WO2011002810A2 - Procédé de construction d'un algorithme de points finaux optimaux - Google Patents

Procédé de construction d'un algorithme de points finaux optimaux Download PDF

Info

Publication number
WO2011002810A2
WO2011002810A2 PCT/US2010/040477 US2010040477W WO2011002810A2 WO 2011002810 A2 WO2011002810 A2 WO 2011002810A2 US 2010040477 W US2010040477 W US 2010040477W WO 2011002810 A2 WO2011002810 A2 WO 2011002810A2
Authority
WO
WIPO (PCT)
Prior art keywords
endpoint
potential
signatures
optimal
real
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Ceased
Application number
PCT/US2010/040477
Other languages
English (en)
Other versions
WO2011002810A4 (fr
WO2011002810A3 (fr
Inventor
Jiangxin Wang
Andrew James Perry
Vijayakumar C. Venugopal
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Lam Research Corp
Original Assignee
Lam Research Corp
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Priority claimed from US12/555,674 external-priority patent/US8983631B2/en
Application filed by Lam Research Corp filed Critical Lam Research Corp
Priority to JP2012518588A priority Critical patent/JP5693573B2/ja
Priority to CN201080027296.6A priority patent/CN102804353B/zh
Priority to KR1020117031561A priority patent/KR101741271B1/ko
Priority to SG2011085149A priority patent/SG176566A1/en
Publication of WO2011002810A2 publication Critical patent/WO2011002810A2/fr
Publication of WO2011002810A3 publication Critical patent/WO2011002810A3/fr
Publication of WO2011002810A4 publication Critical patent/WO2011002810A4/fr
Anticipated expiration legal-status Critical
Ceased legal-status Critical Current

Links

Classifications

    • H—ELECTRICITY
    • H10—SEMICONDUCTOR DEVICES; ELECTRIC SOLID-STATE DEVICES NOT OTHERWISE PROVIDED FOR
    • H10P—GENERIC PROCESSES OR APPARATUS FOR THE MANUFACTURE OR TREATMENT OF DEVICES COVERED BY CLASS H10
    • H10P50/00—Etching of wafers, substrates or parts of devices
    • H10P50/20—Dry etching; Plasma etching; Reactive-ion etching
    • H10P50/24—Dry etching; Plasma etching; Reactive-ion etching of semiconductor materials
    • H10P50/242—Dry etching; Plasma etching; Reactive-ion etching of semiconductor materials of Group IV materials
    • H—ELECTRICITY
    • H01—ELECTRIC ELEMENTS
    • H01J—ELECTRIC DISCHARGE TUBES OR DISCHARGE LAMPS
    • H01J37/00—Discharge tubes with provision for introducing objects or material to be exposed to the discharge, e.g. for the purpose of examination or processing thereof
    • H01J37/32—Gas-filled discharge tubes
    • H01J37/32917—Plasma diagnostics
    • H01J37/32935—Monitoring and controlling tubes by information coming from the object and/or discharge
    • H—ELECTRICITY
    • H01—ELECTRIC ELEMENTS
    • H01J—ELECTRIC DISCHARGE TUBES OR DISCHARGE LAMPS
    • H01J37/00—Discharge tubes with provision for introducing objects or material to be exposed to the discharge, e.g. for the purpose of examination or processing thereof
    • H01J37/32—Gas-filled discharge tubes
    • H01J37/32917—Plasma diagnostics
    • H01J37/3299—Feedback systems
    • H—ELECTRICITY
    • H05—ELECTRIC TECHNIQUES NOT OTHERWISE PROVIDED FOR
    • H05H—PLASMA TECHNIQUE; PRODUCTION OF ACCELERATED ELECTRICALLY-CHARGED PARTICLES OR OF NEUTRONS; PRODUCTION OR ACCELERATION OF NEUTRAL MOLECULAR OR ATOMIC BEAMS
    • H05H1/00—Generating plasma; Handling plasma
    • H05H1/24—Generating plasma
    • H05H1/46—Generating plasma using applied electromagnetic fields, e.g. high frequency or microwave energy
    • H—ELECTRICITY
    • H10—SEMICONDUCTOR DEVICES; ELECTRIC SOLID-STATE DEVICES NOT OTHERWISE PROVIDED FOR
    • H10P—GENERIC PROCESSES OR APPARATUS FOR THE MANUFACTURE OR TREATMENT OF DEVICES COVERED BY CLASS H10
    • H10P14/00—Formation of materials, e.g. in the shape of layers or pillars
    • H10P14/20—Formation of materials, e.g. in the shape of layers or pillars of semiconductor materials
    • H10P14/24—Formation of materials, e.g. in the shape of layers or pillars of semiconductor materials using chemical vapour deposition [CVD]
    • H—ELECTRICITY
    • H10—SEMICONDUCTOR DEVICES; ELECTRIC SOLID-STATE DEVICES NOT OTHERWISE PROVIDED FOR
    • H10P—GENERIC PROCESSES OR APPARATUS FOR THE MANUFACTURE OR TREATMENT OF DEVICES COVERED BY CLASS H10
    • H10P50/00—Etching of wafers, substrates or parts of devices
    • H10P50/20—Dry etching; Plasma etching; Reactive-ion etching
    • H10P50/24—Dry etching; Plasma etching; Reactive-ion etching of semiconductor materials
    • H10P50/246—Dry etching; Plasma etching; Reactive-ion etching of semiconductor materials of Group III-V materials
    • H—ELECTRICITY
    • H10—SEMICONDUCTOR DEVICES; ELECTRIC SOLID-STATE DEVICES NOT OTHERWISE PROVIDED FOR
    • H10P—GENERIC PROCESSES OR APPARATUS FOR THE MANUFACTURE OR TREATMENT OF DEVICES COVERED BY CLASS H10
    • H10P50/00—Etching of wafers, substrates or parts of devices
    • H10P50/20—Dry etching; Plasma etching; Reactive-ion etching
    • H10P50/28—Dry etching; Plasma etching; Reactive-ion etching of insulating materials
    • H10P50/282—Dry etching; Plasma etching; Reactive-ion etching of insulating materials of inorganic materials
    • H10P50/283—Dry etching; Plasma etching; Reactive-ion etching of insulating materials of inorganic materials by chemical means
    • H—ELECTRICITY
    • H10—SEMICONDUCTOR DEVICES; ELECTRIC SOLID-STATE DEVICES NOT OTHERWISE PROVIDED FOR
    • H10W—GENERIC PACKAGES, INTERCONNECTIONS, CONNECTORS OR OTHER CONSTRUCTIONAL DETAILS OF DEVICES COVERED BY CLASS H10
    • H10W20/00—Interconnections in chips, wafers or substrates
    • H10W20/01—Manufacture or treatment
    • H10W20/071—Manufacture or treatment of dielectric parts thereof
    • H10W20/093—Manufacture or treatment of dielectric parts thereof by modifying materials of the dielectric parts
    • H10W20/095—Manufacture or treatment of dielectric parts thereof by modifying materials of the dielectric parts by irradiating with electromagnetic or particle radiation
    • H—ELECTRICITY
    • H10—SEMICONDUCTOR DEVICES; ELECTRIC SOLID-STATE DEVICES NOT OTHERWISE PROVIDED FOR
    • H10W—GENERIC PACKAGES, INTERCONNECTIONS, CONNECTORS OR OTHER CONSTRUCTIONAL DETAILS OF DEVICES COVERED BY CLASS H10
    • H10W20/00—Interconnections in chips, wafers or substrates
    • H10W20/01—Manufacture or treatment
    • H10W20/071—Manufacture or treatment of dielectric parts thereof
    • H10W20/093—Manufacture or treatment of dielectric parts thereof by modifying materials of the dielectric parts
    • H10W20/096—Manufacture or treatment of dielectric parts thereof by modifying materials of the dielectric parts by contacting with gases, liquids or plasmas

