation for only a single frequency of each signal.SELECTED DRAWING: Figure 5(a)

Description

Technical Field

[0001] The present invention generally relates to the analysis of alternating electromagnetic fields from RF frequency plasma systems.

Background Art

[0002] Plasma processing of materials is commonly found in modern industrial manufacturing. Common examples are the etching and deposition of layers for forming transistors during the manufacture of integrated circuits in the semiconductor industry. Plasma processing is also used in the manufacture of, for example, solar panels, flat panel displays, thin film coatings, medical devices, and the like.

[0003] Plasma is typically formed within a vacuum chamber. When the air is evacuated, a gas recipe is added to the chamber at a selected gas pressure. Energy, usually electrical energy, is supplied to the vacuum chamber to excite the gas into a plasma state. The plasma state supplies the ions necessary to modify the surface of the workpiece.

[0004] Electrical energy in the radio frequency (RF) band is commonly used to power plasma reactors. The RF range is typically from several tens of kilohertz to several hundreds of megahertz. The radio waves are coupled from an RF generator to the plasma chamber through an RF power delivery subsystem that includes a matching network to maximize power transfer. The power can be coupled to the plasma in several different ways. In one configuration, an RF powered electrode can be used to excite the plasma through an electric field (E-field) formed between the powered electrode and a counter electrode. The counter electrode can be another electrode or the chamber vessel body and is typically held at ground potential. When the E-field strength is sufficient, dielectric breakdown occurs and plasma is formed. The RF current maintains the discharge and flows between the powered electrode and ground. The electrons stripped from the parent atoms and molecules oscillate back and forth within the RF E-field and in the process ionize the background gas and thus maintain the plasma.

[0005] In another configuration, RF power is coupled to the plasma through an RF antenna. The antenna need not be in direct contact with the background gas. The RF current flowing through the antenna induces a time-varying magnetic field (H-field) perpendicular to the direction of the current flow. The H-field is generally coupled to the chamber through a dielectric window. Once breakdown occurs, the H-field induces an E-field in the plasma that drives the RF current. Free electrons oscillate in the RF field, ionize the background gas, and thus maintain the plasma. There are many other plasma reactor configurations and plasma generation mechanisms.

[0006] Regardless of the type of RF coupling mechanism, an interface region between the electrode or antenna and the plasma necessarily exists. This region is called the plasma sheath. The sheath has a non-linear RF impedance. As a result, harmonics of the fundamental wave drive frequency are excited. Consequently, the voltage and current of the plasma can have a rich harmonic spectrum. The RF harmonic signature of the plasma process is determined by many variables, including fundamental wave plasma parameters, the chemical nature of the plasma, chamber shape, chamber surface condition, and mechanical characteristics of the chamber. Therefore, the harmonic spectrum contains important information regarding the normality and performance of the plasma process.

[0007] When plasma electrons oscillate in the RF field, they ionize and / or excite neutral gas particles. Electron impact excitation events are accompanied by light emission. Much of the emitted light originates from the visible region of the electromagnetic spectrum. The light emission occurs at specific wavelengths depending on the type of gas used. In an RF plasma, the light emission intensity is modulated by the electron motion at the RF drive frequency and its harmonics.

[0008] In certain plasma reactor configurations, there are one or more drive frequencies. This results in the generation of a harmonic spectrum for each fundamental wave drive frequency and intermodulation frequencies. A sensing device is highly desirable for accurately measuring the characteristics of the frequency spectrum generated within an RF powered plasma reactor. The measured frequency / harmonic spectrum signature can be used to monitor the performance of the process in real time.

[0009] Plasma processing systems, by virtue of their design, are often "leaky" to electromagnetic fields. RF time-varying electric and magnetic fields are radiated through unshielded areas or areas that are inappropriately grounded. Luminescence is radiated through opaque areas such as viewports. This provides three means of sensing the RF spectrum radiated from a plasma processing system using a) a time-varying E-field sensor, b) a time-varying magnetic field (B-dot) sensor, and c) a time-varying optical intensity sensor.

[0010] The RF spectrum is typically represented as a graph of signal amplitude versus frequency. The signal amplitude can correspond to the time-varying E-field or B-field intensity. Since both are driven by the electron motion within the plasma system, the time-varying optical signal amplitude closely corresponds to the B-field intensity. These three signals are actually vector quantities having both phase components and amplitudes. The phase angle between the fundamental and harmonic frequencies of an individual spectrum is a valuable dataset to measure because it is very sensitive to changes in the electron motion within the plasma. Also, the phase relationship between the corresponding frequency components of the E-field and B-field spectra has been found to be very sensitive to specific physical phenomena in the plasma process. By processing and analyzing the perceived RF spectrum, the plasma process can be monitored in real time. To avoid the need for detailed calibration of the measured RF spectrum, statistical techniques can be used to set a baseline or fingerprint for specific process conditions. This is typically done for a known "normal" process. For the same setpoint, subsequent processes can be compared to the baseline to identify statistically significant changes. Overall changes can be easily detected in a single data channel; for example, a sudden decrease or increase in the E-field amplitude may indicate a problem with the RF generator. Other changes can be more subtle and may require multivariate analysis using many data channels from the RF spectrum to detect, for example, a slight air leak or a slight wafer misplacement, but still these problems can potentially affect the workpiece. Furthermore, some changes can only be detected with advanced data channels such as the phase measurements provided by the present invention. For example, phase measurements can be very sensitive to specific process endpoints, even when the layer being etched constitutes less than 1% of the surface area of the workpiece when the etching of the layer is completed.

[0011] The probe pair of the E-field and B-dot is commonly used in so-called VI sensors for measuring the voltage and current flowing through a transmission line. Publication number WO2014 / 016357A2 describes a VI sensor device. This sensor is designed to be connected in series with an RF power supply line. Therefore, it includes a section of the transmission line, as is common for this type of sensor. A broadband capacitive pickup (E-field probe) is used to determine the voltage signal of the RF line connected to the plasma. An induction loop (B-dot probe) is used to determine the RF current of the RF line connected to the plasma. The voltage and current pickups are incorporated into the RF transmission line section of the VI sensor structure. Signals representing the current and voltage are passed to an analog-to-digital converter (ADC), and the digitized signals are processed within a field-programmable gate array. Although the in-line VI sensor is a very important tool, it can be difficult to install without significantly changing the plasma system configuration.

