Feedforward control of multilayer stacks during device manufacturing

Abstract

A method for forming a multi-layer stack on a substrate includes processing the substrate in a first process chamber using a first deposition process to deposit a first layer of the multi-layer stack on the substrate, removing the substrate from the first process chamber, measuring a first thickness of the first layer using an optical sensor, determining a target second thickness for a second layer of the multi-layer stack based on the first thickness of the first layer, determining one or more process parameter values ​​for a second deposition process that will achieve the second target thickness for the second layer, and processing the substrate in the second process chamber using the second deposition process with the one or more process parameter values ​​to deposit the second layer of the multi-layer stack having approximately the target second thickness over the first layer.

Description

[Technical Field]

[0001]

[0001] Embodiments of the present disclosure relate to feedforward control of multi-layer stacks during device manufacturing. Additionally, the present embodiments relate to feedforward control of downstream processes in a multi-process manufacturing sequence based on optical measurements performed after upstream processes in the multi-process manufacturing sequence. [Background technology]

[0002] To develop a manufacturing process sequence for forming components on a substrate, engineers perform one or more designs of experiments (DoE) to determine process parameter values ​​for each process in a sequence of processes to be performed in the manufacturing process sequence. In a DoE, multiple different process parameter values ​​are typically tested by processing a substrate with different process parameter values ​​for each manufacturing process. A device or component, including one or more layers deposited and / or etched during the manufacturing process sequence, is then tested at end-of-line, where end-of-line corresponds to the completion of the component or device. As a result of such testing, one or more end-of-line performance metric values ​​are determined. Results of one or more DoEs may be used to determine target process parameter values ​​for one or more process parameters of the manufacturing processes in the manufacturing process sequence and / or to determine target layer properties (also referred to herein as film properties) for layers deposited and / or etched by one or more of the manufacturing processes in the manufacturing process sequence.

[0003] Once target process parameter values ​​and / or target layer characteristics are determined, substrates are processed according to a manufacturing process sequence, and the predetermined process parameter values ​​and / or layer characteristics determined based on the DoE results are used for each process in the manufacturing process sequence. Engineers then expect the processed substrates to have similar properties to the substrates processed during DoE, and they also expect the manufacturing equipment or components containing the layers formed by the manufacturing process sequence to have target end-of-line performance measurements. However, there is often variation between the film properties determined during DoE and the film properties of films on product substrates, which can result in changes in end-of-line performance measurements. Furthermore, each process chamber may be slightly different from other process chambers and can produce films with different film properties. Furthermore, process chambers may change over time, and therefore the films produced by those process chambers may also change over time, even when using the same process recipe. Summary of the Invention

[0004] Some of the embodiments described herein range from a substrate processing system comprising at least one transfer chamber, a first process chamber connected to the at least one transfer chamber, a second process chamber connected to the at least one transfer chamber, an optical sensor configured to perform optical measurements on the first layer after the first layer is deposited on the substrate, and a computing device operably connected to at least one of the first process chamber, the second process chamber, the transfer chamber, or the optical sensor. The first process chamber is configured to perform a first process for depositing a first layer of a multi-layer stack on the substrate, and the second process chamber is configured to perform a second process for depositing a second layer of the multi-layer stack on the substrate. The computing device receives a first optical measurement of the first layer after a first process is performed on the substrate, the first optical measurement indicating a first thickness of the first layer; determines a target second thickness for a second layer of the multi-layer stack based on the first thickness of the first layer; and causes a second process chamber to perform a second process to deposit the second layer having approximately the target second thickness on the first layer.

[0005]

[0005] In additional or related embodiments, a method includes processing a substrate in a first process chamber using a first deposition process to deposit a first layer of a multilayer stack on the substrate, removing the substrate from the first process chamber, measuring a first thickness of the first layer using an optical sensor, determining a target second thickness for a second layer of the multilayer stack based on the first thickness of the first layer, determining one or more process parameter values ​​for a second deposition process that will achieve the second target thickness for the second layer, and processing the substrate in the second process chamber using the second deposition process having one or more process parameter values ​​to deposit a second layer of the multilayer stack having approximately the target second thickness on the first layer.

[0006]

[0006] In some embodiments, a method includes receiving or generating a training dataset including a plurality of data items, each data item of the plurality of data items including a combination of layer thicknesses for a plurality of layers of a multi-layer stack and an end-of-line performance measurement for an apparatus including the multi-layer stack; and training a machine learning model based on the training dataset to receive as input a single layer thickness or at least two layer thicknesses of the multi-layer stack and to output at least one of a target thickness for a single remaining layer of the multi-layer stack, a target thickness for at least two remaining layers of the multi-layer stack, or a predicted end-of-line performance measurement for the apparatus including the multi-layer stack.

[0007] Numerous other features are provided in accordance with these and other aspects of the present disclosure. Other features and aspects of the present disclosure will become more fully apparent from the following detailed description, the claims, and the accompanying drawings.

[0008]

[0008] The present disclosure is illustrated by way of example, and not by way of limitation, in the figures of the accompanying drawings, in which like reference numerals indicate like elements. It should be noted that various references to "an" or "one" embodiment in this disclosure do not necessarily refer to the same embodiment, and such references mean at least one. [Brief explanation of the drawings]

[0009] [Figure 1A] 1 is a top schematic view of a first example manufacturing system according to an embodiment. [Figure 1B]

[0010] FIG. 2 is a schematic top view of a second manufacturing system according to an embodiment. [Figure 2A]

[0011] 1 is a flowchart of a method for performing feed-forward control of one or more processes in a DRAM bitline formation process, according to an embodiment. [Figure 2B]

[0012] 1 illustrates a schematic side view of a portion of a substrate including poly plugs, DRAM bitline stacks, and a hardmask layer, according to an embodiment. [Figure 3]

[0013] 3 shows a simplified side view of a system 300 for measuring layer thickness on a substrate in a cluster tool, according to one embodiment of the present disclosure. [Figure 4]

[0014] 1 is a flowchart for a method of performing feedforward control of one or more downstream processes in a process sequence for a multi-layer stack based on optical measurements of films obtained from one or more previously performed processes in the process sequence, according to an embodiment. [Figure 5]

[0015] 1 is a flowchart for a method of performing feedforward control of a downstream etching process in a process sequence based on optical measurements of films obtained from one or more previously performed deposition processes, according to an embodiment. [Figure 6]

[0016] 1 is a flowchart for a method for performing feedforward control of one or more downstream processes in a process sequence based on optical measurements of a film obtained from one or more previously performed processes in the process sequence, according to an embodiment. [Figure 7]

[0017] 1 is a flowchart for a method of updating the training of a machine learning model used to control a downstream process in a process sequence based on optical measurements of one or more layers formed by one or more processes in the process sequence. [Figure 8]

[0018] 1 is a flowchart for a method of performing a design of experiments (DoE) associated with a manufacturing process sequence for forming one or more layers on a substrate, according to an embodiment. [Figure 9]

[0019] 1 is a flowchart for a method for training a model to determine target thicknesses of one or more remaining layers, process parameter values ​​for forming one or more layers, and / or end-of-line performance measurements based on thickness values ​​of one or more layers formed by one or more processes in a manufacturing process sequence, according to an embodiment. [Figure 10]

[0020] 1 shows a diagrammatic representation of a machine in the example form of a computing device upon which a set of instructions may be executed, for causing the machine to perform any one or more of the methodologies discussed herein. DETAILED DESCRIPTION OF THE INVENTION

[0010]

[0021]

[0003] Embodiments described herein relate to a method for performing feedforward control of one or more processes not yet performed in a manufacturing process sequence based on thickness measurements of one or more layers formed by one or more previously performed processes in the manufacturing process sequence. In one embodiment, the thicknesses of one or more previously formed layers of a multilayer stack are used to determine target thicknesses and / or process parameter values ​​to achieve the target thicknesses for one or more remaining layers to be formed in the multilayer stack. In one embodiment, the thicknesses of one or more layers already formed on a substrate are used to determine target process parameter values ​​to use for an etching process performed to etch the one or more previously deposited layers. In an embodiment, a trained machine learning model is used to determine, based on the thicknesses of the one or more layers, the thicknesses of one or more additional layers to be formed, the process parameter values ​​to use for forming the additional one or more layers, the process parameter values ​​to use for etching the previously deposited one or more layers, and / or predicted end-of-line performance measurements for an apparatus or component including the layer or layers. Embodiments also extend to training a machine learning model to determine, based on an input of the thickness of one or more layers, the thickness of one or more additional layers to be formed, the process parameter values ​​used to form the additional one or more layers, the process parameter values ​​used to etch one or more layers already formed, and / or predicted end-of-line performance measurements of the layer or devices or components that include the layer. Examples of machine learning models that may be trained include linear regression models, Gaussian regression models, and neural networks such as convolutional neural networks.

[0011]

[0022] Traditionally, a one-time DoE is performed to determine recipe set points for process parameters for each manufacturing process in a manufacturing process sequence (e.g., including a sequence of deposition and / or etch processes). Once recipe set points are established for each process in the manufacturing process sequence, an assumption is made that each process chamber executing a recipe for a process in the manufacturing process sequence will use the process parameter set points determined for that process, achieving the film quality and film characteristics determined during DoE for the manufacturing process sequence. However, variations often exist between process chambers and / or process parameters for the process chambers drift over time. Such variations and / or drifts cause these process chambers to achieve process parameter values ​​that differ from the values ​​actually set in the process recipe. For example, a process recipe for a manufacturing process may include a target temperature of up to 200°C, but a first process chamber, when set to 200°C, may actually reach an actual temperature of 205°C. Additionally, a second process chamber, when set to 200°C, may actually reach an actual temperature of 196°C. Such deviations from the predetermined process parameter values ​​of a process recipe can cause one or more properties of a film deposited using a manufacturing process to change from the target properties. For example, two different chambers running the same deposition process may form layers of different thicknesses, where the layer on a first substrate may have a thickness above the target thickness and the layer on a second substrate may have a thickness below the target thickness. This layer may be one layer of a multi-layer stack for the ultimately formed device, and such changes in the film properties can have a detrimental effect on the ultimately formed device.