Definitions

  • Endpoint - a point in time at which a process (e.g., etching of a silicon layer) has reached or is near completion.
  • a process e.g., etching of a silicon layer
  • Endpoint domain an interval in a data set during which an endpoint is thought to occur.
  • An endpoint domain is usually relatively broad and is based on user estimate.
  • PLS-DA Partial Least Squares Discriminant Analysis - a technique for finding relationships between two sets of data.
  • PLS-DA may be used when there are multiple independent variables (in an input matrix X) and possibly multiple dependent variables (in an input matrix Y).
  • the Y variables are not continuous but consist of a set of independent discrete values or classes.
  • PLS-DA may try to find linear combinations of the X variables that can be used to classify the input data into one of the discrete classes.
  • Stepwise regression - refers to fitting a straight line using a least-squares fitting algorithm to the data values in a finite temporal interval of the data from an individual sensor channel.
  • endpoint refers to a point in time at which a process (e.g., etching a silicon layer) has reached or is near completion.
  • the process of identifying an endpoint may be as simple as identifying a signal with the largest change. However, a signal change may not always coincide with an endpoint. Other factors, such as noise in the channel, may cause the signal pattern to change.
  • Fig. 1 shows a simple method for establishing an endpoint algorithm. The method as described in Fig. 1 is usually performed manually by an expert user, for example.
  • test substrate tends to be of the same type as the substrate that may be utilized in a production environment. For example, if a specific patterned substrate is utilized during production, a similar patterned substrate may be employed as a test substrate.
  • a first step 102 data is acquired for a substrate.
  • sensors such as a pressure manometer, an optical emission spectrometer (OES), a temperature sensor, and the like
  • OES optical emission spectrometer
  • a temperature sensor and the like
  • Data for hundreds, if not thousands, of sensor channels may be collected.
  • an expert user may examine one or more signals for changes in the signal patterns.
  • the expert user may employ one or more software programs to assist with the analysis.
  • the software program may be a simple analysis tool that may perform simple calculations and analysis.
  • the software program may be a simple data visualization program that may be employed to graphically illustrate a signal history, for example.
  • the expert user may be looking for a signal change (e.g., change in a signal pattern) as an indication of an endpoint. For example, if a signal is sloping downward, a peak in the signal slope may represent a change.
  • a signal change e.g., change in a signal pattern
  • a peak in the signal slope may represent a change.
  • the expert user may eliminate data values that he believes to be not relevant in identifying an endpoint.
  • One method for reducing the data set includes identifying and eliminating regions in the signal stream within which the expert user does not expect the endpoint to occur.
  • the expert user may limit his search for an endpoint to a target area in the signal stream, usually between a pre-endpoint domain and a post-endpoint domain. Because of the high cost (in expert time) of finding and refining endpoint signatures, the aim is to make the pre-endpoint and post-endpoint domains as large as possible to limit the region left in which to look for endpoint.
  • the expert user may further reduce the data set by only analyzing select signals.
  • the select signals may include signals or combination of signals that, based on the expert user's experience, may contain endpoint data.
  • the combination of signals is usually from a single sensor source.
  • data from different sensor sources are not combined since variations between the sensors may make the correlation analysis difficult, if not impossible, to be performed manually.
  • the expert user may perform a verification analysis to determine the robustness of the signal change as an endpoint candidate. For example, the expert user may analyze the history of the signal to determine the uniqueness of the signal change. If the signal change is not unique (i.e., occurring more than once in the history of the signal), the signal may be eliminated from the data set. The expert user may then resume his tedious task of identifying the "elusive" endpoint in another signal.
  • a set of filters (such as a set of digital filters) may be applied to the data set to remove noise and smooth out the data.
  • filters include, but are not limited to, for example, time series filters and frequency-based filters.
  • the multi-variate analysis may be performed to further reduce the data set.
  • the expert user may be required to define the shape (e.g., curve) of an endpoint feature.
  • the expert user is expected to anticipate the shape of the endpoint even though an endpoint candidate may have yet to be identified.
  • the multi-variate analysis essentially eliminates signals that do not exhibit the desired shape. In an example, if the shape of the endpoint is defined to be a peak, signals that do not exhibit this shape may be eliminated. Accordingly, if the optimal endpoint signature does not have the "expected" shape, the optimal endpoint signature may be missed.
  • the expert user may choose an endpoint algorithm type based on the nature of the transition.
  • the endpoint algorithm type may be based on the shape of the spectral line(s), for example, that may represent the endpoint.
  • the endpoint may be represented by a slope change. Accordingly, the expert user may propose a slope dependent algorithm.
  • the endpoint algorithm may be based on the derivative that may provide the best endpoint signature.
  • the first derivative such as a change in the slope
  • the second derivative of the slope such as an inflection point
  • the ability to identify not only an endpoint signature but also the best endpoint algorithm associated with the endpoint signature may require expertise that few users (even expert users) may possess.
  • the algorithm settings may be optimized and/or tested.
  • the endpoint algorithm Once the endpoint algorithm has been identified, the endpoint algorithm may be converted into a production endpoint algorithm. Since differences may exist between the test environment and the production environment, the setting of the endpoint algorithm may have to be adjusted before the endpoint algorithm may be moved into production. Settings that may be adjusted include but are not limited to, for example, smoothing filters, delay time, specific settings for the algorithm types, and the like.
  • filters that may be employed to smooth the data in a test environment may cause unacceptable real-time delay within a production environment.
  • real-time delay refers to the time difference between a non-filtered signal change and a filtered signal change. For example, a peak in a signal may have occurred at 40 seconds into the process. However, after a filter is applied, the peak may not occur until 5 seconds later. If an endpoint algorithm is applied with the filter settings, the substrate may be over-etched before the endpoint algorithm identifies the endpoint. To minimize the real-time delay, the filters may have to be adjusted.
  • a test may be performed to determine if the settings have been optimized.
  • the endpoint algorithm may be applied to the data set that has been utilized to create the endpoint algorithm. If the endpoint algorithm correctly identifies the endpoint using the adjusted settings, the settings may be considered as optimized. However, if the endpoint algorithm fails to correctly identify the endpoint, the settings may have to be adjusted. This test may have to be performed multiple times (through a trial and error method) before the settings may even be optimized.
  • step 112 is usually considered as an optional step in the creation of an endpoint algorithm.
  • the endpoint algorithm that may be created may not always be the optimal endpoint algorithm for the process.
  • Fig. 1 shows a simple method for establishing an endpoint algorithm.
  • FIG. 2 shows, in an embodiment of the invention, a simple flow chart illustrating a method for constructing an endpoint algorithm.
  • FIG. 3 A and 3B show, in an embodiment of the invention, a simple flow chart illustrating the steps an algorithm engine may execute in discovering the optimal endpoint algorithm.
  • FIG. 4 shows, in an embodiment of the invention, a simple flow chart for implementing the optimal endpoint algorithm within a production environment.
  • FIG. 5 shows, in an embodiment of the invention, a block diagram illustrating an example of an evolution of data sets into a list of optimal endpoint algorithms.
  • inventions are described hereinbelow, including methods and techniques. It should be kept in mind that the invention might also cover articles of manufacture that includes a computer readable medium on which computer-readable instructions for carrying out embodiments of the inventive technique are stored.
  • the computer readable medium may include, for example, semiconductor, magnetic, opto- magnetic, optical, or other forms of computer readable medium for storing computer readable code.
  • the invention may also cover apparatuses for practicing embodiments of the invention.
  • Such apparatus may include circuits, dedicated and/or programmable, to carry out tasks pertaining to embodiments of the invention. Examples of such apparatus include a general-purpose computer and/or a dedicated computing device when appropriately programmed and may include a combination of a computer/computing device and dedicated/programmable circuits adapted for the various tasks pertaining to embodiments of the invention.
  • Embodiments of the invention include methods for constructing an endpoint algorithm for determining an optimal endpoint for a process. Embodiments of the invention also include in-situ methods for applying the endpoint algorithm within a production environment.
  • methods are provided for constructing an endpoint algorithm.
  • the methods may include simple, user-friendly, automated methods that may be utilized by both expert and non-expert users.
  • the methods may include acquiring sensor data, automatically defining an approximate endpoint period, automatically analyzing the data, automatically determining a set of potential endpoint signatures, and automatically importing an optimal endpoint algorithm into production.
