Radical sensor substrate

Abstract

Embodiments disclosed herein include a sensor device. In one embodiment, a sensor device comprises: a substrate; a first sensor of a first type on the substrate wherein a catalytic layer is provided as at least a portion of the first sensor; the second sensor of the first type is located on the substrate and adjacent to the first sensor; and a cover over the substrate wherein an opening through the cover is provided over the first sensor and the second sensor.

Description

[0001] CROSS-REFERENCE TO RELATED APPLICATIONS

[0002] This application claims priority to U.S. patent application No. 17 / 876,392, filed on July 28, 2022, the entire contents of which are incorporated herein by reference. Technical Field

[0003] Embodiments relate to the field of semiconductor manufacturing, and more particularly, to sensor substrates including sensors for detecting one or more process conditions in a semiconductor processing tool. Background Art

[0004] The use of radical species is common in semiconductor processing operations. For example, radical species may be used in etch processes, deposition processes, or similar processes. However, it is difficult to quantify the concentration of radical species in existing semiconductor processing tools. Therefore, it may be necessary to run test substrates to determine the concentration of radical species for a given processing operation. Even so, such analysis does not provide a measure of the radical species itself. Only the ultimate effect of the radical species can be determined.

[0005] In addition to free radical species sensing, other processing parameters may be required to perfect the processing conditions and achieve chamber matching. For example, temperature, pressure and similar parameters may also be required to match the chamber. Summary of the invention

[0006] Embodiments disclosed herein include sensor devices. In one embodiment, a sensor device includes: a substrate; a first sensor of a first type located on the substrate, wherein a catalytic layer is provided as at least a portion of the first sensor; a second sensor of the first type located on the substrate and adjacent to the first sensor; and a cover located over the substrate, wherein an opening through the cover is provided over the first sensor and the second sensor.

[0007] Embodiments disclosed herein may also include a method of determining a process parameter for a semiconductor process recipe. In one embodiment, the method includes: inserting a sensor substrate having a first sensor with a catalytic coating and a second sensor without a catalytic coating into a semiconductor process tool; igniting a plasma in the semiconductor process tool; obtaining a temperature reading using the first sensor and the second sensor; and calculating a radical density of the plasma based on a difference in the temperature readings of the first sensor and the second sensor.

[0008] Embodiments disclosed herein may also include a method of controlling a processing parameter of a semiconductor processing recipe. In one embodiment, the method includes: providing an artificial intelligence (AI) model and / or a machine learning (ML) model of a semiconductor processing tool; obtaining in-situ sensor readings using a sensor provided on a substrate inserted into the semiconductor processing tool; using the sensor readings as input to train the AI ​​model and / or the ML model; and deploying the AI ​​model and / or the ML model to change one or more processing parameters of the semiconductor processing recipe. BRIEF DESCRIPTION OF THE DRAWINGS

[0009] Figure 1 is a plan view of a sensor substrate including a plurality of sensors for determining concentration of free radical species according to one embodiment.

[0010] Figure 2 is a cross-sectional view of a model of a sensor substrate including a sensor for determining the concentration of free radical species and a processor under a cover.

[0011] Figure 3 is a plan view of a thermocouple sensor configuration for determining radical species concentration according to one embodiment.

[0012] Figure 4 is a cross-sectional view of a thermopile for determining radical species concentration according to one embodiment.

[0013] Figure 5 is a schematic diagram of a Wheatstone bridge for determining the concentration of free radical species according to one embodiment.

[0014] Figure 6 is a plan view of a sensor substrate with an (acoustic) resonator sensor for determining the concentration of radical species according to one embodiment.

[0015] Figure 7 is a cross-sectional view of a sensor substrate having a first sensor at a first Z height and a second sensor at a second Z height according to one embodiment.

[0016] Figure 8 is a process flow diagram of a process for measuring radical species density of a plasma using a sensor substrate according to one embodiment.

[0017] Fig. 9 An illustration of a semiconductor processing tool including a tool controller having a machine learning (ML) or artificial intelligence (AI) model that receives feedback from a sensor substrate is provided according to one embodiment.

[0018] Fig.10is a process flow diagram of a process of using a sensor substrate to inform an AI and / or ML algorithm to change one or more processing parameters of a semiconductor tool.