[0012]

[0023] For multilayer stacks, if the thickness of the first layer of the multilayer stack deviates from its target thickness, such deviation can have a detrimental effect on a device including the multilayer stack. However, if a thickness deviation is detected before additional layers of the multilayer stack are deposited, then the target thickness of one or more of those additional layers can be adjusted so that the final multilayer stack has an end-of-line performance measurement similar to that which the multilayer stack would have had if the first layer had its target thickness. Similarly, if one or more of the first two layers of the multilayer stack is detected to have a thickness that deviates from the target thickness before the deposition of the additional layer, then this information can be used to adjust the target thickness for one or more remaining layers of the multilayer stack to improve the end-of-line performance of a device including the multilayer stack. In embodiments, optical sensors are positioned in the transfer chamber, load lock, or via and are used to measure the thickness of the deposited layer after the deposition process. The measured thickness can be used to adjust future processes that result in the deposition of additional layers and / or etching of existing layers to improve the end-of-line performance of a device including the deposited layer.

[0013]

[0024] In one example, the systems and methods described in embodiments herein can be used to provide feed-forward control of one or more layers in a DRAM bit line stack. The DRAM bit line stack can include a barrier metal layer, a barrier layer, and a bit line metal layer. The sensing margin can depend on the thickness of each of the barrier metal layer, the barrier layer, and the bit line metal layer. A machine learning model can be trained to receive the barrier metal layer thickness and / or the barrier layer thickness as input and output a target barrier layer thickness and / or bit line metal layer thickness. The machine learning model can additionally output a predicted sensing margin for the DRAM bit line stack including the barrier metal layer, the barrier layer, and the bit line metal layer with the input and / or output thickness values. In this manner, by measuring the thickness of the layers of the DRAM bit line stack after each layer is formed, the process used to form the next layer or layers can be accurately adjusted for any deviations of the previously formed layers from the target thicknesses for those layers. Such adjustments can improve the sensing margin for a DRAM memory module including the DRAM bit line stack. The same techniques also work for any other type of multi-layer stack to improve other end-of-line performance measures, such as device electrical characteristics.

[0014]

[0025] In an embodiment, a computing device analyzes layers of a multi-layer stack and performs stack-level optimization. Stack-level information may be used, for example, to optimize power-performance-area-and-cost (PPAC) for a device including the multi-layer stack. Feedforward decisions may be made for one unit process using information from one or more previous unit processes. Processing logic may use complex spectra from multiple unit processes as input to one or more generated ML models, allowing optimization of the operation of the entire stack as opposed to optimizing individual processes.

[0015]

[0026] Referring now to the figures, FIG. 1A is an illustration of a cluster tool 100 (also referred to as a system or fabrication system) configured for substrate fabrication, such as post poly plug fabrication, DRAM bitline formation, three-dimensional (3D) NAND fabrication (e.g., ONON gate formation and / or OPOP gate formation), etc., in accordance with at least some embodiments of the present disclosure. The cluster tool 100 includes one or more vacuum transfer chambers (VTMs) 101, 102, a factory interface 104, multiple processing chambers / modules 106, 108, 110, 112, 114, 116, 118, and a process controller 120 (controller). A server computing device 145 may also be connected to the cluster tool 100 (e.g., to the controller 120 of the cluster tool 100). In embodiments having two or more VTMs, such as that shown in FIG. 1A, one or more pass-through chambers (referred to as vias) may be provided to facilitate vacuum transfer from one VTM to another. In an embodiment consistent with that shown in FIG. 1A, two pass-through chambers may be provided (eg, pass-through chamber 140 and pass-through chamber 142).

[0016]

[0027] The factory interface 104 includes a loading port 122 configured to receive one or more substrates, for example, from a Front Opening Unified Pod (FOUP) or other suitable substrate-containing box or carrier, to be processed using the cluster tool 100. The loading port 122 can include one or more loading areas 124a-124c, which can be used to load one or more substrates. Three loading areas are shown, although more or fewer loading areas can be used.

[0017]

[0028] The factory interface 104 includes an atmospheric transfer module (ATM) 126 used to transfer substrates loaded into the loading port 122. More specifically, the ATM 126 includes one or more robotic arms 128 (shown in dotted lines) configured to transfer substrates from the loading areas 124a-124c to the ATM 126 through doors 135 (shown in dotted lines, also referred to as slit valves) that connect the ATM 126 to the loading port 122. Each loading port (124a-124c) typically has one door to allow substrate transfer from the respective loading port to the ATM 126. The robotic arms 128 are also configured to transfer substrates from the ATM 126 to the load locks 130a-130b through doors 132 (shown in dotted lines, one for each load lock) that connect the ATM 126 to air locks 130a-130b. There may be more or less than two load locks, but for illustrative purposes only, two load locks (130a and 130b) are shown, each having a door for connection to the ATM 126. The load locks 130a-b may or may not be batch load locks.

[0018]

[0029] The load locks 130a, 130b can be maintained at either atmospheric or vacuum pressure environments under the control of the controller 120 and serve as intermediate or temporary holding spaces for substrates being transferred to / from the VTMs 101, 102. The VTM 101 includes a robot arm 138 (shown in dotted lines) configured to transfer substrates from the load locks 130a, 130b to one or more of the multiple processing chambers 106, 108 (also referred to as process chambers) or to one or more pass-through chambers 140, 142 (also referred to as vias) without breaking vacuum, i.e., while maintaining the vacuum pressure environment within the VTM 102 and the multiple processing chambers 106, 108 and pass-through chambers 140, 142. The VTM 102 includes a robot arm 138 (dotted line) configured to transfer substrates from airlocks 130a, 130b to one or more of the multiple processing chambers 106, 108, 110, 112, 114, 116, and 118 without breaking vacuum, i.e., while maintaining the vacuum pressure environment within the VTM 102 and the multiple processing chambers 106, 108, 110, 112, 114, 116, and 118.

[0019]

[0030] In certain embodiments, the load locks 130 a , 130 b may be omitted, and the controller 120 may be configured to move the substrate directly from the ATM 126 to the VTM 102 .

[0020]

[0031] A door 134, e.g., a slit valve door, connects each of the load locks 130a, 130b to the VTM 101. Similarly, a door 136, e.g., a slit valve door, connects each processing module to the VTM (e.g., either VTM 101 or VTM 102) to which it is coupled. The multiple processing chambers 106, 108, 110, 112, 114, 116, and 118 are configured to perform one or more processes. Examples of processes that may be performed by one or more of the processing chambers 106, 108, 110, 112, 114, 116, and 118 include a cleaning process (e.g., a pre-cleaning process to remove surface oxide from a substrate), an annealing process, a deposition process (e.g., for depositing a cap layer, a hard mask layer, a barrier layer, a bit line metal layer, a barrier metal layer, etc.), an etching process, etc. Examples of deposition processes that may be performed by one or more of the process chambers include physical vapor deposition (PVD), chemical vapor deposition (CVD), atomic layer deposition (ALD), etc. Examples of etching processes that may be performed by one or more of the process chambers include plasma etch processes. In one exemplary embodiment, process chambers 106, 108, 110, 112, 114, 116, and 118 are configured to perform processes typically associated with a post-poly plug fabrication sequence and / or a dynamic random access memory (DRAM) bit line stack fabrication sequence. In one exemplary embodiment, process chambers 106, 108, 110, 112, 114, 116, and 118 are configured to perform processes typically associated with a 3D NAND formation sequence, such as forming an ONON gate or an OPOP gate, which may include processes for depositing a multilayer stack of alternating layers of insulators and conductors (e.g., SiO2 and SiN, or SiO2 and polysilicon).

[0021]

[0032] In an embodiment, one or more of the components of the cluster tool 100 include optical sensors 147a, 147b configured to measure a property such as a thickness of a layer or film on a substrate. In one embodiment, optical sensor 147a is disposed in via 140 and optical sensor 147b is disposed in via 147b. Alternatively or additionally, one or more optical sensors 147a-b may be disposed within VTM 102 and / or VTM 101. Alternatively or additionally, one or more optical sensors 147a-b may be disposed within load lock 130a and / or load lock 130b. Alternatively or additionally, one or more optical sensors 147a-b may be disposed in one or more of process chambers 106, 108, 110, 112, 114, 116, and 118. One or more optical sensors 147a-b may be configured to measure a film thickness of a layer deposited on a substrate. In one embodiment, optical sensors 147a-b correspond to optical sensor 300 of FIG. 3. In some embodiments, optical sensors 147a-b measure film thickness after each layer of a multi-layer stack is formed on a substrate. One or more optical sensors 147a-b measure film thickness between processes in a manufacturing process sequence and may be used to inform decisions about how to perform further processes in the manufacturing process sequence. In embodiments, optical measurements indicative of film thickness may be performed on a substrate without removing the substrate from a vacuum environment.

[0022]

[0033] The controller 120 (e.g., a tool and equipment controller) may control various aspects of the cluster tool 100, such as gas pressures in the processing chambers, individual gas flows, spatial flow ratios, plasma power in the various process chambers, temperatures of various chamber components, radio frequency (RF) or electrical conditions of the processing chambers, etc. The controller 120 may receive signals from and send commands to any of the components of the cluster tool 100, such as the robot arms 128, 138, the process chambers 106, 108, 110, 112, 114, 116, and 118, the load locks 130a-b, the slit valve doors, the optical sensors 147a-b and / or one or more other sensors, and / or other processing components of the cluster tool 100. Thus, the controller 120 may control the start and stop of processing, adjust the deposition rate and / or thickness of the target layer, adjust the process temperature, adjust the type or mix of deposition compositions, adjust the etch rate, etc. Controller 120 may also receive and process measurement data (eg, optical measurement data) from various sensors (eg, optical sensors 147a-b) and make decisions based on such measurement data.

[0023]

[0034] In various embodiments, controller 120 may be and / or include a computing device, such as a personal computer, a server computer, a programmable logic controller (PLC), a microcontroller, etc. Controller 120 includes (or is) one or more processing devices, which may be general-purpose processing devices, such as a microprocessor, a central processing unit, etc. More specifically, the processing devices may be complex instruction set computing (CISC) microprocessors, reduced instruction set computing (RISC) microprocessors, very long instruction word (VLIW) microprocessors, or processors implementing other instruction sets or combinations of instruction sets. The processing devices may also be one or more special-purpose processing devices, such as an application-specific integrated circuit (ASIC), a field-programmable gate array (FPGA), a digital signal processor (DSP), a network processor, etc. Controller 120 may include data storage devices (e.g., one or more disk drives and / or solid-state drives), main memory, static memory, a network interface, and / or other components. The processor of the controller 120 may execute instructions to perform any one or more of the methodologies and / or embodiments described herein. The instructions may be stored in a computer-readable storage medium, which may include a main memory, a static memory, a secondary memory, and / or a processor (during execution of the instructions).