  • the algorithm engine may be a software program that is based on a function of time relative to a target region for an endpoint (e.g. , endpoint domain). Once the user has defined an approximate endpoint region (e.g., endpoint domain), the algorithm engine may be employed to analyze the data to discover a set of optimal endpoint signatures.
  • endpoint e.g. , endpoint domain
  • the algorithm engine may identify a set of potential shapes that may represent the potential endpoint signatures in a multi-variate analysis. Unlike the prior art, the user is not require to have prior knowledge of the shape for each potential endpoint signature (e.g., peak, valley, step, etc.). Instead, the algorithm engine may generate a list of potential shapes once the algorithm engine has identified the potential endpoint signatures. Thus, the potential endpoint algorithms that may be identified by the algorithm engine are not limited to a single shape (e.g., curve). In an embodiment, the algorithm engine is configured to perform data conditioning and testing of known endpoint candidates in order to identify the best endpoint signatures for a process.
  • the variability of each parameter as a function of time may be derived by performing a stepwise regression to determine the slope of each data input parameter in a series of finite time intervals throughout the history of the process.
  • the time intervals used in the slope calculation may be set to reject noise in the incoming data and also to reject slow drifts in the data that are not associated with the endpoint.
  • OES signals may be grouped according to the degree of change (i.e., slope) that is seen in the variability as the process evolves.
  • degree of change i.e., slope
  • contiguous wavelengths with similar slope variance may be grouped together.
  • slope- based grouping of the OES signals the number of signals that may need to be analyzed and the noise in those signals may be greatly reduced.
  • the result may represent a list of signals and group of signals that are most likely to contain information related to the endpoint.
  • culling may be performed to reduce the number of potential endpoint signatures.
  • a robust endpoint signature is one that is present in all processed substrates. In an example, if an endpoint signature is not a feature in all or a substantial majority of test substrates, then the endpoint signatures is not robust and may be eliminated. However, if an endpoint signature appears on a control substrate, the endpoint signature may also be eliminated since a control substrate is a substrate that has not been etched and therefore should not have produced an endpoint signature.
  • multi-variate analysis may be performed.
  • the results from the analysis may be utilized as input into a Partial Least Squares Discriminant Analysis (PLS-DA) in order to optimize weighting of each individual signal in each slope- based group.
  • PLS-DA Partial Least Squares Discriminant Analysis
  • the PLS-DA may rely on the target region for the endpoint and the shapes provided by the algorithm engine.
  • the results from the PLS-DA from OES signals may be banded and combined with other sensor signals.
  • the PLS-DA may be repeated with the new set of banded signals to produce a compact optimized combination of potential endpoint signatures that may have a high contrast and a low computational load for real-time endpoint calculation.
  • the potential endpoint signatures are converted into endpoint algorithms with minimal possible delay time. Potential endpoint signatures that can not be converted into real-time endpoint algorithms with minimal real time-delay may be eliminated. In other words, a real-time endpoint algorithm may be discarded if the real-time delay associated with the algorithm exceeds the maximum allowable real-time delay.
  • the potential endpoint algorithms may be ranked based on a ratio of useful information to the information that is irrelevant and/or on real-time delay, hereinafter referred to as a fidelity ratio.
  • a fidelity ratio a ratio of useful information to the information that is irrelevant and/or on real-time delay
  • an algorithm with a high fidelity ratio and a low real-time delay is considered a more robust algorithm.
  • one of the real-time endpoint algorithms may be selected and moved into production.
  • FIG. 2 shows, in an embodiment of the invention, a simple flow chart illustrating a method for constructing an endpoint algorithm.
  • a first step 202 data is acquired by a set of sensors within the processing chamber.
  • data is acquired by a set of sensors within the processing chamber.
  • data (such as optical emission, electrical signals, pressure data, plasma data, and the like) are being collected by a set of sensors.
  • the data that is to be utilized in creating the optimal endpoint algorithm may be coming from more than one test substrate.
  • the data may be coming from test substrates that may be processed within different chambers.
  • noise that is associated with the differences between chambers may also be eliminated.
  • the endpoint domain is an approximate and relatively broad time interval within which the algorithm engine is to search for valid endpoint signatures. For example, because of the high speed of the search the user can expand the endpoint domain to incorporate some of what, in the prior art, would have been the pre-endpoint domain. By so doing the algorithm engine can identify endpoint signatures that might occur earlier in the process. These early endpoints reduce the risk of the process damaging underlying
  • an algorithm engine is activated to perform data analysis and to generate a set of optimal endpoint algorithms.
  • data files from more than one substrate may be analyzed.
  • an endpoint algorithm constructed from data files from multiple substrates, although involving a larger quantity of data, tends to be more robust since endpoint features that are not commonly found in the substrates being analyzed may be eliminated.
  • FIGS. 3 A and 3B show, in an embodiment of the invention, a simple flow chart illustrating the steps an algorithm engine may execute in analyzing the data sets and generating a list of optimal endpoint algorithms. To facilitate discussion, Figs. 3 A and 3B will be discussed in conjunction with Fig. 5.
  • Fig. 5 shows a block diagram illustrating an example of an evolution of data sets into a list of optimal endpoint algorithms, in an embodiment.
  • the algorithm engine may perform linear fitting on the available data sets (initial data group 502).
  • each signal may be divided into uniform segments based on time intervals (data group 504).
  • the length of the segments is important. If the segment length is too long, the endpoint may be averaged out and the endpoint may be missed. If the segment is too short, the slope (as described later in step 304) may be affected by noise.
  • a minimal and maximum number for the segment length may be predefined. In an embodiment, the minimum segment length is longer than a 1/10 of a second. In another embodiment, the maximum segment length is shorter than 2 seconds for data collected at 10Hz.
  • the algorithm engine may calculate a slope and its
  • slope noise values may be employed to normalize the slopes (data group 506B).
  • the algorithm engine may perform a multi-variate analysis (such as a partial least square analysis) utilizing the slopes scaled by the slope noise values as inputs to generate an additional list of slopes and slope noise values based on signals from a combination of sensor channels (also included in data group 506A).
  • slope noise values may be employed to normalize the slopes (also included in data group 506B).
  • the algorithm engine may identify signal candidates that may be carrying endpoint data.
  • the algorithm engine may analyze each signal (and its segments) to quantify the amount of variation in the slope for each signal.
  • One method for quantifying the variability in a slope may include calculating the standard deviation of the normalized slope.
  • a high standard deviation may represent a signal with changes occurring in the slope.
  • a high standard deviation may represent a signal that may be carrying potential endpoint data.
  • signals with a high slope variance relative to slope noise
  • OES data may include a high volume of wavelength measurements (at least 2,000 signals)
  • the algorithm engine may reduce the number of OES signals by combining contiguous wavelengths with similar slope variance into signal wavelength bands (data group 510), at a next step 308.
  • the 100 wavelength measurements may be combined into a single signal wavelength band and may be treated as a single unit during the analysis. For example, if there are 2,000 wavelength measurements, then only 10 signal wavelength bands may have to be analyzed.
  • computational loads may be reduced since the numbers of items that have to be analyzed have been significantly reduced.
  • the algorithm engine may identify a list of normalizing signals (data group 506B) that may capture drift and noise in the underlying process. In other words, the algorithm engine may identify signals suitable for normalizing because they have a high slope but low variance (relative to the slope noise).
  • the normalizing signals (data group 512) may represent possible candidates for removing common mode changes (e.g., drift, noise, etc.) in the sensor signals.
  • the algorithm engine may reduce the number of normalized OES signals by combining contiguous wavelengths with similar slope variance into normalized signal wavelength bands (data group 514). Step 312 is somewhat analogous to step 308 except that step 312 is applied to the normalized OES signals.
  • the algorithm may generate a list of high contrast sensor signals (data group 508), high contrast sensor signal wavelength bands (data group 510), normalized signals (data group 512), and normalized wavelength bands (data group 514) for all sensor channels.
  • the signals within each data set are ranked. Since the possibility of endpoint data within each signal has been quantified, the signals within each data set may be ranked. In an example, a signal with a high slope variance may have a higher ranking than a signal with a low slope variance.
  • the algorithm engine may search the high contrast sensor signals and/or bands for possible endpoint signatures within the endpoint domain (data group 516).
  • an endpoint signature may be identified through a set of class features (peak, valley, inflection, etc.).
  • the set of class features may be predefined, in an embodiment.
  • the set of class features may be searched within the different derivative of the signals.
