Applied Materials, Inc.

Rensselaer Polytechnic Institute

Pengyu Han

Lei Lian

Todd J Egan

Michael N Grimbergen

Upendra V Ummethala

Xi-Cheng Zhang

Michael C Kutney

Leonard Michael Tedeschi

Priyadarshi Panda

Prashant Aji

Eli Mor

Chao-Hsien Lee

Alex J Tom

John Anthony O'Malley

Yuting W Chen

Zhengping Yao

Barry Craver

Jui-Che Lin

Shauh-Teh Juang

Yan-Jhu Chen

Wallace Wang

Pin Ham Lu

Wan Hsueh Lai

Shu Yu Chen

An endpoint detection system for enhanced spectral data collection is provided. An optical bundle is coupled to a light source. The optical bundle includes an emitting optical fiber and a receiving optical fiber disposed at a pairing angle relative to the emitting optical fiber. The optical bundle is coupled to a collimator assembly that receives a light beam of incident light from the emitting optical fiber and directs spectral components of the light beam to first and second portions of a substrate surface. The collimator collects reflected spectral components produced by the spectral components directed to the substrate surface. The collimator assembly transmits the reflected spectral components to the receiving fiber, which transmits the reflected spectral components to a light detection component. A processing device coupled to the light detection component determines a reflectance of the substrate surface based on the reflected spectral components.

Endpoint detection system for enhanced spectral data collection

An optical reflectometry system, including a processing chamber, a substrate support wherein the substrate support is configured to accept a substrate, a light source configured to transmit an incident light beam, an optical fiber bundle coupled to the light source and optically coupled to a lens assembly, wherein the lens assembly optically coupled to at least a first optical fiber, and configured to transmit to, and receive from, at least one area of the substrate through the transparent window, an optical splitter disposed within the optical fiber bundle; a return fiber bundle coupled to the optical splitter, and coupled to a detection system, wherein the detection system is configured to reference a reference light beam to a reflected light beam to improve a signal-to-noise ratio, analyze a full spectrum of the reflected light beam, and determine at least one characteristic of the at least one area of the substrate.

Hyper-spectral multi-spot optical reflectometer

A cool cluster comprises one or more transfer chambers; a plurality of process chambers connected to the one or more transfer chambers; and a computing device of the tool cluster. The computing device is to receive first measurements generated by sensors of a first process chamber during or after a process is performed within the first process chamber; determine that the first process chamber is due for maintenance based on processing the first measurements using a first trained machine learning model; after maintenance has been performed on the first process chamber, receive second measurements generated by the sensors during or after a seasoning process is performed within the first process chamber; and determine that the first process chamber is ready to be brought back into service based on processing the second measurements using a second trained machine learning model.

Autonomous substrate processing system

An electronic device manufacturing system configured to receive, by a processor, input data reflecting a feature related to a manufacturing process of a substrate. The manufacturing system is further configured to train a machine-learning model based on the input data reflecting the feature. The manufacturing system is further configured to modify the machine-learning model in view of the virtual knob for the feature.

Methods and mechanisms for generating virtual knobs for model performance tuning

A method for determining processing chamber conditions using sensor data and a machine learning model is provided. The method includes receiving, by a processing device, sensor data that include chamber data indicating a state of an environment of a processing chamber processing a substrate according to a set of process parameters of a current process. The sensor data further include spectral data indicating optical emission spectra (OES) measurements of a plasma disposed within the processing chamber. The method further includes using the sensor data as input to a machine learning model and obtaining one or more outputs that indicate one or more chamber condition metrics. The method further includes determining a recovery status of a processing chamber based on the one or more chamber condition metrics. The method further includes causing a modification to a performance of the processing chamber based on the recovery status of the processing chamber.

Accelerating preventative maintenance recovery and recipe optimizing using machine-learning based algorithm

Implementations disclosed describe a collimator assembly having a collimator housing that includes an interface configured to optically couple to a process chamber that has a target surface, and a port to receive an optical fiber to deliver, to an enclosure formed by the collimator housing, a first (second) plurality of spectral components of light belonging to a first (second) range of wavelengths, and an achromatic lens located, at least partially, within the enclosure formed by the collimator housing, the achromatic lens to direct the first (second) plurality of spectral components of light onto the target surface to illuminate a first (second) region on the target surface, wherein the second region is substantially the same as the first region.