[0019] Fig.11 A block diagram of an exemplary computer system that may be used with a processing tool according to one embodiment is illustrated. DETAILED DESCRIPTION

[0020] The systems described herein include a sensor substrate including sensors for detecting one or more process conditions in a semiconductor processing tool. In the following description, numerous specific details are set forth to provide a thorough understanding of the embodiments. Those skilled in the art will appreciate that the embodiments may be practiced without these specific details. In other cases, well-known aspects are not described in detail to avoid unnecessarily obscuring the embodiments. Furthermore, it should be understood that the various embodiments shown in the drawings are illustrative representations and are not necessarily drawn to scale.

[0021] As mentioned above, in semiconductor processing tools, the ability to determine the concentration of radical species is desirable for many different reasons. For example, knowledge of the concentration of radical species can be used to improve process uniformity and improve chamber matching operations. However, there is currently no equipment that enables in-situ measurement of radical species concentration.

[0022] Therefore, embodiments disclosed herein include sensor substrates that implement free radical species concentration measurements. The free radical species concentration used herein may also be referred to as free radical (radical / free radical) concentration. In one embodiment, free radicals may include atoms or molecules containing one or more unpaired electrons in the valence shell or outer orbital. In one embodiment, the odd number of electrons in the free radicals can make the free radicals unstable and highly reactive. Therefore, free radical species can be used for etching or deposition processing. Some common free radical species include hydrogen (H), oxygen (O) and nitrogen (N). Free radicals can be generated by flowing a gas into a processing chamber and igniting a plasma. However, other methods of forming free radicals may also be used.

[0023] Free radicals can be detected by including a catalytic material on the sensor substrate. The catalytic material can selectively interact with the free radical species. The reaction of the catalytic material and the free radicals may cause a temperature increase. The increase in temperature may be related to the concentration of the free radical species. That is, when the concentration of the free radical species is relatively high, the temperature of the catalytic material will be higher than the temperature when the concentration of the free radical species is relatively low. In order to provide a more accurate measurement of the free radical concentration, the sensor substrate may include a first temperature sensor without a catalytic coating and a second temperature sensor with a catalytic coating. Therefore, a temperature difference measurement can be provided to isolate the heating effect caused by the reaction of the free radical species. In one embodiment, the temperature measurement sensor may include a thermocouple, a thermopile, a Wheatstone bridge, a resonator, or any other temperature sensor architecture.

[0024] While the embodiments disclosed herein are particularly useful for determining free radical species concentrations, embodiments may also include a sensor substrate having functionality for measuring temperature, pressure, and the like within a chamber. For example, a thermocouple, a thermopile, a Wheatstone bridge, or a resonator may be used to measure the temperature within a semiconductor processing tool. Additionally, sensors such as resonators or Pirani sensors can provide pressure measurements. For example, as the pressure changes, the resonant frequency of the resonator may change. Other types of sensors may also be used to determine the amount of deposition or etching of a layer.

[0025] In addition to discovering in-situ processing conditions, embodiments may also include functionality to feed in-situ measurements back to machine learning (ML) and / or artificial intelligence (AI) models. The ML and / or AI models may include digital twin functionality. The measured processing conditions may be used to inform or train the ML and / or AI models. Using ML and / or AI models may be useful for applications such as chamber matching, chamber seasoning, and similar applications.

[0026] Reference now Figure 1, shows a plan view of a sensor substrate 100 according to one embodiment. In one embodiment, the sensor substrate 100 can have a form factor of a semiconductor wafer (e.g., a silicon wafer). For example, the diameter of the sensor substrate 100 can be 200 mm, 300 mm, or 450 mm. However, other form factors can also be used for the sensor substrate 100. In one embodiment, the thickness of the sensor substrate 100 can be about 10 mm thick or thinner, or about 6 mm thick or thinner. The thickness of the sensor substrate 100 may be suitable for handling the sensor substrate 100 using existing robotic handling solutions in a semiconductor processing tool. Therefore, the sensor substrate 100 can be inserted into or retracted from a processing tool without opening the chamber.

[0027] In one embodiment, the sensor substrate 100 may include a substrate 105. The substrate 105 may be any suitable material. In a specific embodiment, the substrate 105 may be or include a circuit board (e.g., an organic board or a ceramic board, such as a printed circuit board (PCB) or the like). The substrate 105 may include electrical wiring (not shown) to electrically couple the sensor 110 with the control model 107.

[0028] Sensor 110 may include any type of sensor suitable for measuring temperature. For example, sensor 110 may include a thermocouple, a thermopile, a thermistor, a resistance bridge (e.g., a Wheatstone bridge), a resonator, and the like. In a particular embodiment, sensors 110 may each include a pair of sensors. The first sensor may not have a catalytic layer and the second sensor may have a catalytic layer. Thus, a temperature differential reading may be provided to accurately measure the free radical concentration within a semiconductor processing tool.