  • filters may be applied to the data groups 508 and 510 in order to remove noise and to smooth out the data.
  • the filters being applied to the data groups may be time symmetric filters.
  • Time symmetric filters utilize equal number of points before and after a particular point to calculate an average value. These filters can only be applied in a post-processing mode rather than during real-time execution of the process. Unlike time asymmetric filters, time symmetric filters tend to introduce minimal time distortion and/or amplitude distortion. As a result, the filtered data may experience minimal real-time delay.
  • each data group may include a plethora of signals. Since each data group has been ranked, in an embodiment, data analysis time may be significantly reduced by reducing the search values. In an example, instead of searching all of the items within data group 508, only the top 10 high contrast sensor signals may be analyzed. The number of items that may be searched may vary. A diminishing return analysis may be performed to determine the optimal number.
  • the algorithm engine may search the ratios of high contrast sensor signals/bands (data groups 508 and 510) to normalizing sensors/bands (data groups 512 and 514) for possible endpoint signatures (data group 518) within the endpoint domain.
  • the algorithm engine may search the ratios of high contrast sensor signals/bands (data groups 508 and 510) to normalizing sensors/bands (data groups 512 and 514) for possible endpoint signatures (data group 518) within the endpoint domain.
  • the algorithm engine may search the data results (data groups 516 and 518) to rank combinations (data group 520). In other words, matching is performed to combine endpoint signatures with similar shapes and time period in order to improve contrast and SNR. In an embodiment, linear combination is performed within the same derivative. In other words, a peak that occurs in the first derivate may not be combined with a peak that occurs in the second derivative even though both may be occurring within the same time interval.
  • the algorithm engine may perform a robustness test to remove potentially non-repeatable endpoint signatures.
  • the robustness test may check for consistency across multiple substrates. In an example, if the potential endpoint signature is not consistent across multiple substrates, the potential endpoint signature may be discarded since the potential endpoint signature may be a result of noise/drift, for example.
  • the robustness test may check for similarity between test substrates and a control substrate (or a set of control substrates).
  • the test substrates are substrates with resist mask with one portion being an exposed silicon area.
  • the control substrate may have the same characteristic as the test substrates except the control substrate may be totally covered by a resist mask. Both the test substrates and the control substrate may undergo the same substrate processing.
  • control substrate since the entire surface of the control substrate is covered with a resist mask, the control substrate should show no sign of etching. Accordingly, the control substrate should have no endpoint. Thus, if a change on the control substrate matches one of the potential endpoint signatures, the matched potential endpoint signature is discarded.
  • the robustness test may include testing for uniqueness.
  • the potential endpoint signature being tested has a peak feature. The rest of the signal may be analyzed to determine if another peak feature is occurring before or after the occurrence of the potential endpoint signature. If another peak is identified, the potential endpoint signature is eliminated.
  • the algorithm engine may perform a multi-variate correlation analysis, such as a correlation-based partial least square discriminate analysis (PLS-DA) to optimize the list of potential endpoint signatures.
  • a multi-variate analysis (such as a correlation-based PLS analysis) usually requires the shape of the endpoint signature to be defined. In other words, the multi-variate analysis needs to know the desired shape of the signature curve.
  • the user is usually the one who has to provide the shape (e.g., peak, valley, slope, etc.) of the endpoint signature. Given that a multi-variate correlation analysis (PLS-DA) to optimize the list of potential endpoint signatures.
  • PLS-DA partial least square discriminate analysis
  • determination of the shape of an endpoint candidate may take hours, if not weeks, the user may normally only be able to provide one shape feature as an input into the multi-variate analysis.
  • the potential endpoint signatures as identified by the algorithm engine may have different shape features.
  • the number of inputs that may be entered in the multi-variate correlation analysis may depend upon the shapes of the potential endpoint signatures that may have been identified.
  • the shape/shapes may be correlated against each signal to generate a correlation matrix between the potential endpoint signature and the signals within each sensor channel.
  • the correlation matrix may include optimal weights and/or loads that may be applied to every signal to maximize the contrast in each potential endpoint signature.
  • the multi- variate analysis may help optimize the list of potential endpoint signatures (data group 522), a multi-variate correlation analysis is not required to identify a list of optimal endpoint algorithms. Also, even though a correlation-based PLS analysis is utilized in the
  • this invention is not limited to a correlation-based PLS analysis but may be any type of correlation-based multi-variate analysis.
  • the algorithm engine may convert the remaining potential endpoint signatures (data group 522) into real-time endpoint algorithms (data group 524) with minimal real-time delay.
  • the algorithm engine is configured to convert the potential endpoint signatures into endpoint algorithms that may be executed during production with minimal real-time delay.
  • settings that may be required by each endpoint algorithm may be calculated automatically.
  • the settings for real-time filters may be automatically optimized to call an endpoint on every processed test substrate with minimal filter delay.
  • the real-time filters may be cascaded and may use initialization of the cascade memory components to minimize the initial transients that occur with infinite impulse response filters. This is particularly important in endpoint algorithms that may have an endpoint close to the start of the data history.
  • the algorithm engine may provide a realtime endpoint algorithm, hi an embodiment, if the algorithm engine is unable to construct a real-time endpoint algorithm then no endpoint algorithm is provided. In an example, if the algorithm engine is unable to construct a real-time endpoint algorithm that is able to call/identify an endpoint on every processed test substrate, then no endpoint algorithm may be provided.
  • the algorithm engine may eliminate endpoint algorithms that may exceed the maximum allowable real-time delay. In an example, if the time required to identify an endpoint may exceed a predefined threshold, the endpoint algorithm may be eliminated since the real-time delay may result in over-etched substrates during production.
  • the algorithm engine may eliminate real-time endpoint algorithms that do not pass a set of robustness criteria.
  • An example of a robustness criterion may include identifying endpoints on all test substrates with minimal real-time delay. In other words, each endpoint algorithm may be required to identify endpoint on all test substrates.
  • Another example of a robustness criterion may include not identifying endpoint on a control substrate. In other words, if an endpoint algorithm is able to identify an endpoint on a control substrate, the endpoint algorithm is not robust and the endpoint algorithm may be discarded.
  • the algorithm engine may rank the real-time endpoint algorithms.
  • the ranking may be based on the fidelity ratio and/or on realtime delay. In an example, if two real-time endpoint algorithms have the same fidelity ratio then the endpoint algorithm with the smaller real-time delay is ranked higher. In another example, if two endpoint algorithms have the same real-time delay then the endpoint algorithm with the higher fidelity ratio may have a higher rank.
  • a real-time endpoint algorithm may be moved into production.
  • the real-time endpoint algorithm that has the highest ranking may be automatically moved into production.
  • the real-time endpoint algorithm that may be moved into production may be user-controlled, thereby enabling the user to choose the endpoint algorithm that may best meet his needs.
  • real-time delay is a concern of a device manufacturer. For this reason, the device manufacturer may prefer to have a less robust endpoint algorithm if a shorter delay time may be provided.
  • Empirical evidence shows that by automating the process, the task of creating an optimal real-time endpoint algorithm may be performed in a few minutes. Further, since the algorithm engine is configured to perform the analysis with minimal input from a human, the process of constructing an endpoint algorithm may now be performed by non-expert users.
  • the user may quickly redefine the endpoint domain and rerun the algorithm engine to generate a new list of endpoint algorithms within a few minutes.
  • FIG. 4 shows, in an embodiment of the invention, a simple flow chart for implementing a real-time endpoint algorithm in a production environment.
  • a recipe may be executed.
  • data may be acquired during substrate processing by a set of sensors.
  • an endpoint algorithm may be employed in-situ to analyze the data to identify the process endpoint.
  • a computing engine may be employed to analyze the data. Since a high volume of data may be collected, the computing engine may be a fast processing module that may be configured to handle a large volume of data. The data may be sent directly from the sensor without first having to go through the fabrication host controller or even the process module controller.
  • the system may make a determination about the endpoint being identified.
  • one or more embodiments of the present invention provide for methods for identifying an optimal real-time endpoint algorithm.
  • automating the analysis the methods essentially eliminate the need for an expert user.
  • a more robust endpoint algorithm may be moved into production.
  • time required for creating an endpoint algorithm has been significantly reduced, a updating or creating new endpoint algorithm is no longer a resource- intensive and time-consuming task.

