Methods, systems and computer readable media for detecting processing chamber conditions

Abstract

A method and system for detecting a condition of a processing chamber and a computer readable medium. The present disclosure provides multiple embodiments of a system and method for detecting a condition of a processing chamber. Embodiments include performing a waferless processing step in a processing chamber to determine the condition of a plurality of chamber walls. Based on an analysis of residual gases generated from the waferless processing step, an operator or process controller can determine whether the chamber walls have deteriorated to the point where they need to be cleaned.

Description

Technical Field

[0001] The present disclosure relates to systems and methods for residual gas analysis in semiconductor manufacturing. Background Art

[0002] Semiconductor manufacturing utilizes various processing steps, including etching and depositing multiple films (i.e., multiple layers) on a blanketed wafer to produce trenches, vias, metal lines, passive and active circuit elements such as capacitors, resistors, inductors, transistors, antennas, and generally form insulating and conductive multiple structures in the production of wafers with semiconductor integrated circuits.

[0003] One processing step includes plasma etching, which is performed by applying electromagnetic energy (usually radio frequency (RF)) to a gas containing a chemically reactive element (such as fluorine or chlorine). The plasma releases positively charged ions that bombard the wafer to remove (etch) material, and releases chemically reactive radicals that react with the etched material to form volatile or non-volatile byproducts. The charge of the ions directs the ions vertically toward the wafer. This produces an almost vertical etch profile, which is critical for tiny features in today's densely packed chip designs. The process chemistry varies depending on the type of film to be etched. Those process chemistries used in dielectric etching applications are typically fluorine-based. Silicon and metal etching use chlorine-based chemistries. In reactive ion etching, the goal is to optimize the balance between physical etching and chemical etching so that the physical bombardment (etching rate) is sufficient to remove the necessary material while the appropriate chemical reactions occur to form consumed volatile byproducts.

[0004] Other processing steps include non-plasma etching, chemical vapor deposition (CVD), and physical vapor deposition (PVD). These steps, as well as the plasma etching step described above, are performed using various precursor gases (i.e., reactive gases and / or reactive gas mixtures, and in some cases, reactive gases and their respective inert carrier gases) provided to a processing chamber holding the wafer. Summary of the invention

[0005] Some embodiments of the present disclosure provide a method for detecting a condition of a processing chamber, comprising: performing a waferless processing step in a processing chamber, wherein the waferless processing step includes introducing one or more precursor gases into the processing chamber, wherein the one or more precursor gases react with each other and with a plurality of residual compounds accumulated on a plurality of walls of the processing chamber, and the reaction produces a residual gas; analyzing a plurality of ion species in the residual gas; and determining a condition of the processing chamber based on the analysis.

[0006] Some embodiments of the present disclosure provide a system for detecting a condition of a processing chamber, comprising: an ion analyzer module and a process controller. The ion analyzer module is configured to be coupled to a processing chamber, and the ion analyzer module is further configured to: receive residual gas from the processing chamber, the residual gas being a result of a waferless processing step performed in the processing chamber, wherein the waferless step includes introducing one or more precursor gases into the processing chamber, the one or more gases reacting with each other and with residual compounds accumulated on multiple walls of the processing chamber, the reaction resulting in residual gas; and analyze multiple ion species in the residual gas. The process controller is configured to determine the condition of the processing chamber based on the analysis.

[0007] Some embodiments of the present disclosure provide a computer-readable medium comprising a plurality of instructions executable by a processor to detect a processing chamber condition by performing the following steps: performing a waferless processing step in the processing chamber, wherein the waferless processing step includes introducing one or more precursor gases into the processing chamber, wherein the one or more precursor gases react with each other and with residual compounds accumulated on multiple walls of the processing chamber, wherein the reactions result in residual gas; analyzing multiple ion species in the residual gas; and determining the condition of the processing chamber based on the analysis. BRIEF DESCRIPTION OF THE DRAWINGS

[0008] The following detailed description and the attached Figure 1 The various aspects of the present disclosure will be best understood by reading together. It is worth noting that, according to the standard practice of the industry, the various features are not drawn to scale. In fact, for the sake of clarity of discussion, the size of the various features may be arbitrarily increased or reduced.

[0009] Figure 1 A system for detecting a processing chamber condition is illustrated according to one embodiment of the present disclosure;

[0010] Figure 2 is a flow chart of a method for detecting a processing chamber condition according to one embodiment of the present disclosure;

[0011] Figure 3 According to one embodiment of the present disclosure, Figure 1 controlling the processing steps performed on the wafer in the processing chamber; and

[0012] Figure 4 According to one embodiment of the present disclosure, Figure 1 The cleaning process performed on the control wafer in the processing chamber.