Adjustable achromatic collimator assembly for endpoint detection systems

An endpoint detection system for enhanced spectral data collection is provided. An optical bundle is coupled to a light source configured to generate incident light. The optical bundle includes two or more sets of optical fibers that each include an emitting optical fiber and a receiving optical fiber. The receiving optical fibers are disposed within the optical bundle at a pairing angle relative to a respective emitting optical fiber. The optical bundle is also coupled to a collimator assembly that includes an achromatic lens. The achromatic lens receives a first light beam of incident light from a first emitting optical fiber and directs spectral components of the first light beam to a first and second portion of a surface of a substrate. The first portion of the substrate surface is substantially the same as the second portion. The achromatic lens collects reflected spectral components that are produced by the spectral components directed to the first and second portions of the substrate surface. The achromatic lens transmits the reflected spectral components to a first receiving fiber of the optical fiber bundle, which transmits the reflected spectral components to a light detection component. A processing device coupled to the light detection component determines a reflectance of the substrate surface based on the reflected spectral components.

Methods and systems for detection of an endpoint of a substrate process are provided. A set of machine learning models are trained to provide a metrology measurement value associated with a particular type of metrology measurement for a substrate based on spectral data collected for the substrate. A respective machine learning model is selected to be applied to future spectral data collected during a future substrate process for a future substrate in view of a performance rating associated with the particular type of metrology measurement. Current spectral data is collected during a current process for a current substrate and provided as input to the respective machine learning model. An indication of a respective metrology measurement value corresponding to the current substrate is extracted from one or more outputs of the trained machine learning model. In response to a determination that the respective metrology measurement satisfies a metrology measurement criterion, an instruction including a command to terminate the current process is generated.

Substrate process endpoint detection using machine learning

A substrate processing system comprises one or more transfer chambers; a plurality of process chambers connected to the one or more transfer chambers; and a computing device connected to each of the plurality of process chambers. The computing device is to receive first measurements generated by sensors of a first process chamber during or after a process is performed within the first process chamber; determine that the first process chamber is due for maintenance based on processing the first measurements using a first trained machine learning model; after maintenance has been performed on the first process chamber, receive second measurements generated by the sensors during or after a seasoning process is performed within the first process chamber; and determine that the first process chamber is ready to be brought back into service based on processing the second measurements using a second trained machine learning model.

A method is disclosed for operating an endpoint detection system of a processing chamber having a ceiling formed therein, a substrate support located internal to the processing chamber, and a substrate resting on the substrate support. A transparent panel is located in the ceiling of the processing chamber, the panel oriented at a first acute angle relative to the substrate and the substrate support. The transparent panel receives an incident light beam from the endpoint detection system at a second acute angle relative to the panel. The transparent panel transmits the incident light beam to the substrate within the processing chamber at an angle perpendicular to the substrate and the substrate support.

Eliminating internal reflections in an interferometric endpoint detection system

A terahertz (THz) anti-reflection device, for example, a broadband tunable THz anti-reflection device, includes a silicon substrate having a plurality of recesses, each of the plurality of recesses having a plurality of cavities of decreasing dimension. The cavities may be nested polygonal cavities, for example, having a square or rectangular cross section. The recesses having the cavities may be positioned at regular periods, for example, periods ranging from 10 &#x3bc;m to 20 &#x3bc;m. The devices may be fabricated by conventional lithographic methods. Also disclosed are methods for modifying terahertz radiation and methods for fabricating anti-reflection devices.

Terahertz radiation anti-reflection devices and methods for handling terahertz radiation

A broadband anti-reflection apparatus for use with terahertz radiation includes a layer having an outer surface comprising a plurality of pyramid structures having about a 30 &#x3bc;m to about a 110 &#x3bc;m period, and wherein reflectance of the terahertz radiation is reduced compared to a layer comprising a planar outer surface. Also disclosed is a method for modifying terahertz radiation which includes receiving terahertz radiation on a device having an anti-reflection layer having an outer surface comprising a plurality of pyramid structures having about a 30 &#x3bc;m to a 110 &#x3bc;m period, and modifying the terahertz radiation passing through the device or processing the terahertz radiation in the device.

Growing community of inventors

San Jose, CA, United States of America

Pengyu Han