Landscapes

  • Physics & Mathematics (AREA)
  • Engineering & Computer Science (AREA)
  • Plasma & Fusion (AREA)
  • Chemical & Material Sciences (AREA)
  • Analytical Chemistry (AREA)
  • Electromagnetism (AREA)
  • Spectroscopy & Molecular Physics (AREA)
  • Drying Of Semiconductors (AREA)
  • Testing And Monitoring For Control Systems (AREA)
  • General Factory Administration (AREA)
  • Chemical Vapour Deposition (AREA)
  • Management, Administration, Business Operations System, And Electronic Commerce (AREA)
  • Container, Conveyance, Adherence, Positioning, Of Wafer (AREA)
  • Indication And Recording Devices For Special Purposes And Tariff Metering Devices (AREA)
  • Complex Calculations (AREA)
  • Combined Controls Of Internal Combustion Engines (AREA)
  • Arrangements For Transmission Of Measured Signals (AREA)
  • Plasma Technology (AREA)
  • Information Retrieval, Db Structures And Fs Structures Therefor (AREA)

Abstract

L'invention porte sur un procédé d'identification automatique d'un algorithme de points finaux optimaux pour qualification d'un point d'extrémité de processus lors d'un traitement de substrat dans un système de traitement au plasma. Le procédé comprend la réception de données de détection à partir d'une pluralité de détecteurs lors du traitement d'au moins un substrat dans le système de traitement au plasma, les données de détecteur comprenant une pluralité de flux de signaux provenant d'une pluralité de canaux de détecteurs. Le procédé comprend également l'identification d'un domaine de points finaux, le domaine de points d'extrémité étant une période approximative pendant laquelle on s'attend à ce que le point final de processus se produise. Le procédé comprend en outre l'analyse de données du détecteur afin de générer un ensemble de signatures pour la fin potentielle. Le procédé comprend également la conversion de l'ensemble de signatures de points finaux potentiels en un ensemble d'algorithme de points finaux optimaux. Le procédé comprend en outre l'importation d'un algorithme de points finaux optimaux de l'ensemble d'algorithme de points finaux optimaux dans un environnement de production.
PCT/US2010/040477 2009-06-30 2010-06-29 Procédé de construction d'un algorithme de points finaux optimaux Ceased WO2011002810A2 (fr)