[0024]

[0035] In one embodiment, the controller 120 includes a feedforward engine 121. The feedforward engine 121 may be implemented in hardware, firmware, software, or a combination thereof. The feedforward engine 121 is optionally configured to receive and process optical measurement data, including results of reflectance measurements performed by an optical sensor such as a spectrometer. The feedforward engine 121 may calculate the optical measurement data (e.g., reflectance signals) after a layer is formed on a substrate and / or after a layer on a substrate is etched to determine one or more target thickness values ​​and / or other target properties for the layer. The feedforward engine 121 may further determine updated target thicknesses and / or other target properties for one or more additional layers of the multilayer stack, determine target process parameter values ​​to use for a process to form a layer having the updated target thicknesses and / or other properties, determine target process parameter values ​​for a process to use for etching the one or more layers, and / or predict one or more end-of-line performance measurements for an apparatus or component that includes the layer. Examples of end-of-line performance measurements that may be measured include signal margin, yield, voltage, power, device operating speed, device latency, and / or other performance variables.

[0025]

[0036] In one embodiment, the feedforward engine 121 includes a predictive model 123 that can correlate film thicknesses and / or other film characteristics of one or more layers with predicted values ​​of end-of-line performance measurements. Additionally or alternatively, the predictive model 123 can output recommended target layer thicknesses and / or other target layer characteristics of to-be-deposited layers based on input of thicknesses and / or other layer characteristics of one or more previously deposited layers. Additionally or alternatively, the predictive model 123 can output target process parameter values ​​for process parameters of one or more yet-to-be-executed processes in a manufacturing process sequence. The yet-to-be-executed processes can be, for example, deposition processes and / or etch processes. In one embodiment, the predictive model 123 is a trained machine learning model, such as a neural network, a Gaussian regression model, or a linear regression model.

[0026]

[0037] The feedforward engine 121 may input measured thicknesses and / or other layer characteristics of one or more already formed layers into the predictive model 123 and receive, as output target thicknesses and / or other target layer characteristics of one or more additional layers, target process parameter values ​​to achieve the target thicknesses, target process parameter values ​​for the etch process to be performed on the one or more layers, and / or predicted values ​​of end-of-line performance measurements. The process recipes to be performed to form the additional layers and / or etch the one or more layers may then be adjusted based on the output of the predictive model 123. Thus, the feedforward engine 121 may predict end-of-line problems during the manufacturing process (i.e., before the end-of-line is reached) and may further adjust one or more process recipes for processes not yet performed in the manufacturing process sequence to avoid the predicted end-of-line problems.

[0027]

[0038] In one example, a first one of the process chambers 106, 108, 110, 112, 114, 116, and 118 may be a deposition chamber for depositing a barrier metal layer, a second one of the process chambers may be a deposition chamber for depositing a barrier layer, and a third one of the process chambers may be a chamber for depositing a bit line metal layer. A manufacturing process sequence may include a first process recipe for depositing the barrier metal layer, a second process recipe for depositing the barrier layer, and a third process recipe for depositing the bit line metal layer. Each of the process recipes may be associated with a target layer thickness achieved by the respective process recipe. The first deposition chamber may execute the process recipe for depositing the barrier metal layer. One or more optical sensors 147a-b may be used to measure the thickness of the barrier metal layer. The feedforward engine 121 may then determine that the measured thickness deviates from the target thickness of the barrier metal layer. The feedforward engine 121 may use the predictive model 123 to determine a new target thickness for the barrier layer and / or the bit line metal layer based on the measured thickness of the barrier metal layer. For example, if the barrier metal layer is too thick, then the thickness of the barrier layer and / or the thickness of the bit line metal layer may be adjusted accordingly (e.g., by increasing and / or decreasing one or both of the target thicknesses of the barrier layer and the bit line metal layer). New process parameter values ​​for the process recipe for forming the barrier layer may be determined, and the second process chamber may execute the adjusted process recipe to form the barrier layer having the new target thickness.

[0028]

[0039] The substrate may be measured again by optical sensors 147a-b to determine the thickness of the barrier layer. The thickness of the barrier metal layer and the thickness of the barrier layer may then be compared to the target thicknesses for these two layers to determine any deviations from the target thicknesses. If such deviations are identified, then feed forward engine 121 may adjust the target thickness of the bit line metal layer. Feed forward engine 121 may use predictive model 123 to determine a new target thickness of the bit line metal layer based on the measured thicknesses of the barrier metal layer and the barrier layer. For example, if the barrier metal layer is too thick and the barrier layer is too thin, then the thickness of the barrier layer and / or the thickness of the bit line metal layer may be adjusted accordingly (e.g., by increasing and / or decreasing one or both of the target thicknesses of the barrier layer and the bit line metal layer). New process parameter values ​​for the process recipe for forming the metal bit line layer may be determined, and the third process chamber may execute the adjusted process recipe to form the metal bit line layer having the new target thickness.

[0029]

[0040] The substrate may be measured again by optical sensors 147a-b to determine the thickness of the metal bit-line layer. The thicknesses of the metal barrier layer, the barrier layer, and the metal bit-line layer may then be used by feed-forward engine 121 to predict the value of an end-of-line performance measurement. If the predicted value falls outside of specification, a decision may be made to discard the substrate rather than expend additional resources to complete the fabrication of a device or component that is predicted to fail final inspection. Additionally or alternatively, if the end-of-line performance measurement falls below a performance threshold, the process chamber that deposited the too-thick or too-thin layer may be removed from service and / or scheduled for maintenance. Thus, feed-forward engine 121 may perform diagnostics regarding the health of the process chamber, and maintenance may be scheduled for the process chamber, if appropriate.

[0030]

[0041] Controller 120 may be operably connected to server 145. Server 145 may be or may include a computing device that acts as a factory floor server that interfaces with some or all tools in a fabrication facility. Server 145 may send instructions to controllers of one or more cluster tools, such as cluster tool 100. For example, server 145 may receive signals from and send commands to controller 120 of cluster tool 100.

[0031]

[0042] In various embodiments, server 145 may be and / or include a computing device such as a personal computer, a server computer, a programmable logic controller (PLC), a microcontroller, etc. Server 145 may include (or be) one or more processing devices, which may be general-purpose processing devices such as a microprocessor, a central processing unit, etc. More specifically, the processing devices may be complex instruction set computing (CISC) microprocessors, reduced instruction set computing (RISC) microprocessors, very long instruction word (VLIW) microprocessors, or processors implementing other instruction sets or combinations of instruction sets. The processing devices may also be one or more special-purpose processing devices such as an application-specific integrated circuit (ASIC), a field-programmable gate array (FPGA), a digital signal processor (DSP), a network processor, etc. Server 145 may include data storage devices (e.g., one or more disk drives and / or solid-state drives), main memory, static memory, a network interface, and / or other components. A processing unit of server 145 may execute instructions to perform any one or more of the methodologies and / or embodiments described herein. The instructions may be stored in a computer-readable storage medium, which may include a main memory, a static memory, a secondary memory, and / or a processing unit (during execution of the instructions).

[0032]

[0043] In some embodiments, server 145 includes feedforward engine 121 and predictive model 123. Server 145 may include feedforward engine 121 and predictive model 123 in addition to or instead of controller 120, which includes feedforward engine 121 and predictive model 123. In some embodiments, controller 120 and / or server 145 correspond to computing device 1000 of FIG.

[0033]

[0044] In some examples, one or more processes may be performed on a substrate in a first cluster tool (e.g., cluster tool 100) to form one or more films on the substrate, and one or more processes (e.g., an etch process optionally performed after performing a lithography process on the substrate) may be performed on the substrate in another cluster tool. Optical measurements may be performed in the first cluster tool and / or the second cluster tool to determine predicted end-of-line performance and / or to make adjustments for one or more further processes to be performed on the substrate. In such embodiments, server 145 may communicate with controllers of both cluster tools to adjust feed-forward control of one or more processes not yet performed based on measured thicknesses of one or more layers formed on the substrate through processes already performed in the manufacturing process sequence.

[0034]

[0045] 1B is an illustration of a cluster tool 150 configured for substrate fabrication, e.g., post-poly plug fabrication, in accordance with at least some embodiments of the present disclosure. Cluster tool 150 includes a vacuum transfer chamber (VTM) 160, a factory interface 164, a plurality of chambers / modules 152, 154, 156 (some or all of which may be process chambers), and a controller 170. A server computing device 145 may also be connected to cluster tool 150 (e.g., to controller 170 of cluster tool 150).

[0035]

[0046] The factory interface 164 includes one or more loading ports configured to receive one or more substrates, for example from front-opening unified pods (FOUPs) 166a, 166b or other suitable substrate-containing boxes or carriers, to be processed using the cluster tool 150.

[0036]

[0047] The factory interface 164 includes an atmospheric transfer module (ATM) used to transfer substrates loaded into the loading port. More specifically, the ATM includes one or more robotic arms configured to transfer substrates from a loading area to the ATM through which the ATM connects to the loading port. The robotic arms are also configured to transfer substrates from the ATM to the load locks 158a-b through doors connecting the ATM to the load locks 158a-b. The load locks 158a-b can be maintained at either atmospheric or vacuum pressure under the control of the controller 170 and serve as intermediate or temporary holding spaces for substrates being transferred to and from the VTM 160. The VTM 160 includes a robotic arm 162 configured to transfer substrates from the load locks 158a-b to one or more of the multiple processing chambers 152, 154, 156 without breaking vacuum, i.e., while maintaining the vacuum pressure environment within the VTM 160 and the multiple chambers 152, 154, 156.

[0037]

[0048] In the illustrated embodiment, optical sensors 157a-b are disposed in load locks 158a-b, respectively, to perform optical measurements on substrates passing through the load locks 158a-b. Alternatively or additionally, one or more optical sensors may be disposed within VTM 160 and / or one of chambers 152, 154, 156.