[0012] Publication number WO2018 / 177965A1 describes a device by McNally et al. that uses a specially designed magnetic loop antenna to sense the plasma current flowing near a plasma chamber viewport from an external location. The antenna is carefully designed and calibrated. The output of the antenna is coupled to a spectrum analyzer for viewing the frequency spectrum detected by the antenna. The inventors describe frequency analysis techniques for detecting the resonance characteristics of the plasma. The antenna operating in an isolated near-field is important for the proper functioning of this antenna. A shield is used to prevent the antenna from detecting signals from matching networks or other far-field signal sources.

[0013] In Publication No. WO2004 / 006298A2, Parsons describes an invention for remotely sensing RF radiation from a plasma system using an RF antenna. The antenna can detect harmonic signals and is coupled to a processing unit for analysis. The processing unit is coupled to a plasma tool controller where the sensed RF signal is used to adjust and maintain the parameters of the plasma process based on the measured signal level.

[0014] In Publication No. US2007 / 022766, Yamazawa et al. describe an apparatus consisting of two magnetic loop antennas positioned inside the plane of a plasma chamber wall. The antennas are placed near two electrodes of a capacitively coupled plasma reactor. The voltage signals generated by the magnetic flux passing through each loop are coupled to a signal processing unit. Thus, the current flowing from the plasma to the chamber wall is calculated.

[0015] In Publication No. US6,441,620B1, Scanlan et al. describe a method for fault verification in plasma processing using data from an in-line VI sensor. For a given baseline plasma process, changes in the magnitudes of multiple Fourier components from the baseline due to changes in multiple process input parameters are determined. These magnitude changes are stored as reference data. During subsequent production runs, the plasma process is monitored for faults. If a fault is found, the baseline process is repeated using the original baseline values and nominally the same input parameter values. The changes in the Fourier components from the original baseline values are determined and compared to the reference data. This comparison is used to determine which plasma subsystem is most likely to have caused the fault.

[0016] From the above, it is clear that the prior art has many drawbacks. It is necessary to address these drawbacks.

Summary of the Invention

[0017] Disclosed herein is a sensing device for monitoring electromagnetic radiation emitted from a plasma processing system, comprising at least two of: (i) a first probe for detecting a time-varying RF electric field, (ii) a second probe for detecting a time-varying RF magnetic field, and (iii) an optical probe for detecting modulated photoluminescence, and further comprising a signal processing unit configured to receive signals from each probe and to monitor electromagnetic radiation with respect to only a single frequency of each signal.

[0018] The signal processing unit may be further configured to determine the amplitude of each signal from each probe and generate amplitude data.

[0019] The signal processing unit may be further configured to analyze the amplitude data to determine a changing average amplitude.

[0020] The signal processing unit may be further configured to identify events as corresponding to amplitudes outside a predetermined limit with respect to the average amplitude.

[0021] Optionally, the signal processing unit is further configured to store amplitude data corresponding to the identified events.

[0022] Optionally, the signal processing unit is further configured to store amplitude data for a predetermined period corresponding to the identified events.

[0023] The signal processing unit may be further configured to discard amplitude data determined not to correspond to the identified events.

[0024] The signal processing unit may be further configured to average the amplitude data of each signal over a predetermined period.

[0025] Optionally, the signal processing unit is further configured to store average amplitude data corresponding to a predetermined period.

[0026] The signal processing unit may include a quadrature phase module and a local oscillator, and the quadrature phase module is configured to multiply each signal from each probe by the signal from the local oscillator in order to select a single frequency.

[0027] The signal processing unit may include a phase-locked loop configured to adjust the local oscillator so as to track each signal from each probe between a lower limit and an upper limit of the frequency.

[0028] The signal processing unit may include a filter and an averaging module configured to convert the output signal from the quadrature phase module to an average value, remove modulation from the output signal, and generate a signal vector of each signal.

[0029] The signal processing unit may further include a vector generation block configured to rotate the phase of the signal vectors all at once to generate a voltage signal vector that is on the real axis and has a zero phase.

[0030] Optionally, the signal processing unit further includes a first-in first-out buffer configured to store the signal vectors after the processing by the vector generation block.

[0031] Optionally, when the plasma process is pulsed, the signal processing unit is configured to analyze amplitude data for at least one of the signals and store the parameters of the pulse based on the amplitude data.

Brief Description of the Drawings

[0032] Here, the present application will be described with reference to the accompanying drawings.

[0033]

Figure 1

Figure 2

Figure 3

Figure 4

Figure 5(a)

Figure 5(b)

Figure 6

Figure 7

Figure 8

Figure 9

Figure 10

Figure 11

Figure 12

DETAILED DESCRIPTION OF THE INVENTION

[0034] (Detailed Description of the Drawings) The present teachings provide an apparatus / device and method for sensing electromagnetic signals radiated from a plasma processing system, processing the sensed data in the form of an RF spectrum, analyzing the spectrum to detect statistically significant plasma process changes, and identifying fault conditions.

[0035] The sensor of the sensing device may include two or more of the following three sensing elements: a first probe for sensing a time-varying electric field, a second probe for sensing a time-varying magnetic field, and an optical probe (e.g., a high-frequency photodiode) for detecting RF-modulated luminescence. The second probe and the optical probe sense the flow of RF current within the plasma system. The first probe can be used together with one of the other two probes to detect a phase change between the two fields. When measured simultaneously and synchronously, it becomes possible to determine the phase angle between the spectral components of the electric and magnetic fields. That is, the sensing device according to the present teachings samples the signals synchronously in such a way as to preserve the phase data between the frequency and harmonics of the signals. This phase measurement value is particularly sensitive to changes in the RF impedance of the plasma. The sensing device also includes a signal processing unit. This unit is designed to monitor the phase angle between the fundamental frequency and its harmonics for each individual spectrum. These phase measurement values are particularly sensitive to changes in the chemical properties within the plasma.