[0029] In one embodiment, there may be any number of sensors 110. For example, five or more sensors 110, ten or more sensors 110, or fifty or more sensors 110 may be included on the sensor substrate 100. The sensors 110 may be arranged in any desired pattern to determine the concentration of free radical species across the substrate. For example, the sensors 110 may be provided near the center of the sensor substrate 100, near the edge of the sensor substrate 100, and between the center and the edge. In one embodiment, the sensing area of ​​each sensor 110 may be 1 mm 2 and 20mm 2 For example, in some embodiments, the sensing area of ​​each sensor 110 may be approximately 10 mm 2 .

[0030] In one embodiment, the control model 107 may include a processor and a memory. The processor may be used to control the sensor 110, and the memory may store temperature readings. The processor may also determine the free radical concentration based on the temperature difference readings. The sampling rate of the sensor 110 may be between 1 Hz and 10 Hz, but the sampling rate of embodiments may also reach 100 Hz or higher. In one embodiment, the control model 107 may also include a battery or other power source to power the sensor 110.

[0031] In an alternative embodiment, one or more of the processor, memory, and power supply may be removed from the sensor substrate 100. For example, wireless power supply, wireless sensor drive (such as for resonator sensors), and wireless communication may be used in some embodiments. Offloading some processor, memory, and power supply applications from the sensor substrate 100 may enable higher operating temperatures of the sensor substrate 100 (e.g., up to 400° C. or up to 500° C.).

[0032] Reference now Figure 2 , shows a cross-sectional view of a sensor substrate 200 according to one embodiment. In one embodiment, the sensor substrate 200 may include a base substrate 205, such as a PCB or a common ceramic or metal carrier. In one embodiment, a plurality of sensors 210 may be provided on the base substrate 205. In one embodiment, each sensor 210 may include a reference sensor 213B and a catalytic sensor 213A. The reference sensor 213B may be exposed or may be completely covered by the cover 212. The catalytic sensor 213A may include a catalytic layer 214. The catalytic layer 214 is selected as a material that preferentially reacts with the free radical species of interest. For example, hydrogen radicals may preferentially react with gold (Au), stainless steel, silver (Ag), platinum (Pt), or copper (Cu). Oxygen may preferentially react with copper, stainless steel, or gold. Nitrogen may preferentially react with copper, gold, or stainless steel. The thickness of the catalytic layer 214 may be about 10 microns or less, about 1 micron or less, or about 500 nanometers or less. Although shown as distinct components, it is to be understood that the reference sensor 213B and the catalytic sensor 213A may be integrated into a single sensor component.

[0033] In one embodiment, the sensor substrate 200 may include a single type of catalytic material for the catalytic layer 214. In other embodiments, different catalytic materials may be provided for the various catalytic layers 214. Thus, the sensor substrate 200 may determine the concentration of more than one type of free radical species. For example, the catalytic layer 214 may include two or more of copper, gold, and stainless steel.

[0034] In one embodiment, the catalytic sensor 213A and the reference sensor 213B may be any suitable sensors, such as those described in more detail below. The temperature of the catalytic sensor 213A may be compared to the temperature of the reference sensor 213B to determine how much of the temperature increase corresponds to the reaction of the free radical species with the catalytic layer 214. Once the amount of heating caused by the reaction of the free radical species is known, the correlation with the concentration of the free radical species may be determined.

[0035] In one embodiment, the sensor substrate 200 may include a cover 212. The cover 212 may protect certain portions of the sensor substrate 200 from being exposed to the processing environment. For example, the control model 207, electrical wiring, and the like may be covered by the cover 212. In one embodiment, the cover 212 may include openings 216. These openings expose the sensor 210 or only the catalytic sensor 213A to the processing environment in order to measure processing conditions, such as free radical species concentrations. The cover 212 may be any suitable material, such as plastic, metal, ceramic, or the like. The cover 212 may withstand the processing environment being investigated. For example, the cover 212 may be a material that is resistant to etching chemicals (e.g., hydrogen etching chemicals).

[0036] Reference now Figure 3 , shows a plan view of a temperature sensor 310 that can be used for a sensor substrate 300 according to one embodiment. As shown, the sensor 310 includes a first sensor 313A and a second sensor 313B. The first sensor 313A may include a catalytic layer 314 on a sensor probe 318. The second sensor 313B may include a bare sensor probe 318.