[0038]

[0049] The controller 170 (e.g., a tool and equipment controller) may control various aspects of the cluster tool 150, such as the gas pressures within the processing chamber, the individual gas flows, spatial flow ratios, the temperatures of various chamber components, the radio frequency (RF) or electrical conditions of the processing chamber, etc. The controller 170 may receive signals from and send commands to any of the components of the cluster tool 150, such as the robot arm 162, the process chambers 152, 154, 156, the load locks 158a-b, the optical sensors 157a-b, the slit valve door, one or more sensors, and / or other processing components of the cluster tool 150. In this manner, the controller 170 may control the start and stop of processing, adjust deposition rates, the type or mix of deposition compositions, etch rates, etc. The controller 170 may also receive and process measurement data (e.g., optical measurement data) from various sensors, such as the optical sensors 157a-b. Controller 170 may be substantially similar to controller 120 of FIG. 1A and may include a feedforward engine 121 (which may include, for example, a predictive model 123).

[0039]

[0050] Controller 170 is operably connected to server 145, which may also be operably connected to controller 120 of FIG. 1A.

[0040]

[0051] In one example, one or more processes are performed on a substrate by various process chambers 106, 116, 118, 114, 110, 112, 108 of cluster tool 100 to form one or more layers on the substrate. The thickness of the one or more layers may be measured using one or more optical sensors 147a-b. The measured thickness may be used by feedforward engine 121 to determine layer thicknesses for one or more yet-to-be-deposited layers, process parameters for processes for forming the yet-to-be-deposited layers, and / or process parameter values ​​for processes for etching previously-deposited layers. The substrate may then be removed from cluster tool 100 and placed in a lithography tool to pattern a mask layer on the substrate. The substrate may then be placed in cluster tool 150. One or more etching processes may then be performed on the substrate by one or more of process chambers 152, 154, 156 of cluster tool 150 to etch one or more films. One or more target process parameter values ​​for the etch process may have been output by the feedforward engine 121 based on the measured thickness(es) of one or more deposition layers. Alternatively or additionally, one or more deposition processes may be performed on the substrate by one or more of the process chambers 152, 154, 156 of the cluster tool 150 to deposit one or more layers of a multi-layer stack. The target thickness(es) of such films may have been output by the feedforward engine 121 based on the measured thickness(es) of the deposited layer(s).

[0041]

[0052] In one embodiment, the process chambers of cluster tool 100 and / or cluster tool 150 are configured to perform one or more DRAM bit line stack processes (e.g., for post-poly plug fabrication). Alternatively, cluster tool 100 and / or cluster tool 150 may be configured to perform other processes, such as a 3D NAND deposition process.

[0042]

[0053] 2A is a flowchart for a method 220 of performing feedforward control of one or more processes in a DRAM bitline formation process, according to an embodiment. FIG. 2B shows a schematic side view of a portion of a substrate 200 including a poly plug 202, a DRAM bitline stack 201 (including a barrier metal 204, a barrier layer 206, and a bitline metal layer 208), and a hard mask layer 210, according to an embodiment. The poly plug 202 can be formed outside of the cluster tool 100. The DRAM bitline stack 201 can be formed inside the cluster tool 100 according to the method 220 without breaking vacuum between the deposition of the various layers of the DRAM bitline stack 201.

[0043]

[0054] In step 225 of method 220, substrate 200 may be loaded into loading port 122 via one or more of loading areas 124a-124c. Robot arm 128 of ATM 126, under the control of controller 120, may transfer substrate 200 having poly plug 202 from loading area 124a to ATM 126. Robot arm 128 may then place substrate 200 into load locks 130a-b, which may be pumped down to a vacuum under the control of controller 120. Controller 120 may then direct robot arm 138 to transfer substrate 200 to one or more of the processing chambers to complete fabrication of substrate 200, i.e., to complete bit line stack processing on top of poly plug 202 on substrate 200.

[0044]

[0055] In step 230, the robot arm 138, under the control of the controller 120, can retrieve the substrate 200 from the load locks 130a-b and place the substrate in a pre-clean chamber (e.g., process chamber 106). The transfer of the substrate 200 from the load locks to the process chamber 106 can be performed without breaking vacuum (i.e., a vacuum pressure environment is maintained within VTM 101 and VTM 102 while the substrate 200 is transferred to the pre-clean chamber). The processing chamber 106 can be used to perform one or more pre-clean processes to remove contaminants that may be present on the substrate 200, such as native oxidation that may be present on the substrate 200.

[0045]

[0056] In step 235, the controller 120 directs the robot arm 138 to open the door 136 and transfer the substrate 200 to the next processing chamber, which may be a barrier metal deposition chamber, such as the process chamber 108. The transfer of the substrate 200 from the process chamber 106 to the process chamber 108 may be performed without drawing a vacuum. The process chamber then performs a deposition process to form a barrier metal layer 204 over the poly plug 202. The barrier metal may be, for example, one of titanium (Ti) and tantalum (Ta).

[0046]

[0057] In step 240, the controller 120 directs the robot arm 138 to remove the substrate 200 from the process chamber 108 and directs the optical sensors 147a-b to generate optical measurements of the barrier metal layer 204 and determine the thickness of the barrier metal layer 204. For example, the controller 120 may direct the robot arm 138 to transfer the substrate under vacuum from the processing chamber 108 to one of the pass-through chambers 140, 142. The controller 120 may direct the optical sensors 147a-b to generate optical measurements of the barrier metal layer 204 while the substrate 200 is in the pass-through chamber 140, 142.

[0047]

[0058] In step 245, controller 120 determines a target thickness for barrier layer 206 based on the measured thickness of barrier metal layer 202. Additionally, controller 120 may determine a target thickness for bitline metal layer 208. Determining the target thickness for the barrier layer and / or barrier metal layer may be performed using, for example, a trained machine learning model, such as feedforward engine 121 and / or predictive model 123. Steps 240, 245 may be performed without breaking vacuum on substrate 200.

[0048]

[0059] In one embodiment, in step 250, the controller 120 instructs the robot arm 139 to transfer the substrate 200 to another process chamber (e.g., process chamber 116) without breaking vacuum and instructs the process chamber to perform an annealing step on the barrier metal layer 204. In some embodiments, steps 240 and / or 245 can be performed after step 250. The annealing process can be any suitable annealing process, such as a rapid thermal processing (RTP) anneal.

[0049]

[0060] In step 255, the controller 120 may instruct the robot arm 139 to transfer the substrate 200 from the pass-through chambers 140, 142 or from an annealing process chamber (e.g., process chamber 116) to a barrier layer deposition chamber (e.g., process chamber 110) without breaking vacuum. The processing chamber 110 may be configured, for example, to perform a barrier layer deposition process on the substrate 200 (e.g., deposit a barrier layer 206 on the barrier metal layer 204). The barrier layer 206 may be, for example, one of titanium nitride (TiN), tantalum nitride (TaN), or tungsten nitride (WN).

[0050]

[0061] In step 260, the controller 120 directs the robot arm 138 or the robot arm 139 to remove the substrate 200 from the barrier layer deposition chamber and directs the optical sensors 147a-b to generate optical measurements of the barrier layer 206 and determine the thickness of the barrier layer 206. For example, the controller 120 may direct the robot arm 139 to transfer the substrate under vacuum from the processing chamber 108 to one of the pass-through chambers 140, 142. The controller 120 may direct the optical sensors 147a-b to generate optical measurements of the barrier layer 206 while the substrate 200 is in the pass-through chamber 140, 142.

[0051]

[0062] In step 265, controller 120 determines a target thickness for bitline metal layer 208 based on the measured thickness of barrier layer 206 and the measured thickness of barrier metal layer 204. Determining the target thickness for bitline metal layer 208 may be performed using, for example, feedforward engine 121 and / or a trained machine learning model, such as predictive model 123. Steps 260 and 265 may be performed without breaking vacuum on substrate 200.

[0052]

[0063] In step 270, the controller 120 may direct the robot arm 139 to transfer the substrate 200 from the processing chamber 110 to, for example, a bit line metal deposition process chamber (e.g., processing chamber 112) without breaking vacuum. The bit line metal deposition chamber may be configured to perform a bit line metal deposition process on the substrate 200 (e.g., deposit a bit line metal layer 208 on the barrier layer 206). The bit line metal layer may be, for example, one of tungsten (W), molybdenum (Mo), ruthenium (Ru), iridium (Ir), or rhodium (Rh).

[0053]

[0064] In step 275, the controller 120 directs the robot arm 139 to remove the substrate 200 from the bitline metal layer deposition chamber and directs the optical sensors 147a-b to generate optical measurements of the bitline metal layer 208 and determine the thickness of the bitline metal layer 208. For example, the controller 120 may direct the robot arm 139 to transfer the substrate under vacuum from the processing chamber 112 to one of the pass-through chambers 140, 142. The controller 120 may direct the optical sensors 147a-b to generate optical measurements of the bitline metal layer 208 while the substrate 200 is in the pass-through chamber 140, 142.

[0054]

[0065] In step 280, the controller 120 predicts the value of an end-of-line performance measurement based on the measured thickness of the metal bit line layer 208, the measured thickness of the barrier layer 206, and the measured thickness of the barrier metal layer 204. The determination of the end-of-line performance measurement may be performed using, for example, a trained machine learning model, such as the feedforward engine 121 and / or the predictive model 123. Steps 275 and 280 may be performed without breaking vacuum on the substrate 200.

[0055]

[0066] In one embodiment, in step 285, the controller 120 directs the robot arm 139 to transfer the substrate 200 to an annealing process chamber (e.g., process chamber 116) without breaking vacuum and directs the process chamber to perform an annealing process on the bitline metal layer 208. In some embodiments, steps 275 and / or 280 can be performed after step 285. The annealing process can be any suitable annealing process, such as a rapid thermal processing (RTP) anneal.

[0056]

[0067] In some embodiments in which an annealing step is performed in step 285, the annealed substrate 200 may be transferred to another processing chamber in step 290 for depositing an optional capping layer 209 on the bit line metal layer 208. For example, the annealed substrate 200 including the bit line metal layer 208 may be transferred under vacuum, for example with the robot arm 139, from the annealing chamber (e.g., processing chamber 116) to a capping layer deposition chamber (e.g., processing chamber 118) for depositing a capping layer on the annealed bit line metal layer 208.

[0057]

[0068] In step 295, the controller 120 may direct the robot arm 139 to transfer the substrate 200 to a hard mask deposition chamber (e.g., processing chamber 114) without breaking vacuum. The hard mask deposition chamber is configured to perform a hard mask deposition process on the substrate 200 (e.g., deposit a hard mask layer 210 over the bit line metal layer 208 and / or cap layer 209). The hard mask may be, for example, one of silicon nitride (SiN), silicon oxide (SiO), or silicon carbide (SiC).