[0013]

Explanation of symbols

[0014] 100: System

[0015] 102: Wafer

[0016] 104: Ion analyzer module

[0017] 106: Process Controller

[0018] 107: Electronic devices

[0019] 108: Processing chamber

[0020] 110: Catheter

[0021] 112:Port

[0022] 114: Communication cable

[0023] 116: Wireless Communication

[0024] 118: Processing gas module

[0025] 120: Communication connection

[0026] 122: Gas duct

[0027] 124: Material

[0028] 125: Clean gas module

[0029] 126: Cleaning gas duct

[0030] 128: wall

[0031] 130: Communication connection

[0032] 200: Method flow chart

[0033] 202: Steps

[0034] 204: Steps

[0035] 206: Steps

[0036] 300: Processing steps

[0037] 302: Segment

[0038] 304: Groove

[0039] 306:Silicon dioxide film

[0040] 308: Photoresist layer

[0041] 310: wall

[0042] 400: Cleaning process DETAILED DESCRIPTION

[0043] The disclosure that follows provides many different implementations or embodiments to implement different features of the subject matter provided. Specific embodiments of components and configurations are described below to simplify the disclosure. Of course, these are merely embodiments and are not intended to be limiting. For example, in the subsequent description, forming a first feature above or on a second feature may include an implementation in which the first and second features are formed in direct contact, and may also include an implementation in which additional features may be formed between the first and second features, so that the first and second features may not be in direct contact. In addition, the disclosure may repeat numbers and / or letters in various embodiments. Such repetition is for the sake of simplicity and clarity, and the repetition itself does not imply a relationship between the various embodiments and / or configurations discussed.

[0044] Additionally, spatially relative terms, such as "below," "below," "lower," "above," "higher," and the like, may be used herein to facilitate description of the relationship of one element or feature to another element or feature as depicted in the drawings. In addition to encompassing the orientation depicted in the drawings, the spatially relative terms are intended to encompass different orientations of the device in use or operation. The device may be otherwise oriented (rotated 90 degrees or at other orientations), and the spatially relative terms used herein interpreted accordingly.

[0045] Wafers are often processed in multiple batches, referred to as lots. Depending on the process step, the wafers may be processed in a process chamber. The process chamber may be temperature and pressure controlled, and may include ports for introducing precursor (i.e., reactive) gases and removing byproducts of the process step (e.g., any residual gases). The process chamber may also include electronics for generating a plasma of the gases introduced into the chamber as part of the process step. Process chambers and associated components are known in the art and will not be described in further detail.

[0046] The walls of a processing chamber are exposed to various byproducts that result from the reaction of precursor gases with the wafers or with other materials in the chamber, such as materials used in a plasma sputtering process, which degrade (i.e., decay) over time. In the context of the present disclosure, degradation of a chamber, and particularly degradation of the walls of a chamber, refers to the accumulation of byproducts on the walls of the chamber, thereby reducing the efficiency of the corresponding processing steps that produced such byproducts. Since, for example, it may be necessary to increase the amount of precursor gas and / or increase the processing time in order to overcome the reduced efficiency at these manufacturing steps, it is advantageous to know when the walls of the chamber need to be cleaned of such contaminants so that future batches of wafers will not reduce yield and / or will not increase the cost of future manufacturing steps.

[0047] Figure 1 A system 100 for detecting processing chamber conditions is illustrated in accordance with one embodiment of the present disclosure. The system 100 includes an ion analyzer module 104 and a process controller 106. In one embodiment, one or more wafers 102 may be located in a processing chamber 108. The one or more wafers 102 may represent a batch of wafers. In other embodiments of the disclosed system 100, the wafers 102 are not located in the processing chamber 108. The processing chamber 108 may be used in conjunction with a plurality of semiconductor manufacturing process steps known to those skilled in the art. For example, the processing chamber may be used in the manufacture of integrated circuits by performing a plurality of series of processing steps on a plurality of batches of semiconductor wafers (also referred to as semiconductor substrates), such as deposition of various layers or films by chemical vapor deposition and / or physical vapor deposition, plasma-based or non-plasma-based etching, ion implantation, and annealing. The wafers 102 may be formed of any type of substrate material, including semiconductor substrates formed of silicon or gallium arsenide.

[0048] In one embodiment, the processing chamber 108 processes one or more batches of wafers, such as the batch of wafers 102. The batch of wafers may undergo multiple manufacturing process steps in the processing chamber 108. After processing one or more batches of wafers in the processing chamber 108, an operator may initiate a process to test the condition of the processing chamber 108, as discussed further below.

[0049] In one embodiment, the ion analyzer module 104 is coupled to the processing chamber 108, for example, via a conduit 110 and a port 112 of the processing chamber 108. The ion analyzer module 104 is also communicatively coupled to the process controller 106, either via a physical connection, such as a communication cable 114, or via wireless communication 116. In an alternative embodiment (not shown), the ion analyzer module 104 is located in the processing chamber 108, and the ion analyzer module 104 communicates with the process controller 106 in the same manner as previously described (i.e., wireless communication or via a physical connection through the port 112 of the processing chamber 108). In another embodiment (not shown), the process controller 106 is included as a component of the ion analyzer module 104. In this embodiment, the ion analyzer module 104 includes electronic devices, such as a processor, memory, hard disk, input / output (I / O), etc., which perform the functions of the process controller 106, as discussed further below.

[0050] In one embodiment, the ion analyzer module 104 is a gas chromatograph configured to analyze multiple ion species in a gas. The gas may include one or more ion species, such as fluorine ions, chloride ions, and / or bromide ions. For example, the gas chromatograph may be configured to measure the concentration of multiple ion species in the gas. In another embodiment, the ion analyzer module 104 includes a gas chromatograph and a mass spectrometer for determining multiple ion species and measuring their respective concentrations. Gas chromatographs and mass spectrometers are known in the art and will not be discussed in further detail.