[0037] Sensor 313 may be a thermocouple sensor. That is, first conductive trace 315A and second conductive trace 315B may be coupled to opposite ends of voltmeter 319. The material of conductive trace 315A is different from that of conductive trace 315B. Figure 3 314. A particular thermocouple architecture is shown in FIG. 314, but it is understood that any thermocouple architecture may be used. As the temperature increases, the voltage of the voltmeter 319 also increases. The measured voltage can be correlated to the temperature of the sensor probe 318. In particular, when exposed to free radical species, the temperature of the catalytic layer 314 increases more than the temperature of the sensor probe 318. By comparing the temperature of the sensor probe 318 of the sensor 313B with the temperature of the sensor probe 318 of the sensor 313A (which is located below the catalytic layer 314), the correlation with the free radical concentration can be determined. In one embodiment, the sensor 313A is substantially similar to the sensor 313B, but does not include the catalytic layer 314.

[0038] Reference now Figure 4, shows a cross-sectional view of a sensor substrate 400 according to one embodiment. In one embodiment, the sensor substrate 400 may include a pair of thermopiles 413A and 413B that pass through the base substrate 405. In one embodiment, the thermopiles 413A and 413B may include alternating conductive traces 421 and 422. Trace 421 may be a different conductive material than trace 422. Therefore, the thermopile 413 is able to read the temperature through the voltage difference between traces 421 and 422. Although Figure 4 A particular thermopile architecture is shown in FIG. 1 , but it is understood that any thermopile architecture may be used in the embodiments disclosed herein.

[0039] In one embodiment, the first thermopile sensor 413A may include a catalytic layer 414. The catalytic layer 414 may be provided on a top surface of the thermopile sensor 413A. The second thermopile sensor 413B may be provided without the catalytic layer 414. In one embodiment, the first thermopile sensor 413A may be substantially similar to the second thermopile sensor 413B, but does not include the catalytic layer 414. Thus, a temperature differential reading between the first thermopile sensor 413A and the second thermopile sensor 413B may be determined. The temperature differential reading may be correlated to the free radical species concentration. In other embodiments, the thermopile sensors 413A and 413B may be coplanar and located on the same surface (and located on a diaphragm to isolate the hot side from the cold side).

[0040] Reference now Figure 5 , shows a schematic diagram of a sensor substrate 500 according to one embodiment. As shown, the sensor architecture can include a resistor bridge architecture. In a specific embodiment, the resistor bridge architecture is a Wheatstone bridge architecture. For example, a set of four resistors R1-R4 can be provided in a ring architecture and connected to a power supply. For example, node A can be coupled to the positive terminal of the power supply and node C can be coupled to the negative terminal of the power supply. Terminals D and B can be coupled to a voltmeter circuit 502 to measure the output voltage (V O ).

[0041] In one embodiment, two opposing resistors R may include a catalytic layer (not shown), while the remaining resistors R may be formed without a catalytic layer. For example, resistors R2 and R3 may include a catalytic layer. Resistors R2 and R3 may also be variable resistors (as indicated by the lines passing through resistors R2 and R3). In other embodiments, resistors R1 and R4 may be resistors with catalytic layers, while resistors R2 and R3 may be formed without catalytic layers. That is, any two resistors on opposite sides of the bridge may be selected as resistors covered with catalytic material. In addition, any number of resistors R1-R4 may be variable resistors.

[0042] Reference now Figure 6 , shows a plan view of a sensor substrate 600 according to one embodiment. As shown, a pair of resonators 613A and 613B can be provided on a substrate 605. Each resonator can include a first interdigitated structure 623 and a second interdigitated structure 624. For example, the resonator 613 can be a surface acoustic wave (SAW) resonator, a bulk acoustic wave (BAW) resonator, or any other type of MEMS resonator architecture. The resonator 613 can be any material suitable for providing a resonator structure. For example, the resonator 613 can include a piezoelectric material. In addition, although three and four interdigitated fingers are shown in the figure, it is understood that any number of interdigitated fingers can be used. Although interdigitated fingers are shown, other resonator architectures can also be used.

[0043] Resonator 613 can detect temperature. For example, a change in the temperature of resonator 613 may cause a change in the resonant frequency of resonator 613. In addition, resonator 613 can also detect changes in pressure or layer thickness (e.g., pressure changes or layer thickness caused by deposition or etching). Changes in pressure and / or layer thickness may also cause predictable changes in the resonant frequency of resonator 613.