[0036] In the present teachings, remote E-field and B-dot probes can be used to detect the RF fields emitted from the plasma system. For additional sensitivity, a photodiode or other optical sensor can be added. Hereinafter, this three-element sensor according to the present teachings may be referred to as a VIO sensor. A specific configuration of the VIO sensor according to the present teachings will be described below.

[0037] In this specification, the terms B dot probe and E field probe are used, but it should be understood that these terms are not intended to limit the sensing device to a "perfect" B dot probe that measures only changes along the B axis or a "perfect" E field probe that measures only changes along the E axis. In practice, the sensing devices of the present teachings function with probes that are not aligned with the E or B axis as long as the probe is sufficiently independent from the electromagnetic field plane. In the present teachings, each probe is susceptible to some influence of the other field, i.e., the E field probe detects a time-varying RF magnetic field and the B dot probe detects a time-varying RF electric field. The probes of the present teachings can more generally be described as two different electromagnetic field probes each in a different vector plane from the other.

[0038] One of the three VIO sensor signals is selected as the reference to which the others are synchronized for phase measurement. In an optimal configuration, the E field is used as the reference to which the B field and / or the optical RF spectrum are synchronized. However, any signal from any of the three probes can be selected. The reason for synchronization is to enable determination of the phase relationships between the corresponding frequency components of the E field spectrum, the B field spectrum, and the RF spectrum.

[0039] The phase of the B field RF frequency component relative to the corresponding E field RF frequency component is related to the shape of the chamber and its mechanical structure. The phase of the optical RF frequency component relative to the corresponding E field RF frequency component is more sensitive to the chemical properties of the gas than to the shape. Electrons vibrating in an electric field excite gas atoms to a higher energy state. The atoms remain in the excited state for some time. When the excited electrons decay back to the ground state, photons of light are emitted. The time until the atoms return to the ground state after being excited is considered the phase shift between the optical RF frequency component and the corresponding electric field frequency component. Therefore, it is important to measure the phase shifts of both the RF magnetic field and the RF optical signal to maximize understanding of the normality and performance of the plasma process.

[0040] In modern plasma tools, to minimize RF leakage through the viewport, the size of the viewport is reduced and an RF shield is added. The RF shield is often more effective at blocking E-field leakage but less effective at blocking B-field leakage. As long as a line of sight to the internal chamber is obtained, optical RF radiation is not blocked by the RF shield. Although photodetectors always require a viewport, as shown in FIG. 1, there are many regions in the plasma system where an electrical RF signal can be remotely detected. Examples of regions of the plasma processing system where the emitted RF signal can be detected include, but are not limited to, the following: (101) Coaxial transmission lines between the RF generator and the matching network (unit), either in-line or through a ground shield (or with the shielded region removed). (102) Inside the matching unit / circuit, e.g., by the housing wall (103) Outside the matching unit through a cooling fan (104) Through the RF housing between the matching unit and the plasma chamber (slots may be required). (105) Through a turbo pump (106) Inside the ICP source region, e.g., by the housing wall (107) Inside the plasma chamber

[0041] Assuming that multiple different locations within the plasma processing system can be selected to monitor electromagnetic radiation, the sensor of the sensing device according to the present teachings has two preferred physical configurations. In one configuration, the sensing elements are co-located within an "antenna" module. The E-field probe, B-dot probe, and high-frequency photodiode are co-located within the module or housing. This module is ideal for attachment to a large viewport that can detect all two or three signals simultaneously. This configuration is shown in FIG. 2. This shows the sensor 200 according to the present teachings within the module 201. The module 201 is located at the viewport 204 of the plasma processing system 100.

[0042] FIG. 2 also shows a sensor 200 connected to the signal processing unit 202 via a signal coupling cable, as will be described in more detail below. The combination of the sensor 200 and the signal processing unit may be regarded as a sensing device.

[0043] In another configuration, the individual sensing elements of the sensor 200 are distributed in different regions of the plasma processing system 100 where they can be most exposed to the radiation signals that they exhibit high sensitivity to. This configuration is shown in FIG. 3. In this embodiment, the photodetector / sensor 301 is located at the viewport (or optical fiber port), the B-dot sensor 302 is mounted inside the ICP plasma source housing, and the E-field probe 303 is mounted on the inner wall of the matching box or along the coaxial cable between the generator and the matching box. When installed between the generator and the matching unit, the harmonic signal level is significantly attenuated by the matching unit, so the E-field probe 303 is exposed to a significantly low harmonic signal level. Nevertheless, a sufficiently sensitive signal processing unit 202 can be used to obtain a stable E-field reference in order to enable accurate measurement of the phase change of the B-dot and / or the optical RF spectrum (when the E-field is selected as the reference).

[0044] The B-dot probe 302 comprises an inductive pick-up loop having a plurality of windings. The time-varying magnetic flux passing through the loop induces a voltage across both ends of the loop output. The loop voltage is coupled to a signal conditioning circuit (also referred to herein as a frequency response leveling circuit). The signal conditioning circuit is used to adjust the output voltage of the inductive loop such that equal amplitude magnetic fluxes over a wide frequency range induce equal or similar voltage levels at the output of the signal conditioning circuit.

[0045] The E-field probe 303 comprises a capacitive pickup. A time-varying electric field charges a capacitor and induces a voltage across its terminals. The output of the capacitive pickup is coupled to another signal conditioning circuit (frequency response leveling circuit) to achieve the same frequency leveling effect as described for the B-dot probe. The signal conditioning circuit is used to adjust the output voltage of the capacitive pickup such that equal amplitude electric fluxes over a wide frequency range induce equal or similar voltage levels at the output of the signal conditioning circuit.

[0046] The photodetector 301 can be a photodiode. The light intensity reaching the photosensitive area of the photodetector induces a certain voltage at the output. The photodetector 301 typically exhibits high sensitivity to light in the wavelength range of 200 nm to 1000 nm, although other ranges can be used if necessary. The detector is designed to have a flat response to light of the same intensity over the entire wavelength range within the specified optical bandwidth. In other words, the voltage generated at the output of the photodetector is approximately equal at different wavelengths of the same intensity. The photodetector has a high frequency bandwidth. This means that the output of the photodetector can respond to time-varying light intensity in the frequency range of several 10 kHz to several 100 MHz and is designed to have a flat frequency response in a similar manner to the B-dot and E-field sensors. Continuously adjusting the response of the sensor over the target frequency range prevents problems such as saturation due to resonance effects at specific frequencies.