[0051] In one embodiment, the ion analyzer module 104 receives residual gas from the processing chamber 108 during processing of a batch of wafers or during one or more wafer-less processing steps. Wafer-less processing steps include any semiconductor processing steps, such as a plurality of production processing steps used in processing of a plurality of batches of wafers performed in the processing chamber 108 without any wafers in the processing chamber 108. The residual gas may contain one or more ion species, such as fluoride ions, chloride ions, and / or bromide ions, but the scope of the present disclosure includes a plurality of ion analyzer modules configured to analyze a plurality of ionic species (generated by any semiconductor manufacturing process steps) in the residual gas.

[0052] The process controller 106 is configured to receive (or retrieve) the results of the gas analysis performed by the ion analyzer module 104. The process controller 106 may include electronic devices 107, such as a microcontroller, memory, hard disk, and / or input / output (i.e., a keypad and display), etc. In one embodiment, the display (electronic device 107) of the process controller 106 displays the results of the ion analyzer module 104 to an operator of the system 100. The results may indicate the concentrations of various ion species measured by the ion analyzer module 104, and therefore indicate the condition of the processing chamber 108 (i.e., the degree to which the processing chamber condition has deteriorated). For example, based on multiple concentrations of one or more (predetermined) ion species being greater than one or more (predetermined) respective baseline values, these results may indicate that the processing chamber 108 undergoes a cleaning process to remove accumulated byproducts (which have accumulated through previous multiple processing steps in the processing of one or more batches of wafers). According to one embodiment of the present disclosure, and as described below in combination Figure 4 As further discussed, the cleaning process cleans (ie, removes and / or replaces) byproducts from the interior walls of the chamber in preparation for receiving future batches of wafers for future processing.

[0053] In one embodiment, the process controller 106 is optionally coupled to a process gas module 118 via a communication connection 120. The process gas module 118 is coupled to the processing chamber 108 via one or more gas conduits 122. The process gas module 118 provides one or more precursor gases (i.e., gases that may react with each other, with the batches of wafers being processed, and / or with other material target species (e.g., sputtering materials) placed in the processing chamber 108 (e.g., material species 124)) and any inert carrier gases used to assist in the delivery of the precursor gases to the processing chamber 108. These reactive precursor gases may be used to etch the wafer 102 and deposit multiple layers of different material compositions on the wafer 102. In one embodiment, the process controller 106 controls the flow of gases (e.g., precursors and / or inert gases) from the process gas module 118 to the processing chamber 108. The process controller 106 may control the flow of gases by controlling the release, ratio, flow rate, and / or shutoff of gases from the process gas module 118 .

[0054] In another embodiment, the process controller 106 is optionally coupled to a clean gas module 125. As discussed in further detail below, the clean gas module 125 provides one or more clean gases to the processing chamber 108 via one or more clean gas conduits 126 to clean the processing chamber 108, for example, a plurality of chamber walls 128. In one embodiment, the process controller 106 controls the flow (i.e., release, ratio, flow rate, and / or shutoff) of the clean gas from the clean gas module 125 to the processing chamber 108 via a communication connection 130.

[0055] In one embodiment, the cleaning gas module 125 includes or has access to a plurality of gases for cleaning the process chamber 108, such as CFx, C4F8, CO, CHxFy, SF6, C5F8, O2, N2, HeO2, C2F6, Ar. These gases may also include carbon tetrafluoride, chlorine, and / or gases such as CH4, NF3, and HBr. The process controller 106 is configured to initiate a cleaning process for the process chamber by controlling the flow of one or more cleaning gases from the cleaning gas module 125 to the process chamber 108. In another embodiment, the process controller 106 is configured to control waferless process steps performed in the process chamber 108. For example, the process controller 106 may select one or more process steps (also referred to as waferless process steps) to be performed in the process chamber 108. This selection may be based on a history of multiple manufacturing process steps performed in the process chamber 108 on multiple batches of wafers. The history may be stored in the memory (electronic device 107). In one embodiment, the selection is based on a history of multiple manufacturing process steps performed on multiple batches of wafers in the processing chamber 108 since a previous clean of the processing chamber 108. In another embodiment, an operator selects one or more waferless process steps to be performed in the processing chamber 108 via the process controller 106, or in another embodiment via input controls (not shown) on the clean gas module 125.

[0056] In another embodiment, the process gas module 118 includes or has access to, for example, manufacturing process gases such as nitrogen trifluoride, chlorine, hydrogen bromide, silicon hexafluoride, and carbon tetrafluoride, alone or in any combination, although the scope of the present disclosure includes all gases and combinations of gases used in various manufacturing process steps. In one embodiment, the process gas module 118 and the clean gas module 125 include (or have access to) separate gas storage tanks, wherein each storage tank contains a separate manufacturing process gas.

[0057] Figure 2 A flow chart 200 of a method for monitoring a processing chamber condition according to an embodiment of the present disclosure is provided.

[0058] In step 202, one or more semiconductor processing steps are performed in the processing chamber 108. For example, in one embodiment of the present disclosure, the process controller 106 may select one or more processing steps (also referred to as waferless processing steps) and / or control the flow of one or more gases (e.g., precursor gases and / or carrier gases used in the selected processing steps) from the process gas module 118 to the processing chamber 108. In one embodiment, the waferless processing steps may be based on a history of multiple processing steps for one or more batches (i.e., multiple lots) of wafers. The history may include the number, type, and characteristics of multiple previous processing steps performed on more wafers or more batches of wafers, such as multiple precursor gases used and multiple operating parameters of the processing chamber 108. In another embodiment, based on the results of the ion analyzer module 104, communicated to an operator via the display (electronic device 107), the operator may input multiple instructions into the input / output interface (electronic device 107) of the process controller 106, or alternatively, directly input instructions into the input / output interface (not shown) of the processing gas module 118 to select one or more processing steps (i.e., also referred to as waferless processing steps) and / or control the release of one or more precursor gases from the processing gas module 118 to the processing chamber 108.