[0044] In one embodiment, the first resonator 613A may include a catalytic layer 614 on the interdigitated structures 623 and 624. The first resonator 613A may be adjacent to a second resonator 613B that does not include the catalytic layer 614. Thus, the second resonator 613B may be used as a reference signal to provide a differential measurement of the temperature increase due to the reaction of the free radical species with the catalytic layer 614.

[0045] In the embodiments described in more detail above, the temperature sensor architecture is provided at a single Z height. These embodiments can be used to provide a spatial analysis of the radical species concentration. That is, the radical species concentration in the XY plane can be determined. However, it is to be understood that the radical species concentration may also change in the Z direction. Therefore, the embodiments described herein may also include structures capable of measuring the radical species density at two or more Z heights within the chamber.

[0046] Figure 7 An example of such an embodiment is shown. As shown, the sensor substrate 700 can include a substrate 705 having multiple thicknesses. For example, the first thickness can be less than the second thickness by a distance H. The distance H can be about 1 mm or greater, or about 5 mm or greater. However, it is understood that in some embodiments, the distance H can also be less than 1 mm. In addition, although two different thicknesses are shown, it is understood that any number of different thicknesses can be provided adjacent to each other.

[0047] In one embodiment, a first sensor 713A may be provided on a first portion of the substrate 705 and a second sensor 713B may be provided on a second portion of the substrate 705. The top of the second sensor 713B may be higher than the top of the first sensor 713 A. Thus, temperature variation (and thus free radical concentration) may be provided for different Z heights.

[0048] In the illustrated embodiment, sensors 713A and 713B are connected to Figure 4 713A. That is, each sensor 713 is a thermopile including a first conductor 721 and a second conductor 722. Both sensors 713 may also be covered by a catalytic layer 714. In one embodiment, a similar sensor is provided adjacent to both the first sensor 713A and the second sensor 713B, but without the catalytic layer 714 (not shown). A sensor without a catalytic converter layer may be provided adjacent to and at the same height as the first sensor 713A and the second sensor 713B.

[0049] Although Figure 7 An example of a thermopile sensor is shown, but it is understood that any temperature sensor architecture described herein can be used to measure the change in radical concentration in the Z direction. For example, thermocouples, resistor bridges, and / or resonators can be used with Figure 7 The architecture shown is similar to the architecture shown.

[0050] Reference now Figure 8 , shows a process flow diagram describing a process 870 for determining a radical density for a plasma process, according to one embodiment.

[0051] In one embodiment, process 870 may begin with operation 871, which includes inserting a sensor substrate into a semiconductor processing tool, the sensor substrate having a first sensor with a catalytic coating and a second sensor without a catalytic coating. In one embodiment, the first sensor may be provided adjacent to the second sensor. The first sensor and the second sensor may be sensors of the same type. For example, the first sensor and the second sensor may include a thermocouple, a thermopile, a resistance bridge, a resonator, or the like. In a particular embodiment, a sensor similar to any of the sensors described in more detail above may be used with the sensor substrate.

[0052] In one embodiment, the catalytic coating may include any suitable material, such as gold, copper, stainless steel, silver, platinum or the like, and have different sensing area sizes or heights and the like. The catalytic coating may be selected to selectively detect free radicals such as hydrogen, oxygen and / or nitrogen. Although described as having a pair of sensors (one with a catalytic coating and one without a catalytic coating), it is understood that the sensor substrate may include any number of sensor pairs. Each sensor pair may include one uncoated sensor and one catalytically coated sensor. Variations / combinations of these sensors may be used to more accurately derive free radical density using algorithms including ML.

[0053] In one embodiment, process 870 may continue to perform operation 872, which includes igniting a plasma in a semiconductor processing tool. In one embodiment, the semiconductor processing tool may be a tool for any suitable semiconductor processing operation. For example, the semiconductor processing tool may be an etching tool, a deposition tool, a plasma processing tool, or the like. In one embodiment, the plasma is ignited above the sensor substrate. When the sensor substrate provides a plurality of sensor pairs, the sensor pairs may be distributed across the substrate to measure spatial radical density. In some embodiments, sensor pairs may also be provided at different Z heights to determine differences in radical species at different Z heights within the chamber.