[0047] Turning to FIG. 4, this shows a remote non-invasive sensing device according to the present teachings having co-located sensing elements and a signal processing unit. That is, FIG. 4 shows the internal configuration of the sensing device, namely the sensor 201 and the signal processing unit 202 of FIG. 2.

[0048] The sensing elements (optical sensor 301, E-field sensor 303, and B-dot probe 302) and the frequency response leveling circuits 402, 402, 403 constitute the analog front end of the VIO sensor according to the present teachings. Each leveling circuit outputs an analog voltage signal proportional to the quantity sensed by the respective sensing elements 301, 302, 303. The analog voltage output is an alternating current (AC) signal in the RF band. A signal processing unit 202 is incorporated in the VIO sensing device to extract the frequency spectrum in a form that can be analyzed and visualized in a useful way. Individual coaxial cables (which can be bundled in a common sheath) carry the AC signal to the signal processing unit 202. A multi-channel ADC is used to sample the signals from each of the sensing elements. A reference signal (which can be, for example, the E-field sensor, but any one of the sensors can be selected as a reference) is coupled to channel 1 of the ADC. The other two channels are synchronized with channel 1. It should be understood that the selected reference signal can be fixed, for example, it can always be the signal from the E-field sensor, or the signal processing unit can dynamically select the reference signal using each measurement or sensing procedure. The strongest or highest intensity signal may be selected, or some other criterion may be used by the signal processing unit to select the reference signal. Any suitable criterion can be selected by one of ordinary skill in the art.

[0049] The high-speed ADC samples the AC waveform generated by the analog front end. Data sampling is performed simultaneously or synchronously on all channels. Sampling is performed synchronously to preserve the phase data between the frequencies and harmonics of the two signals. A person skilled in the art can select appropriate components and specific techniques to ensure that the phase data and harmonics are properly preserved. Typically, a data block of 512 samples is recorded as a first step. The block size is arbitrarily selected and can be varied to meet various requirements. The data block is transferred to a field programmable gate array (FPGA), where a fast Fourier transform (FFT) is performed. The FFT converts the time-domain AC waveform into a frequency spectrum. The frequency spectrum is sent to a microprocessor (not shown) of the signal processing unit for storage and further processing, including averaging. To reduce the signal-to-noise ratio, multiple FFTs are averaged together. Channel phase information is also preserved by averaging.

[0050] In the present teaching, the AC waveforms are asynchronous block by block but are sampled synchronously channel by channel, that is, each time a new waveform is recorded on any one channel, sampling does not start at the same point in the AC cycle, but sampling occurs exactly simultaneously on all channels. As a result, the phase with respect to 0 degrees of the fundamental frequency component and its harmonics changes each time a new FFT is processed. To perform averaging, a phase rotation operation is carried out. The fundamental frequency component of the first ADC channel is selected as a reference. The phase of the composite fundamental frequency component is rotated to a predetermined phase angle φ. The phase shift Δφ used to rotate the composite fundamental frequency component from the measured value to a predetermined value is recorded. The composite fundamental frequency components from all individual ADC channels are also rotated by Δφ. All harmonic frequency components (N) are rotated by N×Δφ, where N indicates the harmonic number and N = 1 is the first harmonic or the fundamental frequency. This aligns each FFT in the phase space, enabling averaging over the entire series of samples, while fulfilling the dual purpose of preserving the phase relationship between the harmonics within each individual spectrum and the phase relationship between the corresponding frequency components of the spectra from different ADC channels.

[0051] When a preset number of averages are completed, the signal processing unit 202 outputs the data set to the user for viewing and further analysis. That is, the data set can be output by the sensing device in the form of a result table to an external computer resource such as a PC. However, further analysis may also be performed (on-board) by the sensing device.

[0052] The data set or result table includes the frequency of each signal fundamental, the magnitude of each signal, the phase of each signal with respect to the selected reference signal, the magnitude of each signal harmonic, the phase of each signal harmonic with respect to the corresponding harmonic on the reference signal, and the phase of the reference signal harmonic with respect to the fundamental of the reference signal for each signal.

[0053] The following exemplary result tables correspond to two signals, signal frequency 1 and signal frequency 2, respectively. These tables show that for two or more signals, the result table of the first signal is essentially reproduced for any additional signal (signal frequency). [Table 1] [Table 2]

[0054] When data is output, information on the amplitude and phase of each frequency component of each sensor channel is reported. Typically, the phase difference between relevant frequency components is calculated and output to the user. The signal processing unit is designed to be able to process multiple fundamental frequency components simultaneously. This is particularly useful when monitoring plasma processes powered by two or more frequencies. The signal processing unit can typically output a complete data set at a selectable speed in the range of about 1 millisecond to 1 second, depending on the requirements of the application.

[0055] Not all RF plasma processes operate in continuous wave (CW) mode. Some are pulsed at frequencies in the range of a few hertz to several tens of kilohertz. For this reason, the signal processing unit 202 may have a built-in external synchronization port. The unit 202 receives a (TTL) signal input from the pulse RF generator of the plasma processing system. The boxcar (averaging) technique enables the capture of the RF waveform at specific times during the pulse RF period. In this scenario, averaging is performed over multiple pulses. Using this technique, an average pulse profile can be constructed with a resolution of 1 microsecond, if necessary. Apart from synchronization, signal processing is performed in the same way. The pulsed RF signal may have one or more power levels during the repeating cycle, but once the TTL synchronization signal becomes available, the signal processing unit can perform the analysis without being hindered.