[0059] One or more waferless processing steps include introducing one or more precursor gases into the processing chamber 108. The one or more gases may react with each other, with the walls 128 of the processing chamber 108, and / or with target species (e.g., sputtered material or wafers 102) in the processing chamber. Chemical reactions of the one or more gases with compounds accumulated on the walls 128 of the processing chamber 108 may result in residual gases. These compounds may accumulate as byproducts of multiple manufacturing process steps (which are performed on one or more batches of wafers 102 in the processing chamber 108). In one embodiment, the concentration of one or more ion species in the residual gas is indicative of the condition of the walls 128 of the processing chamber 108.

[0060] Figure 3 A processing step 300 is depicted being performed on a wafer 102 in a processing chamber 108 in accordance with one embodiment of the present disclosure. The processing step 300 may represent, for example, an etching process, a chemical vapor deposition process, or a physical vapor deposition process performed on the wafer 102 within the processing chamber 108. For ease of illustration, only a small section 302 of the wall 128 of the processing chamber 108 is shown, however, the wall 128 of the processing chamber 108 completely encloses the wafer 102. Depending on the processing step, the processing chamber 108 may be controlled at a predetermined temperature and / or pressure, and the processing chamber 108 may include electronics, such as electrodes (not shown) for generating a plasma.

[0061] In process step 300, one or more precursor gases are introduced into process chamber 108. For example, if the process step is a plasma-based dry etch process step, the precursor gases may include, but are not limited to, nitrogen trifluoride, hydrogen bromide and chlorine, sulfur hexafluoride, chlorine, chlorine and sulfur hexafluoride, or carbon tetrafluoride. The principles discussed below are applicable to other process steps, such as chemical vapor deposition and physical vapor deposition for depositing multiple layers on a wafer, and non-plasma-based dry etching which uses, but is not limited to, such precursor gases, such as xenon difluoride, bromine trifluoride, chlorine trifluoride.

[0062] For purposes of illustration only, process step 300 is a plasma-based dry etching process step using nitrogen trifluoride as a precursor process gas. As depicted, nitrogen trifluoride is provided to process chamber 108 to etch trenches 304 in silicon dioxide film 306 of wafer 102. In this embodiment, wafer 102 has been patterned with photoresist layer 308. In process chamber 108, fluorine plasma is generated using RF power applied to electrodes (not shown) to drive a chemical reaction between the nitrogen trifluoride precursor gas and those portions of wafer 102 not covered by photoresist layer 308. A byproduct of the etching process is silicon tetrafluoride on walls 310 of trench 304, which is a volatile product. However, the plasma in the processing chamber 108, represented by electrons e-, can cause the silicon tetrafluoride to dissociate into free radicals, and the products of this dissociation in the presence of oxygen O2 can cause the accumulation of a layer of silicon oxyfluoride (Si-OF) compound on the wall 310 of the etched trench 304, and the accumulation of a layer of silicon oxyfluoride, silicon oxide, and / or silicon dioxide on the section 302 of the multiple walls 128 of the processing chamber 108 or on the multiple walls 128 of the processing chamber 108.

[0063] During a manufacturing (i.e., processing) cycle, a plurality of processing steps to provide a functional wafer from a batch of wafers (i.e., to manufacture functional wafers) are performed on a batch of wafers (e.g., the plurality of wafers 102) in the processing chamber 108. As the wafers are processed through one or more manufacturing cycles in the processing chamber 108, the thickness of the layer of, for example, silicon oxyfluoride compounds and / or other compounds (e.g., silicon oxide) on the walls 128 of the processing chamber 108 increases. When waferless processing steps (e.g., waferless plasma-based dry etching process step 300) are periodically performed in the processing chamber 108, without a batch of wafers (or alternatively, with a batch of wafers) in the processing chamber 108, when the thickness of the layer of silicon oxyfluoride compound on the walls 128 of the processing chamber 108 is relatively large, the concentration of fluorine ions in the residual gas in the chamber (which is a result of chemical reactions between the precursor gases and the compound accumulated on the walls 128) will be larger because there is more fluorine available for the plasma-driven oxidation / dissociation process and / or for the plasma-driven bond-breaking process (in which energetic plasma charged species break the bonds of fluorine atoms to the compound accumulated on the walls 128). Therefore, the measured value of the concentration of fluorine ions in the residual gas is directly related to the thickness of the residual compound accumulated on the walls 128 of the processing chamber 108. That is, a higher concentration of fluorine ions in the residual gas indicates a thicker layer of residual compounds accumulated on the walls 128 of the processing chamber 108 .

[0064] See again Figure 2 In step 204, the residual gas is analyzed for a plurality of ion species. For example, in one embodiment, the ion analyzer module 104 determines one or more ion species present in the residual gas, including their respective concentrations.