[0058]

[0069] Performing each of the above sequences in an integrated tool (eg, cluster tool 100) further advantageously avoids oxidation of the bitline metal during the grain growth anneal.

[0059]

[0070] After the DRAM bit line stacks and hard mask layer 210 are formed, the substrate 200 may be removed from the cluster tool 100 and processed using a lithography tool to pattern the hard mask 210. The substrate may then be transferred to the cluster tool 150, which may perform one or more etching processes to etch one or more layers of the DRAM bit line stacks. In some embodiments, in step 280, the controller 120 further determines one or more process parameter values ​​for the etching process to be performed on the DRAM bit line stacks based on the thicknesses of the metal barrier layer, the barrier layer, and / or the metal bit line layer. These process parameter values ​​may be communicated to the controller 170. The controller 170 may then direct an etching process chamber (e.g., process chamber 152 or 154) to perform the etching process using the determined one or more etching process parameter values.

[0060]

[0071] The method 220 may result in DRAM bit line stacks with improved end-of-line performance characteristics compared to DRAM bit line stacks formed using conventional processing techniques.

[0061]

[0072] FIG. 3 illustrates a simplified side view of an optical sensor system 300 for measuring layer thicknesses on substrates in a cluster tool, according to one aspect of the present disclosure. The optical sensor system may correspond, for example, to optical sensors 147a-b, 157-b of FIGS. 1A-B in embodiments. System 300 may include, for example, chamber 303, which may be a transfer chamber (e.g., VTM 101, 102), load lock chamber 130a-b, pass-through chamber 140, 142, or other chamber of the cluster tool. In one embodiment, chamber 303 is a measurement chamber attached to a facet of the cluster tool (e.g., to a facet of the VTM).

[0062]

[0073] Chamber 303 includes an interior space at vacuum pressure, which may be part of the vacuum environment of one or more VTMs (e.g., VTMs 101, 102). Chamber 303 may include a window 320. Window 320 may be, for example, a transparent crystal, glass, or other transparent material. The transparent crystal may be made of a transparent ceramic material, or may be made of a durable transparent material such as sapphire, diamond, quartz, silicon carbide, or a combination thereof.

[0063]

[0074] In an embodiment, system 300 further includes a light source 301 (e.g., a broadband light source or other electromagnetic radiation source), an optical combiner 304 (e.g., a collimator or mirror), a spectrometer 325, controllers 120, 170, and optionally a server 145. Light source 301 and spectrometer 325 may be optically coupled to optical combiner 304 through one or more optical fiber cables 332.

[0064]

[0075] In various embodiments, the optical coupler 304 can be adapted to collimate or otherwise transmit light in two directions along the optical path. The first direction can include light from the light source 301 that is collimated and transmitted into the chamber 303 through the window 320. The second direction can be reflected light that is reflected from the substrate 304, passes through the window 320 again, and returns into the optical coupler 304. The reflected light can be collected into the fiber optic cable 332 and then directed to the spectrometer 325 in the second direction along the optical path. Additionally, the fiber optic cable 332 can be coupled between the spectrometer 325 and the light source 301 for efficient transmission of light from the light source 301 to the transparent crystal 120 and back to the spectrometer 325.

[0065]

[0076] In an embodiment, the light source emits light in a spectrum between approximately 200 and 800 nm, and the spectrometer 325 also has a wavelength range between 200 and 800 nm. The spectrometer 325 may be adapted to detect the spectrum of reflected light received from the optical coupler 304, for example, light that reflects off a substrate in the chamber 303, returns through the window 320, and is collected by the optical coupler 304 into the fiber optic cable 332.

[0066]

[0077] The controllers 120 , 170 may be coupled to both the light source 301 , the spectrometer 325 , and the chamber 303 .

[0067]

[0078] In one embodiment, the controller 120, 170 may instruct the light source 301 to receive a light spectrum from the spectrometer 325 after it has been turned on. The controller 120, 170 may also leave the light source off and receive a second spectrum from the spectrometer 325 when the light source 301 is off. The controller 120, 170 may subtract the second spectrum from the first spectrum to determine an instantaneous reflectometry signal. The controller 120, 170 may then mathematically fit the reflectometry signal to one or more thin film models to determine one or more optical thin film properties of the thin film being measured.

[0068]

[0079] In some embodiments, the one or more optical thin film properties may include a film thickness, a refractive index (n), and / or an extinction coefficient (k) value. The refractive index is the ratio of the speed of light in a vacuum to the speed of light in the film. The extinction coefficient is a measure of how much light is absorbed in the film. The controller 120, 170 may use the n and k values ​​to determine the composition of the film. The controller 120, 170 may be further configured to analyze the data of one or more properties of the film. The controller 120, 170 may then use a feedforward engine to determine a target thickness value for the deposited layer, target process parameter values ​​for the deposition process and / or the etch process, and / or end-of-line performance characteristics, as described above. Alternatively, the server 145 may use a feedforward engine to determine a target process parameter value for the deposition process and / or the etch process, and / or end-of-line performance characteristics, as described above.

[0069]

[0080] It should be noted that embodiments are discussed herein with reference to a particular characteristic (i.e., thickness) of one or more layers to determine target thicknesses of the additional layers, process parameter values ​​for additional processes to be performed, and / or end-of-line performance characteristics. However, it should be understood that other layer characteristics of the deposited layer (e.g., refractive index n and / or extinction coefficient k, etc.) that can be determined based on optical measurements can be used instead of or in addition to thickness to determine target thicknesses of the additional layers, process parameter values ​​for additional processes to be performed, and / or end-of-line performance characteristics. Accordingly, any reference herein to the use of thickness measurements should be understood to apply to the use of thickness measurements alone or the use of thickness measurements in conjunction with refractive index and / or extinction coefficient. Additionally, it should be understood that other optically measurable film properties, such as refractive index and / or extinction coefficient, can be substituted for thickness measurements in the embodiments herein.

[0070]

[0081] FIG. 4 is a flowchart for a method 400 of performing feedforward control of one or more downstream processes in a process sequence for a multi-layer stack based on optical measurements of films obtained from one or more previously performed processes in the process sequence, according to an embodiment.

[0071]

[0082] At step 410 of method 400, a first manufacturing process is performed on a substrate in a first process chamber to form a first layer of a multi-layer stack on the substrate. In some embodiments, there are additional layers on the substrate below the first layer. The substrate can then be removed from the process chamber.

[0072]

[0083] An optical sensor is used to perform an optical measurement on the substrate to measure the first thickness of the first layer in step 415. Additionally or alternatively, one or more other properties of the first layer, such as the refractive index and / or extinction coefficient, may be measured using the optical sensor.

[0073]

[0084] In step 420, a computing device (e.g., a controller or server) determines a target thickness for one or more remaining layers of the multilayer stack based on the first thickness (and / or one or more other measured characteristics of the first layer). Additionally or alternatively, the computing device may determine one or more other target characteristics (e.g., a target refractive index, a target surface roughness, a target average grain size, a target grain orientation, etc.) for the one or more remaining layers based on the first thickness (and / or one or more other measured characteristics of the first layer). Additionally or alternatively, in step 420, the computing device may determine target process parameter values ​​for processes to be performed to form the one or more remaining layers. For example, the computing device may determine process parameter values ​​for process parameters such as deposition time, gas flow rate, temperature, pressure, plasma power, etc. for one or more deposition processes to be performed that will approximately result in the determined target layer thickness. Additionally, the computing device may predict one or more end-of-line performance measurements for the device or component including the multilayer stack having the measured thickness and having target thicknesses for one or more remaining layers. If the predicted end-of-line performance measurements are below a performance threshold, in some embodiments, the substrate may be scrapped or reworked. Additionally or alternatively, if the predicted end-of-line performance measurements are below a performance threshold, the process chamber that deposited the first layer may be scheduled for maintenance. In an embodiment, step 420 may be performed by inputting the measured thickness (and / or other characteristics) of the first layer into predictive model 123.

[0074]

[0085] In step 425, processing logic determines process parameter values ​​for one or more process parameters for a second manufacturing process to be performed to form a second layer of the multilayer stack. In one embodiment, the process parameter values ​​are determined by inputting a target thickness (and / or other target characteristics of the next layer to be deposited) into a table, function, or model. The table, function, or model may receive the target thickness (and / or other layer characteristics) and output process parameter values. In one embodiment, the model is a trained machine learning model, such as a neural network (e.g., a convolutional neural network) or a regression model, trained to output process parameter values ​​for a recipe based on the input target thickness and / or other input target characteristics of the layer. In one embodiment, the target process parameter values ​​were determined in step 420.

[0075]

[0086] In step 430, the substrate is transferred to a second process chamber, which performs a second manufacturing process on the substrate using the determined process parameter values ​​to form a second layer of the multi-layer stack on the substrate, after which the substrate may be removed from the second process chamber.

[0076]

[0087] An optical sensor is used to perform optical measurements on the substrate to measure the actual second thickness of the second layer in step 435. Additionally or alternatively, one or more other properties of the second layer, such as the refractive index or extinction coefficient, can be measured using the optical sensor.

[0077]

[0088] In step 440, a computing device (e.g., a controller or server) determines a target thickness for one or more remaining layers of the multilayer stack based on the first thickness of the first layer and the actual second thickness of the second layer (and / or one or more other measured properties of the first layer and the second layer). Additionally or alternatively, the computing device may determine one or more other target properties (e.g., target refractive index, target surface roughness, target average grain size, target grain orientation, etc.) for the one or more remaining layers based on the first thickness (and / or one or more other measured properties of the first layer) and the actual second thickness (and / or one or more other measured properties of the second layer). Additionally or alternatively, in step 440, the computing device may determine target process parameter values ​​for processes to be performed to form the one or more remaining layers. For example, the computing device may determine process parameter values ​​for process parameters such as deposition time, gas flow rate, temperature, pressure, plasma power, etc. for one or more deposition processes to be performed that will approximately result in the determined target layer thickness. Additionally, the computing device may predict one or more end-of-line performance measurements for the device or component including the multilayer stack having the measured first and second thicknesses and target thicknesses for one or more remaining layers. If the predicted end-of-line performance measurements are below a performance threshold, then the substrate may be scrapped or reworked, in some embodiments, and / or the second process chamber may be scheduled for maintenance. Step 440 may, in some embodiments, be performed by inputting the measured thicknesses (and / or other characteristics) of the first and second layers into the predictive model 123. In some embodiments, the same trained machine learning model is used in steps 420 and 440. Alternatively, different trained machine learning models may be used in steps 420 and 440. For example, the trained machine learning model used in step 420 may be trained to receive only a single thickness, while the trained machine learning model used in step 440 may be trained to receive two thickness values.