[0056] In recent years, pulse RF plasma processes have been developed such that the fundamental wave carrier frequency is not fixed but rather dynamic. This enables better control over power matching over the entire pulse period where mechanical movement of the matching position at the required speed is not practical. Typically, a dynamic range of + / - 10% around the carrier frequency is sufficient. The signal processing unit according to the present teachings also enables the dynamic carrier frequency for continuous wave and pulse RF process monitoring. A smart FFT process tracks the movement of the carrier frequency within frequency bins adjacent to where the fundamental frequency is predicted to appear. Thus, the spectrum of the carrier frequency can be acquired at any point within its + / - 10% dynamic tuning range.

[0057] For high-speed event monitoring, another signal processing method can be used by the sensing device (sensor and signal processing unit) according to the present teachings. The aforementioned frequency spectrum analysis requires sampling of the composite waveform and further FFT processing. FFT is a relatively slow process. Thus, a root mean square (RMS) detection technique using a narrow bandpass filter to isolate the fundamental frequency can be used. It should be understood that the sensing device according to the present teachings can operate in two modes, namely (i) the frequency spectrum mode described above, and (i) a high-speed detection mode described below.

[0058] In the high-speed detection mode, the root mean square (RMS) amplitude of each data channel can be determined in real time at high speed (typically a reporting rate of less than 1 microsecond), and high-speed amplitude data is captured. In this high-speed method, measurement of the harmonic spectrum is not possible. However, real-time high-speed processing enables monitoring of the pulse profile without the need for synchronization with a TTL signal. Individual pulse profiles can be captured and analyzed in comparison with the aforementioned average pulse profile obtained by the boxcar technique. Important pulse characteristics such as duty cycle and pulse repetition frequency can be determined for each pulse to construct a correlation metric for defects or faults. Pulse profile monitoring is just one type of high-speed measurement for which the high-speed detection method can be used. Another important type of high-speed event is an RF arc. Arcs are common in industrial plasma processing chambers and occur for numerous reasons. A common type is the microarc. Dielectric particles, for example, form a nanolayer on the chamber wall. When the outer surface of the layer is charged to a certain level, dielectric breakdown of the layer occurs. The arc burns through the layer and reaches the grounded wall behind. At that time, particles are released from the surface, which can pose a significant risk to the wafer being processed. These arcs cause sudden and sharp changes in the voltage and current of the plasma. The severity of the change can be classified from the perspective of the length of time it persists and the relative change in the level of the induced voltage or current. Since it is often a precursor to large-scale or catastrophic arc events that can damage the processing tool, it is very important to measure the signs of microarc discharges. A signal processing unit operating in the high-speed detection mode is ideally suited for detecting such microarc discharge events.

[0059] Furthermore, the high-speed mode provides complete "temporal coverage" of the signals from the sensors / probes. Well-known methods use block techniques that monitor portions of the signal's time, similar to looking through a picket fence, and it is possible to confirm what lies behind without achieving 100% coverage. That is, the prior art relies on discrete sampling, while the present teachings provide continuous monitoring. For event detection, it is important to have 100% temporal coverage of the signal so that events are not missed. That is, continuous monitoring of the entire plasma process can be provided, and amplitude data for the entire process is acquired. The signal processing unit is configured to continuously determine the amplitude of each signal from each probe and (continuously) generate amplitude data.

[0060] Figure 5(a) shows the flow of data from the ADC to the memory for the high-speed operating mode. This all occurs within the VIO sensing device and is performed by a combination of the FPGA and the on-board CPU within the signal processing unit 202. In this exemplary embodiment, all high-speed amplitude data is transferred in blocks from the FPGA to the CPU with a resolution of 1 microsecond. That is, the high-speed amplitude data is transferred from the preprocessed data memory 501 to the CPU's event detection and classification module 502, pulse measurement module 503, and data averaging module 504. The CPU checks for events and locally stores up to 5 ms of data in the RAM (event data memory 505) for each detected event. It will be understood that any length of time can be appropriately selected, and 5 ms is merely an example. In event detection techniques, a moving (changing) average is used to compare with the latest value. In the exemplary embodiment, the moving average is simply the average of the last 1000 values, i.e., the last 1 millisecond. Events are classified as corresponding to amplitudes outside of user-defined limits with respect to the moving average.

[0061] The CPU also averages the measured RMS signal levels and stores the averaged data every 100 milliseconds (in the context data memory 506). Here too, the choice of 100 milliseconds is arbitrary, and any period can be selected. This “context” average, unlike the moving average described above, is, in the exemplary embodiment, the average of the points within each 100 - millisecond block.

[0062] The CPU checks for arc and RF pulse events and stores an arc and / or pulse snapshot of each event in the event data memory 505. That is, the CPU processes all the data in the first - in - first - out buffer (FIFO) 509 and stores only the data corresponding to the events in the event data memory. The average “context” data is stored such that the signal processing unit records the average rms value (i.e., one data point every 100 milliseconds).

[0063] Regarding the pulse (profile) measurement module 503, it provides high - resolution capture of out - of - specification pulses for storage in the event data memory 505. The pulse measurement module 503 provides pulse statistics every 100 milliseconds (i.e., period, duty cycle, average “on” value, maximum and minimum “on” values) to the context data memory.

[0064] The software running on the connected PC can retrieve data from the sensor's event data memory and context data memory and save it to a file on the connected PC.

[0065] It should be understood that the signal processing unit need not capture and store both event data and context data. Depending on the user's requirements, only one of these processes may be performed. However, it is advantageous to capture context data along with event data. Having "context" data provides a way to determine what else was happening when an event occurred, i.e., whether the event occurred immediately after power was applied. Whether the voltage and current were different than normal before and after the event (compared to previously executed plasma processes).

[0066] How the high-speed amplitude data stored in memory 501 is captured and processed using ADC 506, RMS detector 507, and filter and averaging block 508 will be described in more detail with respect to FIG. 5(b).

[0067] The sensing device may also be provided with a communication interface. This can be used to stream data to an external computing resource (e.g., a PC). This interface also enables the computing resource to request data rather than having data streamed to it.

[0068] The block diagram shown in FIG. 5(b) shows the FPGA logic used for the aforementioned high-speed (RMS). The FPGA logic shown in FIG. 5(b) used for RMS detection is very different from the logic used for harmonic spectrum detection (spectrum mode). It will be understood that the configurations described herein are merely exemplary and that those skilled in the art may use alternative logic configurations to achieve the same high-speed detection mode of the sensing device according to the present teachings.