[0065] In step 206, a condition of the processing chamber 108 is determined based on the analysis of step 204. In one embodiment, the analysis is performed by the ion analyzer module 104, while in another embodiment, the analysis is performed by the process controller 106 based on data received from the ion analyzer module 104. In yet another embodiment, if the analysis indicates that the concentration of one or more ion species is greater than one or more respective (predetermined) baseline values, then the process controller 106 determines that the processing chamber 108 requires cleaning to remove multiple layers of process step byproducts (i.e., residual compounds) accumulated on the plurality of walls 128 of the processing chamber 108. In another embodiment, the indication of the presence of the predetermined ion species in the residual gas indicates that the processing chamber 108 is to be cleaned to remove multiple layers of process step byproducts accumulated on the plurality of walls 128 of the processing chamber 108.

[0066] In one embodiment, upon determining that the process chamber 108 needs to be cleaned, the process controller 106 automatically controls the flow of cleaning gas from the cleaning gas module 125 to the process chamber 108 to clean the plurality of chamber walls 128. In another embodiment, the process controller 106 includes a display (or an external computer or other processing device, including a display, or an external display, communicatively coupled to the system 100) for displaying the results of the analysis to an operator. In this embodiment, the operator decides whether to clean the process chamber 108 based on these results, and the process controller 106 includes an input / output interface (electronic device 107) for the operator to input instructions to the process controller 106 to control the flow of cleaning gas for cleaning the process chamber 108. In another embodiment, if the processing step of step 202 is not a waferless processing step, but is performed when a batch of wafers 102 are in the processing chamber 108, the process controller is configured to automatically control the flow of precursor gas from the processing gas module 118 to the processing chamber 108, for example, by shutting off this flow and controlling the flow of cleaning gas from the cleaning gas module 125 to the processing chamber 108 to clean multiple chamber walls 128.

[0067] Figure 4 A cleaning process 400 is depicted as being performed in a processing chamber 108 in accordance with one embodiment of the present disclosure. In one embodiment of the cleaning process 400, one or more cleaning gases are introduced into the processing chamber 108. For purposes of providing illustrative embodiments only, the following discussion depicts a carbon tetrafluoride-based cleaning process in which carbon tetrafluoride and chlorine are introduced into the chamber. However, the scope of the present disclosure encompasses other cleaning gases or compositions of cleaning gases for removing and / or replacing byproducts of manufacturing process steps, such as byproducts from etching, chemical vapor deposition, and physical vapor deposition, from the walls 128 of the processing chamber 108. Although the wafer 102 is also shown within the processing chamber 108 for ease of illustration, in one embodiment of the present disclosure, the cleaning process 400 is performed in the processing chamber 108 without any wafer present in the processing chamber 108.

[0068] When RF power is applied to an electrode (not shown) in the processing chamber 108, a fluorine plasma is generated, thereby driving a chemical reaction between the cleaning gas in the presence of oxygen and compounds such as silicon oxide and / or silicon dioxide (i.e., silicon oxide compounds) accumulated on the section 302 of the walls 128 of the processing chamber 108 (or on the walls 128). The chemical reaction causes the silicon oxide compounds (also referred to as silicon oxide-based passivation films) on the walls 128 of the processing chamber 108 to be replaced by a carbon-rich layer (CF), resulting in the generation of volatile silicon oxyfluoride oxides (Si-OF), which can be removed from the processing chamber 108.

[0069] In some embodiments, the system 100 can employ artificial intelligence or machine learning techniques that can estimate (e.g., anticipate) whether the concentration of one or more ion species in the residual gas is greater than one or more respective baseline values ​​before or during a next manufacturing process step, based on, for example, a residual gas analysis history, a manufacturing process history, a process chamber cleaning history, and / or possible future manufacturing process steps. The possible future manufacturing process steps may be based on manufacturing process steps pre-programmed by an operator of the system 100 and stored in memory and / or based on the manufacturing process history.

[0070] In some embodiments, one or more artificial intelligence or machine learning techniques may be implemented at least in part via machine learning circuitry. In one embodiment, the electronics of the process controller 106 may include machine learning circuitry that implements one or more artificial intelligence or machine learning techniques.

[0071] As used herein, "artificial intelligence" is used broadly to describe any computational intelligence systems and methods that can learn knowledge based on data histories and / or training data and, for example, use such learned knowledge to adjust its approach to solving one or more problems, for example, by inferring the conditions of the processing chamber 108 based on received inputs, such as: gas analysis (e.g., data corresponding to measured residual gases) received in real time from the ion analyzer module 104, and / or gas analysis histories, manufacturing process histories, processing chamber cleaning histories, and / or possible future multiple manufacturing process steps retrieved from the memory of the process controller 106.

[0072] Machine learning generally refers to a subfield or category of artificial intelligence, and is used herein to broadly describe any algorithm, mathematical model, statistical model, or the like implemented in one or more computer systems or circuits (e.g., circuits of process controller 106) and builds one or more models based on sample data (or training data) to make predictions or decisions.

[0073] The machine learning circuitry of the process controller 106 may employ, for example, neural networks, deep learning, convolutional neural networks, Bayesian program learning, support vector machines, computer vision, and pattern recognition techniques to solve problems such as predicting or determining the status of the processing chamber 108. Further, the machine learning circuitry may implement any one or combination of the following computational algorithms or techniques: classification, regression, supervised learning, unsupervised learning, feature learning, clustering, decision trees, image recognition, or the like.