[0078]

[0089] In one embodiment where the multi-layer stack includes two layers, the computing device determines predicted end-of-line performance measurements but does not determine target thicknesses for any remaining layers in step 440. In such an embodiment, method 400 may end at step 440.

[0079]

[0090] In step 445, processing logic may determine process parameter values ​​for one or more process parameters for a third manufacturing process to be performed to form a third layer of the multi-layer stack. In one embodiment, the process parameter values ​​are determined by inputting a target thickness (and / or other target characteristics of the next layer to be deposited) into a table, function, or model. The table, function, or model may receive the target thickness (and / or other layer characteristics) and output process parameter values. In one embodiment, the model is a trained machine learning model, such as a neural network (e.g., a convolutional neural network) or a regression model, trained to output process parameter values ​​for a recipe based on the input target thickness and / or other input target characteristics of the layer. In one embodiment, the target process parameter values ​​were determined in step 440.

[0080]

[0091] In step 450, the substrate is transferred to a third process chamber, which performs a third manufacturing process on the substrate using the determined process parameter values ​​to form a third layer of the multi-layer stack on the substrate, after which the substrate may be removed from the third process chamber.

[0081]

[0092] An optical sensor is used to perform optical measurements on the substrate to measure the actual third thickness of the third layer in step 455. Additionally or alternatively, one or more other properties of the third layer, such as the refractive index and / or extinction coefficient, may be measured using the optical sensor.

[0082]

[0093] In step 460, a computing device (e.g., a controller or server) determines a predicted end-of-line performance measurement based on the first thickness of the first layer, the measured second thickness of the second layer, and the measured third thickness of the third layer (and / or one or more other measured characteristics of the first layer, second layer, and third layer). If the end-of-line performance measurement is below a performance threshold, then the substrate may be scrapped or reworked, in some embodiments. Step 460 may be performed, in embodiments, by inputting the measured thicknesses (and / or other characteristics) of the first layer, second layer, and third layer into predictive model 123. In some embodiments, the same trained machine learning model is used in steps 420, 440, and 460. Alternatively, different trained machine learning models may be used in steps 420, 440, and 460. If there are additional layers to be deposited after the third layer, then in step 460, the computing device may additionally or alternatively determine a target thickness for the next layer and / or target process parameter values ​​to achieve the target thickness. Steps similar to steps 450-460 may then be performed for the next layer.

[0083]

[0094] FIG. 5 is a flowchart for a method 500 of performing feedforward control of a downstream etching process in a process sequence based on optical measurements of films obtained from one or more previously performed deposition processes, according to an embodiment.

[0084]

[0095] At step 510 of method 500, a first manufacturing process is performed on a substrate in a first process chamber to form a layer on the substrate. In some embodiments, there is an additional layer on the substrate below the first layer. In some embodiments, the layer is a layer of a multi-layer stack. The substrate can then be removed from the process chamber.

[0085]

[0096] An optical sensor is used to perform an optical measurement on the substrate to measure the first thickness of the first layer in step 515. Additionally or alternatively, one or more other properties of the first layer, such as the refractive index and / or extinction coefficient, may be measured using the optical sensor.

[0086]

[0097] In step 520, a computing device (e.g., a controller or server) determines target process parameter values ​​for one or more process parameters of an etch process to be performed on the deposition layer based on the first thickness (and / or one or more other measured properties of the first layer). In addition, the computing device may predict one or more end-of-line performance measurements for equipment or components that include the layer. If the predicted end-of-line performance measurements fall below a performance threshold, then in some embodiments, the substrate may be scrapped or reworked and / or maintenance may be scheduled for the process chamber. Step 520 may, in embodiments, be performed by inputting the measured thickness (and / or other properties) of the layer into predictive model 123.

[0087]

[0098] In step 530, the substrate is transferred to a second process chamber (e.g., an etch process chamber), which performs an etch process on the substrate using the determined process parameter values ​​to etch the layer. In one example, the layer deposited in step 510 may have been thicker than the target thickness, and the etch time for the etch process may be increased to accommodate the thicker layer. The substrate may then be removed from the second process chamber.

[0088]

[0099] An optical sensor may optionally be used to perform optical measurements on the substrate to measure the post-etch thickness of the layer in step 535. Additionally or alternatively, the optical sensor may be used to measure one or more other post-etch properties of the layer.

[0089]

[0100] In step 540, a computing device (e.g., a controller or server) may determine a predicted end-of-line performance measurement based on the layer thickness and / or the post-etch thickness of the layer (and / or one or more other measured properties of the layer). If the predicted end-of-line performance measurement is below a performance threshold, in some embodiments, the substrate may be scrapped or reworked. Step 540 may, in embodiments, be performed by inputting the measured thickness (and / or other properties) of the layer into predictive model 123. In some embodiments, steps 520 and 540 use the same trained machine learning model. Alternatively, steps 520 and 540 may use different trained machine learning models.

[0090]

[0101] FIG. 6 is a flowchart for a method 600 of performing feedforward control of one or more downstream processes in a process sequence based on optical measurements of a film obtained from one or more previously performed processes in the process sequence, according to an embodiment.

[0091]

[0102] In step 605 of method 600, a first manufacturing process is performed on the substrate in a first process chamber to form a layer on the substrate. In some embodiments, there are additional layers on the substrate below the first layer.

[0092]

[0103] An optical sensor is used to perform an optical measurement on the substrate to measure a first thickness of the first layer in step 610. Additionally or alternatively, one or more other properties of the first layer, such as a refractive index and / or an extinction coefficient, may be measured using the optical sensor.

[0093]

[0104] In step 615, a computing device (e.g., a controller or server) determines one or more process parameter values ​​for one or more future processes to be performed on the substrate based on the first thickness (and / or one or more other measured characteristics of the first layer). If additional layers are to be deposited on the substrate, the computing device may also optionally determine target thicknesses for one or more remaining layers. Additionally or alternatively, the computing device may determine one or more other target characteristics (e.g., target refractive index, target surface roughness, target average grain size, target grain orientation, etc.) for the one or more remaining layers based on the first thickness (and / or one or more other measured characteristics of the first layer). In addition, the computing device may predict one or more end-of-line performance measurements for an apparatus or component that includes the first layer having the measured thickness. If the predicted end-of-line performance measurements fall below a performance threshold, then the substrate may be scrapped or reworked, and / or the process chamber that deposited the first layer on the substrate may be scheduled for maintenance, in some embodiments. In an embodiment, step 615 may be performed by inputting the measured thickness (and / or other properties) of the first layer into predictive model 123.

[0094]

[0105] In step 620, the substrate is transferred to a second process chamber, which performs a second manufacturing process on the substrate using the determined process parameter values. The second manufacturing process can be, for example, a deposition process, an etching process, an annealing process, or some other process. For example, the second manufacturing process can be a deposition process to form a second layer of a multi-layer stack on the substrate.

[0095]

[0106] In step 625, the optical sensor may be used to perform optical measurements on the substrate after completion of the second manufacturing process. If the second process is a deposition process, then the optical measurements may measure one or more properties (e.g., thickness) of the additional deposited layer.

[0096]

[0107] In step 630, a computing device (e.g., a controller or server) may determine one or more process parameter values ​​for process parameters of one or more additional processes to be performed on the substrate based on the first thickness of the first layer and the optical measurements of the substrate (e.g., the second thickness of the second layer) determined in step 625. Additionally or alternatively, the computing device may determine a predicted value of an end-of-line performance measurement. If the predicted end-of-line performance measurement falls below a performance threshold, then the substrate, in some embodiments, may be scrapped or reworked, and / or the second process chamber may be scheduled for maintenance. Step 630, in embodiments, may be performed by inputting the measured thicknesses (and / or other properties) of the first and / or second layers into predictive model 123.

[0097]

[0108] In step 635, processing logic determines whether additional processes are to be performed whose results are measured using optical sensors. If additional processes are to be performed, the method returns to block 620, where the next process is performed in the next process chamber. If additional processes are not to be performed, the method proceeds to step 640. In step 640, once the device or component is complete (or reaches a stage of completion where one or more performance measurements can be measured), measurements are taken to determine end-of-line performance measurements. For example, sensing margins and / or other electrical characteristics of the device may be measured. The results of the measured end-of-line performance measurements, along with the measurement results determined in steps 610 and / or 625, may then be used to further train the machine learning model used in steps 615 and 630. For example, predictive model 123 may be continuously trained as new product lots are completed. As a result, the accuracy of predictive model 123 may continue to improve over time.

[0098]

[0109] 7 is a flowchart for a method 700 for updating the training of a machine learning model used to control a downstream process in a process sequence based on optical measurements of one or more layers formed by one or more processes in the process sequence. Method 700 may be used, for example, to periodically retrain predictive model 123. Method 700 may be performed by processing logic, which may include hardware, software, firmware, or a combination thereof. In an embodiment, method 700 is performed by controller 120, 170 and / or server 145 of FIGS. 1A-B.

[0099]

[0110] At operation 705 of method 700, end-of-line measurements are performed on a device or component including the multilayer stack to determine end-of-line performance measurements. At step 710, processing logic determines film thicknesses of one or more layers in the multilayer stack. The thickness of each respective layer may have been measured after deposition of that layer. For example, the layer thickness may have been measured according to any of methods 400-600. At step 715, processing logic generates training data items including the film thicknesses of the one or more layers and the end-of-line performance measurements. At step 720, processing logic then performs supervised learning on the trained machine learning model (e.g., predictive model 123) using the training data items to update the training of the machine learning model.

[0100]

[0111] 8 is a flowchart for a method 800 of performing a design of experiments (DoE) associated with a manufacturing process sequence for forming one or more layers on a substrate, according to an embodiment. Although shown in a particular sequence or order, the order of operations can be changed unless otherwise specified. Therefore, the illustrated embodiment should be understood as an example only, and the illustrated processes can be performed in a different order, and some processes can be performed in parallel. Additionally, in various embodiments, one or more processes can be omitted. Thus, not all processes are performed in all embodiments. Other process flows are possible.