[0069] Regarding FIG. 5(b), the signal processing unit of the high-speed mode sensing device tracks only one fundamental frequency and determines the amplitude and phase of the sensor signal at an update rate faster than 1 microsecond. This data is stored in a first-in first-out (FIFO) buffer 509 that is ready to be retrieved by the CPU. The CPU interface enables the FPGA registers to be memory mapped within the CPU memory space. The CPU sets up a direct memory access (DMA) transfer from the FPGA's FIFO directly to memory.

[0070] The RMS FPGA logic has two clock domains, one related to the ADC clock sampling rate and the other clock domain related to the CPU interface. The high-level diagram shown in FIG. 5(b) shows the signal path from the ADC 506 to the FIFO memory 509. The quadrature phase (IQ) block 510 multiplies the V, I, or O signal (signals from individual sensing elements) by a local oscillator signal that effectively selects the frequency to be monitored.

[0071] The phase-locked loop (PLL) control logic 511 is used to adjust the local oscillator to track the incoming signal between the lower and upper limits of the frequency. The filter and averaging block 508 converts the instantaneous IQ signal to an average value. The IQ output includes modulation signals related to the sampling frequency and the signal frequency in addition to the required "DC" signal. The filter and averaging logic 508 removes the modulation from the IQ output signal to generate the VIO signal vector. The VIO signal vector is then passed to the vector generation block 512 every microsecond, where the VIO signal vector is phase rotated all at once so that the V signal is real, i.e., the vector lies on the real axis and has zero phase. Then, V, and I r and I i and / or O r and O i (subscripts r and i represent the real and imaginary components respectively) are stored in the FIFO 509 that is ready to be read by the CPU of the signal processing unit.

[0072] One of ordinary skill in the art will appreciate that the high-speed (RMS) detection techniques described herein are not limited to the VIO sensing devices of the present teachings, and can be used with well-known prior art sensors such as the aforementioned VI sensor devices, or any sensor capable of supplying an appropriate signal. Not all signals (V, I, and O) are required for this high-speed detection technique.

[0073] The VIO remote sensing device according to the present teachings can be calibrated to provide absolute measurements of the E field, B dot field, and modulated light intensity, but this is not essential for the types of applications described herein. The following description presents a method for setting a baseline for a known normal process that relies only on relative signal strengths. A series of process executions from a known normal process are used to determine an acceptable process window. The normal process window follows the normal distribution of the measured variables. The baseline setting process involves measuring samples of the distribution of all variables. Statistical techniques are used to determine whether the execution of a new process is within the acceptable process window. To facilitate statistical analysis, a database is employed where each measurement is timestamped and stored for later retrieval. This statistical analysis can be performed on an on-board or external computer.

[0074] In one example of the baseline setting sequence, for a selected sample size (n), it is necessary to calculate the mean (μ) and standard deviation (s) for each variable. These parameters are stored in memory. The distance from the mean, measured by the standard deviation, can be determined for each new measurement. Other methods can also be used.

[0075] Individual variables may not be sensitive to specific types of process variations. Each frequency spectrum can include 15 harmonic amplitudes along with 15 associated phases for each fundamental frequency. Using three ADC data channels and spectral interphase parameters, hundreds of variables are acquired in each measurement. Therefore, a multivariable model is implemented to obtain optimal sensitivity. The approach of using the distance from the average analysis is just one of many ways that can be used to determine the deviation within a multivariate data set when compared to a baseline sample. Principal component analysis techniques and neural networks may be used. The sensitivity can also be further improved by using the derivative of the measured data set.

[0076] Due to the large number of variables that a VIO sensing device can generate, it can sometimes be useful to include only the most sensitive variables in the calculations to obtain optimal sensitivity. Alternatively, all of the measured variables can be included. Figure 6 shows a spectral fingerprint. In this example, after setting a baseline for the plasma processing chamber, a set of five wafers was sequentially passed through the processing chamber. The X-axis indicates the spectral component number, where 1 - 15 are the E-field amplitudes of the fundamental wave and the subsequent 14 harmonics, 16 - 30 are the B-dot amplitudes of the same frequency components, and 31 - 45 are the phase angles between channels for each frequency component. Element 46 of this spectrum is the result of the multivariate distance from the average calculation for each wafer. The calculated values have a standard deviation of approximately 1 for all five wafers, indicating that the process was in a normal state throughout. To prevent false alarms, a threshold can be implemented, for example, only distances from the average exceeding 6 sigma are considered a fault (fault score).

[0077] The spectrum shown in FIG. 6 has 45 spectral components (and the results of the defect score calculation). Including the optical RF spectrum, the number of spectral components becomes 75. Further, adding the harmonic phase measurements for the fundamental frequency adds an additional 14 spectral components per live data channel. This provides a data set that is very sensitive to most of the defects that may be encountered during plasma processing. The defect score method provides a relatively simple defect detection technique in which a large data set is condensed into a single defect score with statistical significance. However, it is also possible that different defects can return the same score. When defect classification is required, a more sophisticated approach is needed. For defect classification, spectral pattern recognition techniques can be used so that intelligent algorithms can be trained to recognize specific defect signatures or fingerprints.

[0078] Using the sensing devices and methods described so far, various process defects and events can be detected, including the following. a) The condition of the chamber walls b) Process endpoints c) Malfunctions of pressure control valves, gas flow problems, leaks d) Wafer displacement e) RF arc events f) Abnormal RF pulse events

[0079] The techniques described herein can be performed using data from the sensing device by the sensing device or by an external computer connected to the sensing device.