Priority Applications (4)

Application Number Priority Date Filing Date Title
JP2012518588A JP5693573B2 (ja) 2009-06-30 2010-06-29 最適なエンドポイント・アルゴリズムを構築する方法
CN201080027296.6A CN102804353B (zh) 2009-06-30 2010-06-29 构建最优终点算法的方法
KR1020117031561A KR101741271B1 (ko) 2009-06-30 2010-06-29 최적 종말점 알고리즘 구성 방법
SG2011085149A SG176566A1 (en) 2009-06-30 2010-06-29 Methods for constructing an optimal endpoint algorithm

Applications Claiming Priority (6)

Application Number Priority Date Filing Date Title
US22210209P 2009-06-30 2009-06-30
US22202409P 2009-06-30 2009-06-30
US61/222,102 2009-06-30
US61/222,024 2009-06-30
US12/555,674 US8983631B2 (en) 2009-06-30 2009-09-08 Arrangement for identifying uncontrolled events at the process module level and methods thereof
US12/555,674 2009-09-08

Publications (3)

Publication Number Publication Date
WO2011002810A2 true WO2011002810A2 (fr) 2011-01-06
WO2011002810A3 WO2011002810A3 (fr) 2011-04-14
WO2011002810A4 WO2011002810A4 (fr) 2011-06-03

Family

ID=43411705

Family Applications (5)

Application Number Title Priority Date Filing Date
PCT/US2010/040478 Ceased WO2011002811A2 (fr) 2009-06-30 2010-06-29 Agencement pour identification d'évènements non commandés au niveau du module de traitement et procédés liés
PCT/US2010/040456 Ceased WO2011002800A2 (fr) 2009-06-30 2010-06-29 Procédé et agencement pour surveillance et commande in-situ de processus pour des outils de traitement au plasma
PCT/US2010/040468 Ceased WO2011002804A2 (fr) 2009-06-30 2010-06-29 Procédé et appareil de prédiction de l'uniformité de la vitesse de gravure pour la qualification d'une chambre de plasma
PCT/US2010/040477 Ceased WO2011002810A2 (fr) 2009-06-30 2010-06-29 Procédé de construction d'un algorithme de points finaux optimaux
PCT/US2010/040465 Ceased WO2011002803A2 (fr) 2009-06-30 2010-06-29 Procédé et appareil pour maintenance préventive et prédictive de chambres de traitement

Family Applications Before (3)

Application Number Title Priority Date Filing Date
PCT/US2010/040478 Ceased WO2011002811A2 (fr) 2009-06-30 2010-06-29 Agencement pour identification d'évènements non commandés au niveau du module de traitement et procédés liés
PCT/US2010/040456 Ceased WO2011002800A2 (fr) 2009-06-30 2010-06-29 Procédé et agencement pour surveillance et commande in-situ de processus pour des outils de traitement au plasma
PCT/US2010/040468 Ceased WO2011002804A2 (fr) 2009-06-30 2010-06-29 Procédé et appareil de prédiction de l'uniformité de la vitesse de gravure pour la qualification d'une chambre de plasma

Family Applications After (1)

Application Number Title Priority Date Filing Date
PCT/US2010/040465 Ceased WO2011002803A2 (fr) 2009-06-30 2010-06-29 Procédé et appareil pour maintenance préventive et prédictive de chambres de traitement

Country Status (6)

Country Link
JP (5) JP5629770B2 (fr)
KR (5) KR101741274B1 (fr)
CN (5) CN102473590B (fr)
SG (5) SG176564A1 (fr)
TW (5) TWI484435B (fr)
WO (5) WO2011002811A2 (fr)

Families Citing this family (20)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN102332383B (zh) * 2011-09-23 2014-12-10 中微半导体设备(上海)有限公司 等离子体刻蚀工艺的终点监控方法
US10128090B2 (en) 2012-02-22 2018-11-13 Lam Research Corporation RF impedance model based fault detection
US9502221B2 (en) * 2013-07-26 2016-11-22 Lam Research Corporation Etch rate modeling and use thereof with multiple parameters for in-chamber and chamber-to-chamber matching
CN104730372B (zh) * 2013-12-13 2018-08-10 朗姆研究公司 基于rf阻抗模型的故障检测
US10192763B2 (en) * 2015-10-05 2019-01-29 Applied Materials, Inc. Methodology for chamber performance matching for semiconductor equipment
US10269545B2 (en) * 2016-08-03 2019-04-23 Lam Research Corporation Methods for monitoring plasma processing systems for advanced process and tool control
US9972478B2 (en) * 2016-09-16 2018-05-15 Lam Research Corporation Method and process of implementing machine learning in complex multivariate wafer processing equipment
US11067515B2 (en) * 2017-11-28 2021-07-20 Taiwan Semiconductor Manufacturing Co., Ltd. Apparatus and method for inspecting a wafer process chamber
CN108847381A (zh) * 2018-05-25 2018-11-20 深圳市华星光电半导体显示技术有限公司 测试基板及延长测试基板使用寿命的方法
US10651097B2 (en) * 2018-08-30 2020-05-12 Lam Research Corporation Using identifiers to map edge ring part numbers onto slot numbers
US20200266037A1 (en) * 2019-02-14 2020-08-20 Advanced Energy Industries, Inc. Maintenance for remote plasma sources
DE102019209110A1 (de) * 2019-06-24 2020-12-24 Sms Group Gmbh Industrielle Anlage, insbesondere Anlage der metallerzeugenden Industrie oder der Aluminium- oder Stahlindustrie und Verfahren zum Betreiben einer industriellen Anlage, insbesondere einer Anlage der metallerzeugenden Industrie oder der Aluminium- oder Stahlindustrie
GB202010471D0 (en) * 2020-07-08 2020-08-19 Univ Exeter Control of processing equipment
KR102882466B1 (ko) * 2020-10-15 2025-11-05 에이에스엠 아이피 홀딩 비.브이. 예측 유지보수 방법 및 예측 유지보수 장치
JP7289992B1 (ja) * 2021-07-13 2023-06-12 株式会社日立ハイテク 診断装置及び診断方法並びにプラズマ処理装置及び半導体装置製造システム
US12147212B2 (en) * 2021-12-21 2024-11-19 Applied Materials, Inc. Diagnostic methods for substrate manufacturing chambers using physics-based models
WO2023148967A1 (fr) * 2022-02-07 2023-08-10 株式会社日立ハイテク Dispositif de diagnostic, procédé de diagnostic, système de dispositif de fabrication à semi-conducteur et système de fabrication de dispositif à semi-conducteur
US12191126B2 (en) 2022-02-15 2025-01-07 Applied Materials, Inc. Process control knob estimation
US20240021450A1 (en) * 2022-07-15 2024-01-18 Verity Instruments, Inc. Control for semiconductor processing systems
CN119013763A (zh) * 2023-03-17 2024-11-22 株式会社日立高新技术 状态预测设备、状态预测方法和状态预测系统