[0054] In one embodiment, process 870 may continue to perform operation 873, which includes obtaining temperature readings using the first sensor and the second sensor. Since the first sensor has a catalytic coating on it, the temperature of the first sensor is expected to be higher than the temperature of the second sensor. The second sensor is used as a reference to separate the effects of the reaction of the free radicals with the catalytic coating.

[0055] In one embodiment, process 870 may continue to perform operation 873, which includes calculating a radical density of the plasma based on the difference in temperature readings of the first sensor and the second sensor. In particular, the radical density may be associated with a temperature difference between the first sensor and the second sensor. For example, a relatively high temperature difference corresponds to a high radical density, while a relatively low temperature difference corresponds to a low radical density.

[0056] Reference now Fig. 9 , shows a diagram of a semiconductor processing tool 950 according to one embodiment. In one embodiment, the semiconductor processing tool 950 can include a chamber 951 having a lid 953. A plasma 954 can be ignited within the chamber 951. A sensor substrate 900 (which can be, for example, similar to any sensor substrate described herein) can be provided on a chuck 952, a pedestal, or the like.

[0057] In one embodiment, the semiconductor processing tool 950 may further include a tool controller 955. The tool controller 955 may include a processor 958. The processor 958 may provide instructions to the semiconductor processing tool 950 to process the substrate. The processor 958 may access a memory 957 and an AI / ML model 956. The tool controller 955 may be coupled to the chamber 951 to control processing within the semiconductor processing tool 950, and to the sensor substrate 900.

[0058] In one embodiment, the AI / ML model 956 can obtain free radical concentration measurements (or any other type of measurement) from the sensor substrate 900 and use this information to refine one or more processing conditions within the semiconductor processing tool 950. The AI / ML model may include digital twin features of the semiconductor processing tool 950. For example, the digital twin may include a digital representation of the processing environment. The digital twin may include a model that is derived from physics-based equations of chamber structure and processing conditions (such as gas flow rates, RF frequencies, temperatures, pressures, etc.). Therefore, the data derived from the one or more sensors can be used as feedback information to more accurately model the behavior of the semiconductor processing tool 950.

[0059] In addition, the AI / ML model 956 can also be used for other processes, such as chamber quartering operations, chamber matching, and similar processes. For example, appropriate chamber quartering can be identified by monitoring changes in radical density. That is, when the chamber is properly quartered, the radical density distribution throughout the chamber may be more uniform. In addition, the AI / ML model 956 can also be used to implement chamber matching operations between chambers. For example, the processing conditions can be changed to match the radical density metric. That is, instead of simply matching processing parameters between tools (e.g., the same gas flow rate, the same frequency, the same duration, etc.), the AI / ML model 956 allows matching processing conditions between chambers, such as a specific radical density. More specifically, processing parameters refer to the settings of one or more knobs that control a process, while processing conditions refer to the actual conditions experienced in the chamber (e.g., temperature, pressure, radical density, etching rate, deposition rate, etc.).

[0060] Furthermore, while specific reference is made to matching of radical density, similar matching of other processing conditions is possible. For example, the sensor substrate may include temperature measurements, pressure measurements, deposition thickness measurements, etching measurements, and similar measurements. The AI / ML model 956 may use this information to more appropriately match processing conditions between chambers, rather than simply matching processing parameters.

[0061] Reference now Fig.10, shows a process flow diagram of a process 1080 for improving processing using a sensor substrate and an AI / ML model, according to one embodiment.

[0062] In one embodiment, process 1080 may begin with operation 1081, which includes providing an AI / ML model of a semiconductor processing tool. For example, the AI / ML model may be similar to AI / ML model 956 described in more detail above. For example, the AI / ML model may include a physics-based model that is informed and / or trained by one or more sensors provided within the semiconductor processing tool.

[0063] In one embodiment, process 1080 may continue with operation 1082, which includes obtaining in-situ sensor readings using sensors provided on a substrate inserted into a semiconductor processing tool. For example, these sensors may be suitable for determining radical density concentration, temperature, pressure, etching rate, deposition rate, or the like. In a particular embodiment, the sensor substrate may be similar to any of the sensor substrates described in more detail herein.

[0064] In one embodiment, process 1080 may continue to operation 1083, which includes training the AI / ML model using the sensor readings as input. In one embodiment, the training data may be incorporated into a physics-based model, or any other type of model employed by the AI / ML model. The improved model may enable a more accurate mapping of processing parameters to actual processing conditions within the chamber. Since the processing conditions are matched rather than just the processing parameters, performance may be increased and chamber matching may be improved.