[0080] It should be noted that the categories of e) and f) above do not require the same baseline setting process as described above. The condition of the chamber wall is an important consideration for many plasma processes. During the etching or deposition of materials in wafer manufacturing, layers are formed on the chamber wall. These layers can significantly change the plasma impedance, especially when they are non-conductive. This can change the characteristics of the plasma and cause the process conditions to exceed the acceptable process window. When it is known that the chamber wall is clean, the baseline of the process can be set using the method described above. When the chamber wall reaches a predetermined contamination level, a chamber cleaning process is executed. Accurate feedback about when the chamber is clean enough is highly desirable. Figure 7 shows the defect score calculated for the chamber during the cleaning process. A score of 2σ was determined to represent a sufficiently clean chamber. This technique is used to warn the user when the chamber reaches a clean state. In the example shown in Figure 7, the chamber reached a clean state after approximately 800 seconds.

[0081] Plasma processing often involves multiple etching and deposition steps to create features and structures on a substrate surface. The etching step typically requires complete removal of one material layer to expose the underlying material. The etching process duration must be carefully selected so that the features are created within the tolerance of the dimensional specifications. If the etching process is not run long enough, the layer will be under-etched, i.e., the layer will not be completely removed to expose the underlying layer. If the etching process is run too long, the layer will be over-etched, i.e., the target layer is completely removed but the underlying layer may be damaged when the etching process continues. The term etching endpoint is used to define the point in time when the etching process has completely removed the target layer. Immediately after the layer is completely removed, the plasma process starts to etch the underlying material. Optical emission spectroscopy (OES) techniques are often used to monitor the plasma composition and detect the presence of particles from the underlying material. Once these particles are detected, it can be considered that the etching of the target layer is complete and the process can be correctly terminated. When the concentration of foreign particles is very low, OES does not always function. It has been shown that the defect score technique using the VIO sensing device according to the present teachings can exhibit higher sensitivity for low open area etching endpoints compared to the OES technique. Figure 8 shows an example of an etching endpoint. Using data from the beginning of the process, a baseline of the process is set. It is determined that the 4σ score is the optimal time to end the process. The RF harmonic spectrum, especially the harmonic phase component (with respect to the fundamental wave), shows very high sensitivity to small changes in the chemical properties of the plasma seen at the etching endpoint. Changes in the plasma composition due to new materials entering the process can be detected to very low levels using the RF spectrum and the T-score method already described. The optical RF spectrum detector of the present invention should not be confused with the standard OES techniques mentioned. In the present invention, an RF spectrum is captured by measuring the modulation of the light intensity using a high-frequency bandwidth photodiode.The standard OES system monitors the time-averaged optical intensity emitted from the plasma process and divides it into a spectrum of optical wavelengths. Specific wavelengths are used to identify various atomic or molecular species. Thus, OES is useful for determining the chemical composition of the plasma process. However, when the concentration of the species is low, the limit of resolution becomes a problem.

[0082] Many plasma processing chambers are operated under vacuum. The chamber pressure is precisely regulated. The gas flow and gas concentration are also precisely controlled. Malfunctions in the pressure and / or gas control system can have catastrophic consequences for the substrate being processed. To test the sensitivity of the sensing device according to the present teachings for detecting problems in the gas flow, control experiments were conducted in a plasma processing chamber that runs a silane plasma for silicon deposition applications. A baseline spectrum was recorded. The silane concentration was increased by 10% and decreased by 10% relative to the baseline conditions, while keeping all other parameters constant. For both conditions, RF fingerprints were recorded. The fingerprint fault score space of the 45-channel RF spectrum is shown in FIG. 9. Channel 46 shows the result of the multivariate fault score. There are obvious differences in the spectral fingerprints, and the fault scores exceed 5σ in both cases, making it possible to easily detect changes in the silane concentration. Much lower concentration fluctuations are detectable across a wide range of plasma gas chemistries.

[0083] The positioning of a wafer on a pedestal (or electrostatic chuck) of a plasma processing device is extremely important for wafer processing. There are several reasons why a wafer can be displaced on the chuck. A robotic arm can malfunction and place it in the wrong position. More generally, debris from other parts of the tool can fall onto the chuck, and then the wafer placed thereon can be lifted from the chuck surface. Regardless of the cause, if the displacement of the wafer is not detected, the wafer may be processed incorrectly and discarded from the line. The displacement of the wafer affects the plasma impedance and the resulting harmonic spectrum. FIG. 10 shows the effect on the B dot field measured remotely using a VIO sensor device. The main chart shows a plot of the magnitude of the fundamental wave B dot field versus time during the plasma process. Data for six correctly placed wafers are shown together with data for the displaced wafers. The processing of the displaced wafers was shortened. The inset in FIG. 10 also shows the harmonic spectra of the E field signals and B dot signals for both correctly placed and displaced wafers. In this example, the differences are very clear without the need for statistical analysis. Nevertheless, this example demonstrated the ability of the VIO sensing device according to the present teachings to detect misaligned wafers and prevent wafer discard events.

[0084] As described above, when the central signal processing module is configured for high-speed (RMS) detection, short-lived (RF) events can be detected remotely. FIG. 11 shows an example of such an event, an arc, detected in a plasma process. In particular, FIG. 11 shows the upper graph, indicating the presence of an arc within a typical plasma process. The lower graph in FIG. 11 shows a high-resolution view of the arc with a resolution of 1 microsecond.

[0085] It will be appreciated that the VIO sensing device described herein provides a non-invasive method for RF plasma arc detection. Early detection of arc discharges within a plasma processing chamber is highly desirable as it can avoid product scrap if corrective action is taken. Arc signatures are present in the E-field signal, B-dot signal, and optical RF signal. Thus, as shown in FIG. 3, one or all of the sensing elements can be used to remotely detect arcs at various locations around the plasma system.