[0074] As an example, the machine learning circuit may utilize an artificial neural network to develop, train, or update one or more machine learning models that may be used to detect or determine the condition of the processing chamber 108. An example artificial neural network may include a plurality of interconnected "neurons" that exchange information with each other. These connections have digital weights that can be adjusted based on experience, and thus the neural network can adapt to multiple inputs and have learning capabilities. The "neurons" may be included in a plurality of separate layers that are connected to each other, such as an input layer, a hidden layer, and an output layer. The neural network may be trained by providing training data (e.g., past data indicating the condition of the processing chamber 108) to the input layer. Through training, the neural network may generate and / or modify a hidden layer that represents weighted connections that map the training data provided at the input layer to known output information (e.g., gas analysis data) at the output layer. The relationship between the plurality of neurons of the input layer, the hidden layer, and the output layer formed through the training process may include a plurality of weighted connection relationships, such as may be stored as one or more machine learning models that are in or accessible to the machine learning circuit.

[0075] The machine learning circuitry may be implemented in one or more processors having access to a plurality of instructions, which may be stored in any computer-readable storage medium, which may be executed by the machine learning circuitry to perform any of the operations or functions described herein.

[0076] The present disclosure provides a system and method for monitoring processing chamber conditions without compromising the production of multiple batches of wafers manufactured in the processing chamber. Periodically, as determined by an operator of the system, or for example by a reduction in the yield of multiple batches of wafers processed in the processing chamber, or by using an ion analyzer module to collect and analyze residual gases produced during the processing of multiple batches of wafers, waferless processing steps are performed in the processing chamber 108 to determine the condition of the chamber walls 128. Based on the analysis of the residual gases produced by the waferless processing steps, the operator or process controller can determine whether the chamber walls 128 have deteriorated to the point where they should be cleaned. By using the ion analyzer module and the above-described embodiments of the present disclosure to determine when the chamber walls should be cleaned, the stability of the yield of future batches of wafers and the cost of future chamber processing steps may be maintained.

[0077] In one embodiment, a method for detecting a condition of a processing chamber includes performing a waferless processing step in a processing chamber. The waferless processing step includes introducing one or more precursor gases into the processing chamber. The one or more precursor gases react with each other and with residual compounds accumulated on multiple walls of the processing chamber to produce residual gas. The method also includes analyzing multiple ion species in the residual gas and determining a condition of the processing chamber based on the analysis.

[0078] In another embodiment, a system for detecting a condition of a processing chamber includes an ion analyzer module configured to be coupled to the processing chamber. The ion analyzer module is also configured to receive residual gas from the processing chamber, which is a result of a waferless processing step performed in the processing chamber. The waferless processing step includes introducing one or more precursor gases into the processing chamber. The one or more precursor gases react with each other and with residual compounds accumulated on multiple walls of the processing chamber, thereby resulting in residual gas. The ion analyzer module also analyzes multiple ion species in the residual gas. The system also includes a process controller configured to determine the condition of the processing chamber based on this analysis.

[0079] In yet another embodiment, a computer readable medium includes a plurality of instructions executable by a processor to detect a processing chamber condition by performing a step of performing a waferless processing step in a processing chamber. The waferless processing step includes introducing one or more precursor gases into the processing chamber. The one or more precursor gases react with each other and with residual compounds accumulated on multiple walls of the processing chamber to produce a residual gas. The computer readable medium also includes further instructions executable by the processor for detecting the processing chamber condition by performing a step of analyzing multiple ion species in the residual gas, and determining a condition of the processing chamber based on the analysis.

[0080] Several embodiments are summarized above so that those skilled in the art can better understand the various aspects of the present disclosure. Those skilled in the art will appreciate that they may easily use the present disclosure as a basis for the design and modification of other processes and structures to achieve the same purpose as the embodiments introduced herein, or to achieve the same advantages.

[0081] For example, in one embodiment of the present disclosure, a computer readable medium includes a plurality of instructions executable by a processor for determining a process chamber condition by executing a plurality of steps of controlling a semiconductor processing step in a process chamber, wherein the process step results in the generation of a residual gas in the chamber, analyzing a plurality of ion species in the residual gas, and determining the process chamber condition based on the analysis. The semiconductor processing step includes introducing one or more gases into the process chamber. The one or more gases react with each other and with residual compounds accumulated on the walls of the process chamber to generate the residual gas.

[0082] Some embodiments of the present disclosure provide a method for detecting a condition of a processing chamber, comprising: performing a waferless processing step in the processing chamber, wherein the waferless processing step includes introducing one or more precursor gases into the processing chamber, wherein the one or more precursor gases react with each other and with a plurality of residual compounds accumulated on a plurality of walls of the processing chamber, and the reaction produces a residual gas; analyzing a plurality of ion species in the residual gas; and determining a condition of the processing chamber based on the analysis.

[0083] In some embodiments, in the method for detecting a condition of a processing chamber, the waferless processing step includes an etching processing step, a chemical vapor deposition processing step, or a physical vapor deposition processing step.

[0084] In some embodiments, in the method for detecting a process chamber condition, one or more precursor gases further react with a target substance in the process chamber, wherein the target substance is a solid material, and wherein the waferless processing step is a plasma sputtering processing step.

[0085] In some embodiments, in a method for detecting a process chamber condition, the plurality of ion species includes one or more of a fluoride ion, a chloride ion, and a bromide ion.

[0086] In some embodiments, in a method for detecting a processing chamber condition, the processing chamber condition is based on whether the residual gas has a concentration of one or more of the plurality of ion species greater than one or more respective baseline values.