[0101]

[0112] In step 805 of method 800, multiple versions of the fabrication process sequence are performed. Each version of the fabrication process sequence uses a different combination of process parameter values ​​for one or more processes in the sequence, resulting in a multilayer stack having a different combination of layer thicknesses. In one embodiment, the multilayer stack is a DRAM bit line stack, and each version of the DRAM bit line stack has a different combination of layer thicknesses for the barrier metal layer, the barrier layer, and the bit line metal layer. In some examples, the optimal layer thickness combination for the multilayer stack is known a priori, and that optimal combination of layer thicknesses, as well as one or more additional combinations of layer thicknesses above and / or below the optimal thickness, can be tested. For example, for a DRAM bit line stack, the optimal layer thicknesses can be 2 nm for the metal barrier layer, 3 nm for the barrier layer, and 20 nm for the metal bit line layer. Different versions of the DRAM bit line stack can be produced. Some versions vary only one of the thicknesses above or below the optimum thickness, some versions vary two of the thicknesses above and / or below the optimum thickness, and some versions vary all three of the thicknesses above and / or below the optimum thickness. As one example, approximately 300 substrates are processed to produce a multilayer stack having a range of thickness combinations. For each version of the manufacturing process sequence, one or more further processes can be performed on the substrates to produce a testable device or component.

[0102]

[0113] In step 810, one of the versions of the manufacturing process sequence is selected.

[0103]

[0114] In step 815, one or more measurements are performed on the representative substrate fabricated using the selected version of the fabrication process sequence to determine properties of one or more layers of the multilayer stack on the representative substrate. For example, destructive metrology measurements may be performed to determine the thickness of each layer of the multilayer stack on the substrate. Alternatively, the measurements may be performed in-line during the fabrication of the multilayer stack (e.g., by performing non-destructive optical measurements of each layer of the multilayer stack after the layer is formed).

[0104]

[0115] In step 820, a device or component may be fabricated using the substrate having the multilayer stack formed using a selected sequence of fabrication processes. In some embodiments, step 820 is performed before step 810. Examples of devices that may be formed include DRAM memory modules and 3D NAND memory modules.

[0105]

[0116] In step 825, one or more end-of-line performance measurements are measured on the fabricated device or component including the multi-layer stack formed by the selected version of the fabrication process. The performance measurements may include sensing margin, voltage, power, device speed, device latency, yield, and / or other performance parameters. In some embodiments, one or more electrical measurements are performed on the device or component to determine one or more electrical characteristics of the device or component. The electrical characteristics may correspond to or be end-of-line performance measurements of the device or component. For example, a sensing margin is the percentage of voltage supplied to a gate for a memory unit that is actually detected at the gate. A larger sensing margin is better than a smaller sensing margin because a device with a larger sensing margin may function with less voltage (e.g., a smaller voltage may be applied to the gate of a memory unit to change the state of the gate).

[0106]

[0117] Data items are generated for the selected version of the sequence of the manufacturing process in step 830. The data items may be training data items including layer thicknesses for each layer of the multi-layer stack and one or more end-of-line performance measurements.

[0107]

[0118] In step 835, a determination is made as to whether there are any remaining versions of the manufacturing process sequence that have not yet been tested (and for which data items have not yet been generated). If there are any untested versions of the manufacturing process sequence that remain, the method returns to step 810, where a new version of the manufacturing process sequence is selected and tested. If all of the versions of the manufacturing process sequence have been tested, the method continues to step 840.

[0108]

[0119] A training data set is generated in step 840. The training data set includes data items generated for each version of the sequence of the manufacturing process.

[0109]

[0120] FIG. 9 is a flowchart for a method 900 for training a model to determine target thicknesses of one or more remaining layers, process parameter values ​​for forming one or more layers, and / or end-of-line performance measures based on thickness values ​​of one or more layers formed by one or more processes in a manufacturing process sequence, according to an embodiment. Method 900 may be performed using the components described with reference to FIGS. 1A-3 , as will become apparent. For example, method 900 may be performed by controller 120, controller 170, and / or server 145 in an embodiment. At least some steps of method 900 may be performed by processing logic, which may include hardware (e.g., circuits, dedicated logic, programmable logic, microcode, etc.), software (e.g., instructions executed on a processing device to perform a hardware simulation), or a combination thereof. Although shown in a particular sequence or order, the order of operations can be changed unless otherwise specified. Therefore, the illustrated embodiment should be understood as an example only, and the illustrated processes may be performed in a different order, and some processes may be performed in parallel. Additionally, in various embodiments, one or more processes may be omitted. Thus, not all processes are performed in all embodiments, and other process flows are possible.

[0110]

[0121] At step 905 of method 900, processing logic receives a training data set (e.g., as may have been generated according to method 800). The training data set may include multiple data items, each of which includes one or more layer thicknesses and end-of-line performance measurements for a version of a sequence of a manufacturing process.

[0111]

[0122] In step 910, processing logic receives input of thicknesses for one or more layers of a multi-layer stack on the substrate and trains the model to output at least one of target thicknesses for one or more remaining layers of the multi-layer stack, target process parameter values ​​for process parameters of one or more future manufacturing processes to be performed on the substrate, and / or predicted end-of-line performance measurements.

[0112]

[0123] In one embodiment, the model is a machine learning model, such as a regression model trained using regression. Examples of regression models are regression models trained using linear regression or Gaussian regression. In one embodiment, at step 915, processing logic performs linear regression or Gaussian regression using the training dataset to train the model. A regression model predicts a value of Y given known values ​​of an X variable. The regression model may be trained using regression analysis, which may include interpolation and / or extrapolation. In one embodiment, parameters of the regression model are estimated using least squares. Alternatively, Bayesian linear regression, percentage regression, least absolute deviation, nonparametric regression, scenario optimization, and / or distance metric learning may be performed to train the regression model.

[0113]

[0124] In one embodiment, the model is a machine learning model such as an artificial neural network (also simply referred to as a neural network). The artificial neural network may be, for example, a convolutional neural network (CNN) or a deep neural network. In one embodiment, at step 920, processing logic performs supervised machine learning to train the neural network.

[0114]

[0125] Artificial neural networks generally include a feature representation component with a classifier or recurrent layer that maps features to a target output space. For example, a convolutional neural network (CNN) includes multiple layers of convolutional filters. Lower layers may perform pooling and address nonlinearities. A multilayer perceptron is typically added on top of the lower layers to map the features extracted by the convolutional layers to a decision (e.g., a classification output). Neural networks can be deep networks with multiple hidden layers or shallow networks with zero or a few (e.g., one or two) hidden layers. Deep learning is a class of machine learning algorithms that uses a cascade of multiple nonlinear processing units for feature extraction and transformation. Each successive layer uses the output from the previous layer as input. Neural networks can learn in a supervised (e.g., classification) and / or unsupervised (e.g., pattern analysis) manner. Some neural networks (e.g., deep neural networks) include a hierarchy of layers, with different layers learning different levels of representation corresponding to different levels of abstraction. In deep learning, each level learns to transform its input data into a more abstract and complex representation.

[0115]

[0126] Training a neural network can be accomplished with supervised learning methods, which involve feeding a training dataset of labeled inputs through the network, observing its outputs, defining an error (by measuring the difference between the output and the labeled values), and using techniques such as deep gradient descent and backpropagation to adjust the network's weights across all its layers and nodes so that the error is minimized. In many applications, repeating this process across many labeled inputs in the training dataset results in a network that can generate the correct output when presented with inputs different from those included in the training dataset. In high-dimensional settings, such as large images, this generalization is achieved when a sufficiently large and diverse training dataset is available.

[0116]

[0127] In an embodiment, the input is a feature vector including film properties (e.g., film thickness, etc.) of one or more layers, and the labels are performance measures such as end-of-line performance measures (e.g., electrical values ​​such as sensing margins). In one embodiment, a neural network is trained to receive as input film properties of one or more deposited layers and output one or more predicted performance measures, film properties of layers yet to be deposited, and / or process parameter values ​​for future processes to be performed on already deposited layers and / or to deposit additional layers.

[0117]

[0128] In step 925, the trained model is deployed. The trained model may be deployed, for example, to one or more process chambers and / or controllers of a cluster tool. Additionally or alternatively, the trained model may be deployed to a server connected to one or more controllers (e.g., controllers of one or more process chambers and / or controllers of one or more cluster tools). Deploying the trained model may include storing the trained model in a feedforward engine of the controller and / or server. Once the trained model is deployed, the controller and / or server may use the trained model to perform feedforward control of one or more manufacturing processes in a manufacturing process sequence.

[0118]

[0129] FIG. 10 shows a diagrammatic representation of a machine, in the exemplary form of a computing device 1000, upon which a set of instructions for causing the machine to perform any one or more of the methodologies discussed herein may be executed. In alternative embodiments, the machine may be connected (e.g., networked) to other machines in a local area network (LAN), an intranet, an extranet, or the Internet. The machine may operate as a server or a client machine in a client-server network environment, or as a peer machine in a peer-to-peer (or distributed) network environment. The machine may be a personal computer (PC), a tablet computer, a set-top box (STB), a personal digital assistant (PDA), a mobile phone, a web appliance, a server, a network router, switch, or bridge, or any machine capable of executing a set of instructions (sequential or non-sequential) that specify actions to be taken by the machine. Furthermore, while only a single machine is illustrated, the term “machine” shall also be interpreted to include any collection of machines (e.g., computers) that individually or jointly execute a set (or sets) of instructions to perform any one or more of the methodologies discussed herein.

[0119]

[0130] The example computing device 1000 includes a processing unit 1002, a main memory 1004 (e.g., read-only memory (ROM), flash memory, dynamic random access memory (DRAM) (such as synchronous DRAM (SDRAM) or Rambus DRAM (RDRAM)), static memory 1006 (e.g., flash memory, static random access memory (SRAM)), and a secondary memory (e.g., data storage device 1018), which communicate with each other via a bus 1030.

[0120]

[0131] Processing unit 1002 represents one or more general-purpose processors, such as a microprocessor, a central processing unit, etc. More specifically, processing unit 1002 may be a complex instruction set computing (CISC) microprocessor, a reduced instruction set computing (RISC) microprocessor, a very long instruction word (VLIW) microprocessor, a processor implementing other instruction sets, or a processor implementing a combination of instruction sets. Processing unit 1002 may also be one or more special-purpose processing units, such as an application-specific integrated circuit (ASIC), a field-programmable gate array (FPGA), a digital signal processor (DSP), a network processor, etc. Processing unit 1002 is configured to execute processing logic (instructions 1022) to perform the processes and steps discussed herein.