[0086] The pulsed RF plasma process is widely used in the manufacture of advanced semiconductor nodes. By pulsing the plasma, access to various plasma chemical properties that cannot be achieved in continuous wave RF mode is provided. These plasmas are typically pulsed in a frequency range of a few Hertz to several tens of kilohertz. For the process yield, it is essential to achieve highly reproducible plasma conditions for each pulse. To achieve reproducible plasma conditions, the variations in pulse characteristics must be kept within specific thresholds. The number of wafer defects is known to be correlated with, for example, pulse variations. The high-speed (RMS) detection mode of the VIO sensing device is ideally suited for analyzing each pulse profile to extract key pulse characteristics such as pulse repetition frequency, duty cycle, pulse-on time, etc. The signal processing unit (pulse profile measurement module 503) can sample the incoming data signal every microsecond (although other time frames are also possible). This provides a detailed snapshot of each pulse profile within the target pulse frequency range. Each pulse profile is analyzed in real time and the important pulse parameters are stored. A pulse profile containing 1-microsecond samples is discarded if it is within the tolerance limits set by the user. On the other hand, if the pulse is outside the tolerance range, the high-precision pulse profile is stored for further analysis. This "exception"-based approach limits the amount of data generated, i.e., only the high-precision profiles are stored for pulses that do not meet the tolerance criteria. FIG. 12 shows a pulse profile captured by the B-dot sensor of the VIO sensing device according to the present teachings. This was an irregular pulse diagnosed as having a duty cycle outside the acceptable tolerance range. In particular, FIG. 12 provides an example of capturing a single pulse profile with a 1-microsecond resolution in a 5-millisecond window.

[0087] It should be understood that the statistical analysis and processing of the detection data or RF spectrum described with respect to FIGS. 6-12 above can be performed locally on the sensing device of the present teachings or remotely in computing resources (e.g., a PC) connected to the sensing device.

[0088] In view of the above, it will be understood that the present teachings provide a device and method for sensing RF signals emitted from a plasma processing system, processing the signals in the form of an RF spectrum, analyzing the spectrum to determine a plurality of amplitude and phase components, and performing a statistical analysis to identify a fault condition of the plasma processing system based on variations in the amplitude and phase components.

[0089] The sensor of the sensing device is composed of an E-field probe, a B-dot probe, and / or an optical RF detector. The E-field probe (or any of the probes) can be used as a reference for other probes and is used to sense a time-varying electric field. The B-dot probe and the optical RF detector are used to sense the movement of time-varying electrons or plasma current in the reactor. The individual probes may be co-located or distributed at suitable locations around the plasma system.

[0090] Unlike prior art designs, the sensing device is not coupled to a spectrum analyzer or a network analyzer. These analyzers are expensive and cumbersome to use. Also, they do not enable the advanced detection capabilities required. Instead, a signal processing unit is used. The signal processing unit provides new information regarding remotely sensed RF spectra that has not been reported in prior art remote RF plasma sensor designs. The new measurements are: a) harmonic phase, phase between spectra from different sensors, spectral analysis in pulsed RF and frequency-modulated plasma processes, RMS detection of arcs, and RMS detection of individual pulsed RF profiles. Statistical techniques are described based on spectral fingerprints of known "normal" plasma process conditions. Variations in the phase and amplitude of spectral components are analyzed and a fault score is generated for each new process measurement. Thus, a threshold can be set to alert the user of detected process faults. The phase measurements are particularly sensitive to small changes in the chemical nature of the plasma and the impedance of the plasma. The ability to measure phase makes the sensing device of the present teachings, as described, a very useful diagnostic tool for detecting subtle process changes that occur, for example, during low open area endpoint determination where standard techniques are currently inadequate.

[0091] Three means are provided herein for sensing the RF spectrum radiated from a plasma processing system using: a) a time-varying E-field sensor, b) a time-varying magnetic field sensor, and c) a time-varying optical intensity sensor. The sensing device according to the present teachings provides non-invasive sensing of the radiated RF spectrum through these three mechanisms or pairs thereof, which simultaneously and synchronously provide additional data channels not provided by prior art designs.

[0092] The invention is not limited to the embodiments described herein and can be amended or modified without departing from the scope of the invention.

Claims

1. A sensing device for monitoring electromagnetic radiation emitted from a plasma processing system, comprising: (i) a first probe for detecting a time-varying RF electric field; (ii) a second probe for detecting a time-varying RF magnetic field; and (iii) at least two of optical probes for detecting modulated photoluminescence, further comprising a signal processing unit configured to receive signals from each probe and monitor the electromagnetic radiation with respect to only a single frequency of each signal.

2. The sensing device according to claim 1, wherein the signal processing unit is further configured to continuously determine the amplitude of each signal from each probe to generate amplitude data.

3. The sensing device according to claim 2, wherein the signal processing unit is further configured to analyze the amplitude data to determine a changing average amplitude.

4. The sensing device according to claim 3, wherein the signal processing unit is further configured to identify events corresponding to amplitudes outside a predetermined limit with respect to the average amplitude.

5. The sensing device according to claim 4, wherein the signal processing unit is further configured to store amplitude data corresponding to the identified events.

6. The sensing device according to claim 5, wherein the signal processing unit is further configured to store amplitude data for a predetermined period corresponding to the identified events.

7. The sensing device according to claim 6, wherein the signal processing unit is further configured to discard amplitude data determined not to correspond to the identified events.

8. The sensing device according to any one of claims 2 to 7, wherein the signal processing unit is further configured to average the amplitude data of each signal over a predetermined period.

9. The sensing device according to claim 8, wherein the signal processing unit is further configured to store the averaged amplitude data corresponding to the predetermined period.

10. The sensing device according to any one of claims 1 to 9, wherein the signal processing unit includes a quadrature phase module and a local oscillator, and the quadrature phase module is configured to multiply each signal from each probe by a signal from the local oscillator to select the single frequency.

11. The sensing device according to claim 10, wherein the signal processing unit includes a phase-locked loop configured to adjust the local oscillator so as to track each signal from each probe between a lower limit and an upper limit of a frequency.

12. The sensing device according to claim 10, wherein the signal processing unit further includes a filter and an averaging module configured to convert an output signal from the quadrature phase module into an average value, remove modulation from the output signal, and generate a signal vector of each signal.

13. The sensing device according to claim 12, wherein the signal processing unit further includes a vector generation block configured to rotate the phase of the signal vectors all at once to generate a voltage signal vector that is on the real axis and has a zero phase.

14. The sensing device according to claim 13, wherein the signal processing unit further includes a first-in first-out buffer configured to store the signal vectors after processing by the vector generation block.

15. The sensing device according to claim 1, wherein when the plasma process is pulsed, the signal processing unit is configured to analyze amplitude data for at least one of the signals and store pulse parameters based on the amplitude data.

Images