[0087] In some embodiments, the method for monitoring a processing chamber condition further comprises cleaning a plurality of walls of the processing chamber when the concentration of the one or more ionic species is greater than the one or more respective baseline values.

[0088] In some embodiments, in a method for detecting a condition of a processing chamber, cleaning a plurality of walls of the processing chamber includes introducing carbon tetrafluoride and chlorine gas into the processing chamber.

[0089] In some embodiments, in a method for detecting a processing chamber condition, the concentration of the one or more ion species is greater than the one or more baseline values, representing an excess accumulation of one or more residual compounds on multiple walls of the processing chamber, and wherein carbon tetrafluoride and multiple chlorine gases react with the one or more residual compounds, resulting in one or more of the one or more residual compounds being replaced by a carbon-rich compound.

[0090] In some embodiments, the method for detecting the condition of a processing chamber further includes performing multiple processing steps on one or more batches of wafers in the processing chamber, wherein performing the multiple processing steps precedes performing waferless processing steps, analyzing multiple ion species in the residual gas, and determining the condition of the processing chamber; and wherein performing waferless processing steps, analyzing multiple ion species in the residual gas, and determining the condition of the processing chamber are performed periodically.

[0091] In some embodiments, in a method for detecting a process chamber condition, the one or more precursor gases include one or more of nitrogen trifluoride, hydrogen bromide and chlorine, sulfur hexafluoride, chlorine, chlorine and sulfur hexafluoride, and carbon tetrafluoride.

[0092] Some embodiments of the present disclosure provide a system for detecting a condition of a processing chamber, comprising: an ion analyzer module and a process controller. The ion analyzer module is configured to be coupled to the processing chamber, and the ion analyzer module is further configured to: receive residual gas from the processing chamber, the residual gas being a result of a waferless processing step performed in the processing chamber, wherein the waferless step includes introducing one or more precursor gases into the processing chamber, the one or more gases reacting with each other and with residual compounds accumulated on multiple walls of the processing chamber, the reaction resulting in residual gas; and analyze multiple ion species in the residual gas. The process controller is configured to determine the condition of the processing chamber based on the analysis.

[0093] In some embodiments, in the system for detecting a condition of a processing chamber, the waferless processing step includes an etching process step, a chemical vapor deposition process step, or a physical vapor deposition process step.

[0094] In some embodiments, in a system for detecting a processing chamber condition, the plurality of ion species includes one or more of a fluoride ion, a chloride ion, and a bromide ion.

[0095] In some embodiments, in a system for detecting a condition of a processing chamber, the condition of the processing chamber is based on whether the residual gas has a concentration of one or more of the plurality of ion species greater than one or more respective baseline values.

[0096] In some embodiments, in the system for monitoring a processing chamber condition, the process controller is further configured to initiate cleaning of a plurality of chamber walls when the concentration of the one or more ion species is greater than the one or more respective baseline values.

[0097] In some embodiments, in a system for detecting a condition of a processing chamber, the processing chamber is configured to initiate cleaning of a plurality of chamber walls by controlling flow of a plurality of cleaning gases into the processing chamber.

[0098] In some embodiments, in a system for monitoring processing chamber conditions, the one or more precursor gases include one or more of nitrogen trifluoride, hydrogen bromide and chlorine, sulfur hexafluoride, chlorine, chlorine and sulfur hexafluoride, and carbon tetrafluoride.

[0099] In some embodiments, in the system for detecting a condition of a processing chamber, the process controller is further configured to control waferless processing steps performed in the processing chamber.

[0100] In some embodiments, a system for detecting a process chamber condition further includes a process chamber.

[0101] Some embodiments of the present disclosure provide a computer-readable medium comprising a plurality of instructions executable by a processor to detect a processing chamber condition by performing the following steps: performing a waferless processing step in the processing chamber, wherein the waferless processing step includes introducing one or more precursor gases into the processing chamber, wherein the one or more precursor gases react with each other and with residual compounds accumulated on multiple walls of the processing chamber, wherein the reactions result in residual gas; analyzing multiple ion species in the residual gas; and determining the condition of the processing chamber based on the analysis.

[0102] Those skilled in the art will also understand that these equivalent constructions do not depart from the spirit and scope of the present disclosure, and that they may make various changes, substitutions, and alterations without departing from the spirit and scope of the present disclosure.

Claims

1. A method for detecting a condition of a processing chamber, characterized in that: Include: performing a waferless processing step in a processing chamber, wherein the waferless processing step includes introducing one or more precursor gases into the processing chamber, the one or more precursor gases reacting with each other and with a plurality of residual compounds accumulated on a plurality of walls of the processing chamber, the reaction between the one or more precursor gases and the plurality of residual compounds accumulated on a plurality of walls of the processing chamber producing a residual gas, wherein the residual compounds include silicon oxyfluoride, and the residual gas includes fluoride ions resulting from a chemical reaction between the silicon oxyfluoride and the precursor gas; analyzing a plurality of ion species in the residual gas during the waferless processing step, the ion species including the fluorine ion; determining a condition of the walls of the processing chamber based on the analyzing the ion species in the residual gas, wherein the determining the condition of the walls of the processing chamber comprises: using artificial intelligence or machine learning techniques, the condition of the processing chamber is based on whether the residual gas has a concentration of the fluorine ions that is greater than a baseline value; and When the concentration of the fluoride ions is greater than the baseline value, cleaning the walls of the processing chamber, wherein cleaning the walls of the processing chamber includes introducing carbon tetrafluoride and chlorine gas into the processing chamber, and wherein the carbon tetrafluoride and the chlorine gas react with the one or more residual compounds to cause one or more of the one or more residual compounds to be replaced by a carbon-rich compound.