[0065] In one embodiment, process 1080 may continue to perform operation 1084, which includes deploying the AI / ML model to change one or more processing parameters. As additional data is obtained from the sensor substrate, the AI / ML model may more accurately understand how changing one or more processing parameters will change actual processing conditions.

[0066] In some embodiments, the AI / ML model is deployed to improve processing within a single chamber. For example, the AI / ML model can determine when a chamber is appropriately seasoned. In other embodiments, the AI / ML model is deployed to improve chamber matching between different chambers.

[0067] Reference now Fig.11, a block diagram of an exemplary computer system 1100 of a processing tool is illustrated, according to one embodiment. In one embodiment, the computer system 1100 is coupled to the processing tool and controls the processing in the processing tool. The computer system 1100 can be connected (e.g., networked) with other machines in a local area network (LAN), an intranet, an extranet, or the Internet. The computer system 1100 can operate as a server or client machine in a client and server network environment, or as a peer machine in a peer-to-peer (or distributed) network environment. The computer system 1100 can be a personal computer (PC), a tablet PC, a set-top box (STB), a personal digital assistant (PDA), a cellular phone, a web appliance, a server, a network router, a switch or a bridge, or any machine capable of executing a set of instructions (in sequence or otherwise) that specify actions to be taken by the machine. Further, while a single machine is illustrated for computer system 1100, the term "machine" shall also be taken to include any collection of machines (e.g., computers) that individually or jointly execute a set (or multiple sets of instructions) to perform any one or more of the methodologies described herein.

[0068] The computer system 1100 may include a computer program product, or software 1122, having a non-transitory machine-readable medium having instructions stored thereon, which may be used to program the computer system 1100 (or other electronic device) to perform a process in accordance with an embodiment. A machine-readable medium includes any mechanism for storing or transmitting information in a form readable by a machine (e.g., a computer). For example, a machine-readable (e.g., computer-readable) medium includes a machine-readable (e.g., computer) storage medium (e.g., read-only memory ("ROM"), random access memory ("RAM"), disk storage medium, optical storage medium, flash memory device, etc.), a machine-readable (e.g., computer) transmission medium (electrical, optical, acoustic or other form of propagated signals (e.g., infrared signals, digital signals, etc.)), etc.

[0069] In one embodiment, the computer system 1100 includes a system processor 1102, a main memory 1104 (e.g., read-only memory (ROM), flash memory, dynamic random access memory (DRAM) such as synchronous dynamic random access memory (SDRAM) or Rambus DRAM (RDRAM)), etc.), a static memory 1106 (e.g., flash memory, static random access memory (SRAM), etc.), and an auxiliary memory 1118 (e.g., a data storage device) that communicate with each other via a bus 1130.

[0070] The system processor 1102 represents one or more general-purpose processing devices, such as a microsystem processor, a central processing unit, or the like. In more detail, the system processor may be a complex instruction set computing (CISC) microsystem processor, a reduced instruction set computing (RISC) microsystem processor, a very long instruction word (VLIW) microsystem processor, or a system processor that implements other instruction sets, or a system processor that implements a combination of instruction sets. The system processor 1102 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 system processor (DSP), a network system processor, or the like. The system processor 1102 is configured to execute processing logic 1126 to perform the operations described herein.

[0071] The computer system 1100 may further include a system network interface device 1108 for communicating with other devices or machines. The computer system 1100 may also include a video display unit 1110 (e.g., a liquid crystal display (LCD), a light emitting diode display (LED), or a cathode ray tube (CRT)), an alphanumeric input device 1112 (e.g., a keyboard), a cursor control device 1114 (e.g., a mouse), and a signal generating device 1116 (e.g., a speaker).

[0072] The secondary memory 1118 may include a machine-accessible storage medium 1132 (or more specifically, a computer-readable storage medium) having stored thereon one or more sets of instructions (e.g., software 1122) embodying any one or more of the methodologies or functions described herein. The software 1122 may also reside completely or at least partially within the main memory 1104 and / or within the system processor 1102 during execution of the software by the computer system 1100, the main memory 1104 and the system processor 1102 also constituting machine-readable storage media. The software 1122 may further be transmitted or received over the network 1120 via the system network interface device 1108. In one embodiment, the network interface device 1108 may operate using radio frequency coupling, optical coupling, acoustic coupling, or inductive coupling.