Family Cites Families (59)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US5272872A (en) * 1992-11-25 1993-12-28 Ford Motor Company Method and apparatus of on-board catalytic converter efficiency monitoring
JP3301238B2 (ja) * 1994-10-25 2002-07-15 三菱電機株式会社 エッチング方法
JPH08148474A (ja) * 1994-11-16 1996-06-07 Sony Corp ドライエッチングの終点検出方法および装置
JPH09306894A (ja) * 1996-05-17 1997-11-28 Sony Corp 最適発光スペクトル自動検出システム
US6197116B1 (en) * 1996-08-29 2001-03-06 Fujitsu Limited Plasma processing system
JP3630931B2 (ja) * 1996-08-29 2005-03-23 富士通株式会社 プラズマ処理装置、プロセスモニタ方法及び半導体装置の製造方法
US5993615A (en) * 1997-06-19 1999-11-30 International Business Machines Corporation Method and apparatus for detecting arcs
WO1999014394A1 (fr) * 1997-09-17 1999-03-25 Tokyo Electron Limited Dispositif et procede de detection et de prevention de formation d'arc dans un systeme a plasma haute frequence
US5986747A (en) 1998-09-24 1999-11-16 Applied Materials, Inc. Apparatus and method for endpoint detection in non-ionizing gaseous reactor environments
US8617351B2 (en) * 2002-07-09 2013-12-31 Applied Materials, Inc. Plasma reactor with minimal D.C. coils for cusp, solenoid and mirror fields for plasma uniformity and device damage reduction
JP2001338856A (ja) * 2000-05-30 2001-12-07 Tokyo Seimitsu Co Ltd 半導体製造システムのプロセスコントローラ
JP4554037B2 (ja) * 2000-07-04 2010-09-29 東京エレクトロン株式会社 消耗品の消耗度予測方法及び堆積膜厚の予測方法
US6567718B1 (en) * 2000-07-28 2003-05-20 Advanced Micro Devices, Inc. Method and apparatus for monitoring consumable performance
US6391787B1 (en) * 2000-10-13 2002-05-21 Lam Research Corporation Stepped upper electrode for plasma processing uniformity
US6821794B2 (en) 2001-10-04 2004-11-23 Novellus Systems, Inc. Flexible snapshot in endpoint detection
JP2003151955A (ja) * 2001-11-19 2003-05-23 Nec Kansai Ltd プラズマエッチング方法
WO2003102724A2 (fr) * 2002-05-29 2003-12-11 Tokyo Electron Limited Procede et systeme de traitement, de stockage et de manipulation de donnees
US6825050B2 (en) * 2002-06-07 2004-11-30 Lam Research Corporation Integrated stepwise statistical process control in a plasma processing system
US20040031052A1 (en) * 2002-08-12 2004-02-12 Liberate Technologies Information platform
US6781383B2 (en) * 2002-09-24 2004-08-24 Scientific System Research Limited Method for fault detection in a plasma process
CN100407215C (zh) * 2002-09-30 2008-07-30 东京毅力科创株式会社 用于监视和控制半导体生产过程的方法和装置
BRPI0314881B1 (pt) * 2002-10-25 2019-01-08 S&C Electric Company sistema e método para controle de distribuição de energia elétrica através de uma rede
JP4365109B2 (ja) * 2003-01-29 2009-11-18 株式会社日立ハイテクノロジーズ プラズマ処理装置
US6969619B1 (en) * 2003-02-18 2005-11-29 Novellus Systems, Inc. Full spectrum endpoint detection
JP2004295348A (ja) * 2003-03-26 2004-10-21 Mori Seiki Co Ltd 工作機械の保守管理システム
US20060006139A1 (en) * 2003-05-09 2006-01-12 David Johnson Selection of wavelengths for end point in a time division multiplexed process
US7101805B2 (en) * 2003-05-09 2006-09-05 Unaxis Usa Inc. Envelope follower end point detection in time division multiplexed processes
JP2004335841A (ja) * 2003-05-09 2004-11-25 Tokyo Electron Ltd プラズマ処理装置の予測装置及び予測方法
US7062411B2 (en) * 2003-06-11 2006-06-13 Scientific Systems Research Limited Method for process control of semiconductor manufacturing equipment
JP4043408B2 (ja) * 2003-06-16 2008-02-06 東京エレクトロン株式会社 基板処理装置及び基板処理方法
US6902646B2 (en) * 2003-08-14 2005-06-07 Advanced Energy Industries, Inc. Sensor array for measuring plasma characteristics in plasma processing environments
KR100567745B1 (ko) * 2003-09-25 2006-04-05 동부아남반도체 주식회사 스퍼터링용 타겟의 수명예측 장치 및 수명예측방법
US8036869B2 (en) * 2003-09-30 2011-10-11 Tokyo Electron Limited System and method for using first-principles simulation to control a semiconductor manufacturing process via a simulation result or a derived empirical model
US7930053B2 (en) * 2003-12-23 2011-04-19 Beacons Pharmaceuticals Pte Ltd Virtual platform to facilitate automated production
US7233878B2 (en) * 2004-01-30 2007-06-19 Tokyo Electron Limited Method and system for monitoring component consumption
US7146237B2 (en) * 2004-04-07 2006-12-05 Mks Instruments, Inc. Controller and method to mediate data collection from smart sensors for fab applications
JP2006004992A (ja) * 2004-06-15 2006-01-05 Seiko Epson Corp 研磨装置管理システム、管理装置、管理装置制御プログラム及び管理装置制御方法
TWI336823B (en) * 2004-07-10 2011-02-01 Onwafer Technologies Inc Methods of and apparatuses for maintenance, diagnosis, and optimization of processes
US7292045B2 (en) * 2004-09-04 2007-11-06 Applied Materials, Inc. Detection and suppression of electrical arcing
JP4972277B2 (ja) * 2004-11-10 2012-07-11 東京エレクトロン株式会社 基板処理装置の復帰方法、該装置の復帰プログラム、及び基板処理装置
US7828929B2 (en) * 2004-12-30 2010-11-09 Research Electro-Optics, Inc. Methods and devices for monitoring and controlling thin film processing
JP4707421B2 (ja) * 2005-03-14 2011-06-22 東京エレクトロン株式会社 処理装置，処理装置の消耗部品管理方法，処理システム，処理システムの消耗部品管理方法
JP2006328510A (ja) * 2005-05-30 2006-12-07 Ulvac Japan Ltd プラズマ処理方法及び装置
TWI338321B (en) * 2005-06-16 2011-03-01 Unaxis Usa Inc Process change detection through the use of evolutionary algorithms
US7409260B2 (en) * 2005-08-22 2008-08-05 Applied Materials, Inc. Substrate thickness measuring during polishing
US7302363B2 (en) * 2006-03-31 2007-11-27 Tokyo Electron Limited Monitoring a system during low-pressure processes
US7413672B1 (en) * 2006-04-04 2008-08-19 Lam Research Corporation Controlling plasma processing using parameters derived through the use of a planar ion flux probing arrangement
US7829468B2 (en) * 2006-06-07 2010-11-09 Lam Research Corporation Method and apparatus to detect fault conditions of plasma processing reactor
KR20080006750A (ko) * 2006-07-13 2008-01-17 삼성전자주식회사 반도체소자 제조용 플라즈마 도핑 시스템
US20080063810A1 (en) * 2006-08-23 2008-03-13 Applied Materials, Inc. In-situ process state monitoring of chamber
CN100587902C (zh) * 2006-09-15 2010-02-03 北京北方微电子基地设备工艺研究中心有限责任公司 在线预测刻蚀设备维护的方法
JP2008158769A (ja) * 2006-12-22 2008-07-10 Tokyo Electron Ltd 基板処理システム、制御装置、設定情報監視方法および設定情報監視プログラムを記憶した記憶媒体
US7548830B2 (en) * 2007-02-23 2009-06-16 General Electric Company System and method for equipment remaining life estimation
US7674636B2 (en) * 2007-03-12 2010-03-09 Tokyo Electron Limited Dynamic temperature backside gas control for improved within-substrate process uniformity
US8055203B2 (en) * 2007-03-14 2011-11-08 Mks Instruments, Inc. Multipoint voltage and current probe system
JP2008311338A (ja) * 2007-06-13 2008-12-25 Harada Sangyo Kk 真空処理装置及びこれに用いる異常放電予知装置、並びに、真空処理装置の制御方法
KR100892248B1 (ko) * 2007-07-24 2009-04-09 주식회사 디엠에스 플라즈마 반응기의 실시간 제어를 실현하는 종말점 검출장치 및 이를 포함하는 플라즈마 반응기 및 그 종말점 검출방법
US20090106290A1 (en) * 2007-10-17 2009-04-23 Rivard James P Method of analyzing manufacturing process data
JP4983575B2 (ja) * 2007-11-30 2012-07-25 パナソニック株式会社 プラズマ処理装置およびプラズマ処理方法