[0121]

[0132] Computing device 1000 may further include a network interface device 1008. Computing device 1000 may also include a video display unit 1010 (e.g., a liquid crystal display (LCD) or a cathode ray tube (CRT)), an alphanumeric input device 1012 (e.g., a keyboard), a cursor control device 1014 (e.g., a mouse), and a signal generating device 1016 (e.g., a speaker).

[0122]

[0133] The data storage device 1018 may include a machine-readable storage medium (or more specifically, a computer-readable storage medium) 1028 on which one or more sets of instructions 1022, embodying any one or more of the methodologies or functions described herein, are stored. The instructions 1022 may also reside, completely or at least partially, within the main memory 1004 and / or within the processing unit 1002 during execution thereof by the computer system 1000. The main memory 1004 and the processing unit 1002 also constitute computer-readable storage media.

[0123]

[0134] The computer-readable storage medium 1028 may also be used to store a software library containing the feedforward engine 121 and / or a method for invoking the feedforward engine 121. While the computer-readable storage medium 1028 is shown in the exemplary embodiment to be a single medium, the term "computer-readable storage medium" should be interpreted to include a single medium or multiple media (e.g., a centralized or distributed database, and / or associated caches and servers) that store one or more sets of instructions. The term "computer-readable storage medium" should also be interpreted to include any medium that can store or encode a set of instructions for execution by a machine, causing the machine to perform any one or more of the methodologies described herein. Accordingly, the term "computer-readable storage medium" should be interpreted to include, but is not limited to, non-transitory computer-readable media such as solid-state memory, optical media, and magnetic media.

[0124]

[0135] The modules, components, and other features described herein (e.g., in connection with Figures 1A-3) may be implemented as discrete hardware components or integrated into the functionality of hardware components such as ASICS, FPGAs, DSPs, or similar devices. In addition, modules may be implemented as firmware or functional circuitry within a hardware device. Furthermore, modules may be implemented as any combination of hardware devices and software components, or may be implemented solely in software.

[0125]

[0136] Some portions of the detailed description have been presented in terms of algorithms and symbolic representations of operations on data bits within a computer memory. These algorithmic descriptions and representations are the means used by those skilled in the data processing arts to most effectively convey the substance of their invention to others skilled in the art. An algorithm is here, and generally, conceived to be a self-consistent sequence of steps leading to a desired result. These steps require physical manipulations of physical quantities. Usually, though not necessarily, these quantities take the form of electrical or magnetic signals capable of being stored, transferred, combined, compared, and otherwise manipulated. It has proven convenient, principally for reasons of common usage, to refer to these signals as bits, values, elements, symbols, characters, terms, numbers, or the like.

[0126]

[0137] It should be borne in mind, however, that all of these and similar terms are to be associated with the appropriate physical quantities and are merely convenient labels applied to such quantities. As will become apparent from the following description, unless otherwise specified, throughout this specification, discussions utilizing terms such as "receiving," "identifying," "determining," "selecting," "providing," "storing," and the like, should be understood to refer to the operations and processes of a computer system or similar electronic computing device that manipulates and converts data represented as physical quantities (electronic quantities) in the computer system's registers and memory into other data that are similarly represented as physical quantities in the computer system's memory or registers, or other storage, transmission, or display device for such information.

[0127]

[0138] Embodiments of the present invention also relate to apparatus for performing the processes herein. This apparatus may be specially constructed for the purposes discussed, or may comprise a general-purpose computer system that is selectively programmed by a computer program stored on the computer system. Such computer program may be stored on a computer-readable storage medium such as, but not limited to, any type of disk, including floppy disks, optical disks, CD-ROMs, and magneto-optical disks, read-only memory (ROM), random-access memory (RAM), EPROM, EEPROM, magnetic disk storage media, optical storage media, flash memory devices, other types of machine-accessible storage media, or any type of medium suitable for storing electronic instructions, each coupled to a computer system bus.

[0128]

[0139] The foregoing description sets forth numerous specific details, such as examples of particular systems, components, methods, etc., to provide a thorough understanding of some embodiments of the present disclosure. However, it will be apparent to those skilled in the art that at least some embodiments of the present disclosure may be practiced without these specific details. In other instances, well-known components or methods are not described in detail or are presented in simple block diagram form to avoid unnecessarily obscuring the present disclosure. Thus, the specific details described are merely exemplary. Certain implementations may differ from these example details and still be considered within the scope of the present disclosure.

[0129]

[0140] Throughout this specification, a reference to "one embodiment" or "an embodiment" means that a particular feature, structure, or characteristic described in connection with an embodiment is included in at least one embodiment. Thus, the appearances of the phrase "in one embodiment" or "in an embodiment" in various places throughout this specification are not necessarily all referring to the same embodiment. In addition, the term "or" is intended to mean an inclusive "or" rather than an exclusive "or." When the term "about" or "approximately" is used herein, it is intended to mean that the stated nominal value is accurate to within ±10%.

[0130]

[0141] Although the steps of the methods herein are illustrated and described in a particular order, the order of the steps of each method may be changed such that certain steps are performed in reverse order and certain steps are performed at least partially concurrently with other steps. In alternative embodiments, the order of the separate steps or substeps may be intermittent and / or alternating.

[0131]

[0142] It should be understood that the above description is intended to be illustrative, and not limiting. Many other embodiments will become apparent to those skilled in the art upon reading and understanding the above description. The scope of the disclosure should, therefore, be determined with reference to the appended claims, along with the full scope of equivalents to which such claims are entitled.

Claims

1. at least one transfer chamber; a first process chamber connected to the at least one transfer chamber, the first process chamber configured to perform a first process for depositing a first layer of a multi-layer stack on a substrate; a second process chamber connected to the at least one transfer chamber, the second process chamber configured to perform a second process for depositing a second layer of the multi-layer stack on the substrate; and a third process chamber connected to the at least one transfer chamber, the third process chamber configured to perform a third process for depositing a third layer of the multi-layer stack on the substrate; an optical sensor configured to perform an optical measurement on the first layer after it has been deposited on the substrate, and to perform another optical measurement on the combination of the first and second layers after it has been deposited; a computing device operably connected to at least one of the first process chamber, the second process chamber, the third process chamber, the transfer chamber, or the optical sensor; Equipped with a device including the multilayer stack is fabricated on the substrate; The computing device receiving a first optical measurement of the first layer after the first process is performed on the substrate, the first optical measurement indicating a first thickness of the first layer; predicting, with a trained machine learning model, a target second thickness for the second layer of the multilayer stack and a target third thickness for the third layer of the multilayer stack based at least in part on the first thickness of the first layer; predicting, by the trained machine learning model, prior to actually depositing the multi-layer stack, one or more end-of-line performance measurements of the device based at least in part on the first thickness of the first layer, the predicted second thickness of the second layer to be deposited, and the predicted third thickness of the third layer to be deposited, wherein the one or more predicted end-of-line performance measurements are selected from the group consisting of signal margin, voltage, device latency, and sensing margin; causing the second process chamber to perform the second process to deposit the second layer on the first layer having approximately the target second thickness; receiving second optical measurements of the first layer and the second layer after the second process is performed on the substrate, the second optical measurements indicating an actual second thickness of the second layer, the first thickness of the first layer and the actual second thickness of the second layer providing a total thickness of a partial multilayer stack; predicting, with the trained machine learning model, a new target third thickness of the third layer of the multilayer stack based on the total thickness of the partial multilayer stack, including the first thickness of the first layer and the actual second thickness of the second layer; and predicting, by the trained machine learning model, prior to actually depositing the third layer of the multilayer stack, the one or more updated end-of-line performance measurements for the apparatus based at least in part on the total thickness of the partial multilayer stack, including the first thickness of the first layer and the actual second thickness of the second layer, and the predicted new target third thickness of the third layer to be deposited; causing the third process chamber to perform the third process to deposit the third layer having approximately the new target third thickness on the second layer; automatically scheduling maintenance of a process chamber when the one or more end-of-line performance measures predicted by the trained machine learning model fall below a performance threshold; A substrate processing system.

2. To determine the target third thickness for the third layer of the multi-layer stack, the computing device: inputting the first thickness of the first layer and the actual second thickness of the second layer into the trained machine learning model, the model being trained to determine, for input of the first thickness of the first layer and the actual second thickness of the second layer, a target third thickness of the third layer that, when combined with the first thickness of the first layer and the actual second thickness of the second layer, results in an optimal end-of-line performance measure for a device including the multilayer stack. The substrate processing system according to claim 1 ,

3. the optical sensor is further configured to perform the optical measurement on the third layer; The computing device receiving a third optical measurement of the third layer after the third process is performed on the substrate, the third optical measurement indicating an actual third thickness of the third layer; determining the one or more predicted end-of-line performance measures for the device including the multilayer stack based on the first thickness of the first layer, the actual second thickness of the second layer, and the actual third thickness of the third layer; The substrate processing system of claim 1 , further comprising:

4. To determine the one or more predicted end-of-line performance measures for the device including the multi-layer stack, the computing device: inputting the first thickness of the first layer, the actual second thickness of the second layer, and the actual third thickness of the third layer into the trained machine learning model, the model being trained to predict the one or more predicted end-of-line performance measures for the device including the multi-layer stack for the input of the first thickness of the first layer, the actual second thickness of the second layer, and the actual third thickness of the third layer. The substrate processing system according to claim 3 , wherein the substrate processing system performs the steps of:

5. 10. The substrate processing system of claim 1, wherein the multi-layer stack comprises a dynamic random access memory (DRAM) bit line stack.

6. To determine the target second thickness for the second layer of the multi-layer stack, the computing device: inputting the first thickness of the first layer into the trained machine learning model, the model being trained to output the target second thickness of the second layer that, when combined with the first thickness of the first layer, results in an optimal end-of-line performance measure for a device including the multilayer stack. The substrate processing system according to claim 1 ,

7. The substrate processing system of claim 6 , wherein the trained machine learning model comprises a neural network.

8. 7. The substrate processing system of claim 6, wherein the trained machine learning model is further trained to output at least one of a target third thickness of the third layer of the multi-layer stack or the one or more predicted end-of-line performance measurements for the apparatus including the multi-layer stack.

9. The substrate processing system of claim 1 , wherein the optical sensor comprises a spectrometer configured to measure the first thickness using reflectometry.

10. The substrate processing system of claim 1 , wherein the optical sensor is a component of the transfer chamber, a load lock chamber connected to the transfer chamber, or a pass-through station.

Images