2. The method according to claim 1, characterized in that The waferless processing step includes an etching process step, a chemical vapor deposition process step, or a physical vapor deposition process step.

3. The method according to claim 1, characterized in that The one or more precursor gases further react with a target substance in the processing chamber, wherein the target substance is a solid material, and wherein the waferless processing step is a plasma sputtering processing step.

4. The method according to claim 1, characterized in that The ion types further include one or more of a chloride ion and a bromide ion.

5. The method according to claim 1, characterized in that: It also includes: performing multiple processing steps on one or more batches of wafers in the processing chamber, wherein the multiple processing steps are performed before performing the waferless processing step, analyzing the ion types in the residual gas, and determining the condition of the processing chamber; and wherein performing the waferless processing step, analyzing the ion types in the residual gas, and determining the condition of the processing chamber are performed periodically.

6. The method according to claim 1, characterized in that The one or more precursor gases include one or more of nitrogen trifluoride, hydrogen bromide and chlorine, sulfur hexafluoride, chlorine, chlorine and sulfur hexafluoride, and carbon tetrafluoride.

7. A system for detecting the condition of a processing chamber, characterized in that: Include: An ion analyzer module configured to be coupled to a processing chamber, the ion analyzer module further configured to: receiving a residual gas from the processing chamber, the residual gas being a result of a waferless processing step performed in the processing chamber, wherein the waferless processing step includes introducing one or more precursor gases into the processing chamber, the one or more gases reacting with each other and with residual compounds accumulated on walls of the processing chamber, the residual gas resulting from the reaction between the one or more precursor gases and the plurality of residual compounds accumulated on walls of the processing chamber, wherein the residual compounds include silicon oxyfluoride, the residual gas including fluoride ions resulting from a chemical reaction between the silicon oxyfluoride and the precursor gases; and analyzing a plurality of ion species in the residual gas during the waferless processing step, the ion species including the fluorine ion; and A process controller configured as follows: A condition of the walls of the processing chamber is determined based on the analyzing the ion species in the residual gas, wherein the process controller includes machine learning circuitry implementing one or more artificial intelligence or machine learning techniques, the condition of the processing chamber is based on whether the residual gas has a concentration of the fluoride ions that is greater than a baseline value, and the process controller is further configured to initiate a cleaning of the walls of the processing chamber when the concentration of the fluoride ions is greater than the baseline value, wherein initiating the cleaning of the walls of the processing chamber includes introducing carbon tetrafluoride and chlorine gas into the processing chamber, and wherein the carbon tetrafluoride and the chlorine gas react with the residual compounds to cause one or more of the residual compounds to be replaced by a carbon-rich compound.

8. The system according to claim 7, characterized in that The waferless processing step includes an etching process step, a chemical vapor deposition process step, or a physical vapor deposition process step.

9. The system according to claim 7, characterized in that The plurality of ion species further comprises one or more of a chloride ion and a bromide ion.

10. The system according to claim 7, characterized in that Wherein the processing chamber is configured to initiate the cleaning of the chamber walls by controlling the flow of the cleaning gas into the processing chamber.

11. The system according to claim 7, characterized in that The one or more precursor gases include one or more of nitrogen trifluoride, hydrogen bromide and chlorine, sulfur hexafluoride, chlorine, chlorine and sulfur hexafluoride, and carbon tetrafluoride.

12. The system according to claim 7, characterized in that The process controller is further configured to control the waferless processing steps performed in the processing chamber.

13. The system according to claim 7, characterized in that Also included is the processing chamber.

14. A computer readable medium, characterized in that The invention comprises a plurality of instructions executable by a processor to detect the processing chamber status and operate the processing chamber by executing the following steps: performing a waferless processing step in a processing chamber, wherein the waferless processing step includes introducing one or more precursor gases into the processing chamber, the one or more precursor gases reacting with each other and with residual compounds accumulated on walls of the processing chamber, the reaction between the one or more precursor gases and the plurality of residual compounds accumulated on walls of the processing chamber resulting in a residual gas, wherein the residual compound comprises silicon oxyfluoride, the residual gas comprising fluoride ions resulting from a chemical reaction between the silicon oxyfluoride and the precursor gas; analyzing a plurality of ion species in the residual gas during the waferless processing step, the ion species including the fluorine ion; determining a condition of the walls of the processing chamber based on the analyzing, wherein the condition of the walls of the processing chamber is based on whether the residual gas has a concentration of the fluoride ions that is greater than a baseline value; and initiating a cleaning of the walls of the processing chamber when the concentration of the fluoride ions is greater than the baseline value, wherein initiating the cleaning of the walls comprises introducing carbon tetrafluoride and chlorine gas into the processing chamber, and wherein the carbon tetrafluoride and the chlorine gas react with the residual compounds to cause one or more of the residual compounds to be replaced by a carbon-rich compound; Wherein said analyzing said ion species in said residual gas and determining said condition of said walls of said processing chamber based on said analysis are performed by said processor and performed by a machine learning circuit implementing one or more artificial intelligence or machine learning techniques.

Images