[0073] Although the machine-accessible storage medium 1132 is shown as a single medium in the exemplary embodiment, the term "machine-readable storage medium" should also be considered to include a single medium or multiple media (such as a centralized or distributed database and / or associated cache memory and server) storing the one or more sets of instructions. The term "machine-readable storage medium" should also be considered to include any medium capable of storing or encoding a set of instructions for execution by a machine and causing the machine to perform any one or more methodologies. Therefore, the term "machine-readable storage medium" should be considered to include (but not limited to) solid-state memory and optical and magnetic media.

[0074] In the foregoing description, specific exemplary embodiments have been described. It will be apparent that various modifications may be made thereto without departing from the scope of the following claims. The specification and drawings should therefore be viewed in an illustrative rather than a restrictive sense.

Claims

1. A sensor device, comprising: substrate; a first sensor of a first type located on the substrate, wherein a catalytic layer is provided as at least a portion of the first sensor; a second sensor of the first type, located on the substrate and adjacent to the first sensor; as well as A cover is positioned over the substrate, wherein an opening through the cover is provided over the first sensor and the second sensor. 2 . The sensor device of claim 1 , wherein the sensor of the first type is a temperature sensor. The sensor device according to claim 2 , wherein the temperature sensor is a thermocouple or a thermopile. The sensor device of claim 2 , wherein the first type of sensor utilizes a change in electrical resistance to measure temperature.

5. A sensor device according to claim 4, wherein the first type of sensor is part of a Wheatstone bridge, the Wheatstone bridge having two resistors for the first sensor, wherein the two resistors for the first sensor include the catalytic layer, and wherein two resistors are provided for the second sensor. The sensor device of claim 2 , wherein the first type of sensor is a resonator.

7. The sensor device of claim 6, wherein the resonator is a surface acoustic wave (SAW) resonator, a bulk acoustic wave (BAW) resonator, or another type of MEMS resonator.

8. The sensor device of claim 1, wherein the catalytic layer comprises platinum, copper, gold, or silver.

9. The sensor device according to claim 1, further comprising: Electronic components on the substrate to enable data storage, signal processing, and / or wired / wireless I / O.

10. The sensor device according to claim 1, further comprising: a third sensor of the first type, located on the substrate, wherein a catalytic layer is provided as at least a portion of the third sensor, wherein the catalytic layer of the third sensor is different from the catalytic layer of the first sensor; and The fourth sensor of the first type is located on the substrate and adjacent to the third sensor, wherein the third sensor and the fourth sensor are protruded from the substrate at a height higher than that of the first sensor and the second sensor.

11. A method for determining a processing parameter of a semiconductor processing recipe, the method comprising: inserting a sensor substrate having a first sensor with a catalytic coating and a second sensor without a catalytic coating into a semiconductor processing tool; igniting a plasma in the semiconductor processing tool; obtaining a temperature reading with the first sensor and the second sensor; as well as A radical density of the plasma is calculated based on a difference in the temperature readings of the first sensor and the second sensor.

12. The method of claim 11, wherein the sensor substrate further comprises a third sensor having a catalytic coating different from the catalytic coating on the first sensor, and wherein a second radical density is calculated based on a difference in the temperature readings of the second sensor and the third sensor.

13. The method of claim 11, wherein the first sensor and the second sensor comprise thermocouples, thermopiles, resistive temperature sensors, or resonator sensors.

14. The method of claim 11, wherein the catalytic coating comprises platinum, copper, gold, stainless steel, or silver.

15. A method of controlling a process parameter of a semiconductor process recipe, the method comprising: Providing artificial intelligence (AI) models and / or machine learning (ML) models for semiconductor processing tools; obtaining in-situ sensor readings with a sensor provided on a substrate inserted into the semiconductor processing tool; Using the sensor readings as input to train the AI ​​model and / or the ML model; as well as Deploy the AI ​​model and / or the ML model to change one or more processing parameters of the semiconductor processing recipe.

16. The method of claim 15, wherein the sensor comprises a temperature sensor, a pressure sensor, a deposition or etching thickness sensor, or a free radical sensor.

17. The method of claim 15, wherein the sensor comprises a thermocouple, a thermopile, a resistive temperature sensor, or a resonator sensor.

18. The method of claim 15, wherein the AI ​​model and / or the ML model is a digital twin of the semiconductor processing tool.

19. The method of claim 18, wherein the digital twin comprises a physics-based model of the semiconductor processing tool.

20. The method of claim 15, wherein the sensor readings are provided to the AI ​​model and / or the ML model using a wireless communication link.

Images