Also Published As

Publication number Publication date
WO2011002800A3 (fr) 2011-04-07
KR101708077B1 (ko) 2017-02-17
CN102474968B (zh) 2015-09-02
WO2011002803A3 (fr) 2011-03-03
TW201112302A (en) 2011-04-01
WO2011002804A2 (fr) 2011-01-06
JP2012532462A (ja) 2012-12-13
KR20120037420A (ko) 2012-04-19
JP5624618B2 (ja) 2014-11-12
WO2011002811A2 (fr) 2011-01-06
WO2011002811A3 (fr) 2011-02-24
KR20120101293A (ko) 2012-09-13
CN102804353A (zh) 2012-11-28
JP2012532463A (ja) 2012-12-13
JP2012532464A (ja) 2012-12-13
TWI495970B (zh) 2015-08-11
KR101741271B1 (ko) 2017-05-29
WO2011002810A4 (fr) 2011-06-03
KR20120037421A (ko) 2012-04-19
SG176147A1 (en) 2011-12-29
CN102473631A (zh) 2012-05-23
KR20120037419A (ko) 2012-04-19
CN102473590B (zh) 2014-11-26
CN102804353B (zh) 2015-04-15
TWI509375B (zh) 2015-11-21
TW201129936A (en) 2011-09-01
TWI536193B (zh) 2016-06-01
CN102804929A (zh) 2012-11-28
JP5629770B2 (ja) 2014-11-26
KR101708078B1 (ko) 2017-02-17
TWI484435B (zh) 2015-05-11
CN102474968A (zh) 2012-05-23
JP5599882B2 (ja) 2014-10-01
TW201108022A (en) 2011-03-01
JP2012532460A (ja) 2012-12-13
WO2011002803A2 (fr) 2011-01-06
WO2011002800A2 (fr) 2011-01-06
CN102473590A (zh) 2012-05-23
JP2012532461A (ja) 2012-12-13
SG176565A1 (en) 2012-01-30
TWI480917B (zh) 2015-04-11
WO2011002810A3 (fr) 2011-04-14
KR20120047871A (ko) 2012-05-14
SG176564A1 (en) 2012-01-30
CN102804929B (zh) 2015-11-25
JP5693573B2 (ja) 2015-04-01
KR101741272B1 (ko) 2017-05-29
SG176566A1 (en) 2012-01-30
TW201129884A (en) 2011-09-01
SG176567A1 (en) 2012-01-30
CN102473631B (zh) 2014-11-26
KR101741274B1 (ko) 2017-05-29
WO2011002804A3 (fr) 2011-03-03
TW201115288A (en) 2011-05-01

Similar Documents

Publication Publication Date Title
WO2011002810A2 (fr) Procédé de construction d'un algorithme de points finaux optimaux
US8538572B2 (en) Methods for constructing an optimal endpoint algorithm
CN112884193A (zh) 预测装置、预测方法及记录介质
US11456194B2 (en) Determining critical parameters using a high-dimensional variable selection model
JP2015503763A5 (fr)
US9696717B2 (en) Apparatus and method of segmenting sensor data output from a semiconductor manufacturing facility
CN111258866B (zh) 计算机性能预测方法、装置、设备及可读存储介质
KR102425936B1 (ko) 장치 진단 장치, 플라스마 처리 장치 및 장치 진단 방법
CN120346903B (zh) 一种高纯度铁粉的多级磁选设备及其磁选方法
JP2008077665A (ja) 工程管理方法およびその方法を利用したシステム
Muzellec et al. Gradient strikes back: How filtering out high frequencies improves explanations
CN116359201B (zh) 一种药品鉴定方法、系统、存储介质和电子设备
CN114324187A (zh) 监测刻蚀腔体颗粒的方法、装置、服务器及可读存储介质
CN118824380A (zh) 荧光信号的处理方法、装置、存储介质及处理器
KR101340227B1 (ko) 헬리콥터 비행시험 데이터 처리 방법 및 컴퓨터 판독 가능한 기록매체
JP7129433B2 (ja) シークエンシング分析の関連において正規化を実施するための方法および装置
CN114385294B (zh) 一种交互式陷印方法及系统
CN119577344B (zh) 一种基于ai的实验设备数据处理方法
CN120994484B (zh) 测试用例选择方法、电子设备、存储介质及程序产品
CN119985369B (zh) 数据处理方法、装置、测试设备及存储介质
Bergmann Data Mining for Manufacturing: Preventive Maintenance, Failure prediction, Quality Control
CN118760976A (zh) 一种基于变点检测的隧道掘进机掘进参数处理方法
JP2022149047A5 (fr)
CN118427104A (zh) 一种基于分布式发压机集群的压力测试方法和系统
CN120652786A (zh) 用于智能键合银丝的控制逻辑分析方法及系统

Legal Events

Date Code Title Description
WWE Wipo information: entry into national phase

Ref document number: 201080027296.6

Country of ref document: CN

121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 10794664

Country of ref document: EP

Kind code of ref document: A2

WWE Wipo information: entry into national phase

Ref document number: 2012518588

Country of ref document: JP

ENP Entry into the national phase

Ref document number: 20117031561

Country of ref document: KR

Kind code of ref document: A

NENP Non-entry into the national phase

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 10794664

Country of ref document: EP

Kind code of ref document: A2