Asml Netherlands B.v.

Brion Technologies, Inc.

Onetta, Inc.

Kla-tencor Technologies Corporation

Other

Kla Tencor Corporation

Yen-Wen Lu

Yu Long Cao

Jun Ye

Been-Der Chen

Luoqi Chen

Xun Chen

Jing Su

Rafael C Howell

Jen-Shiang Wang

Jiangwei Li

Peng Liu

Yi Zou

William S Wong

Youping Zhang

Stefan Hunsche

James Norman Wiley

Marinus Jochemsen

Fei Liu

Eric Richard Kent

David Alles

Lin Lee Cheong

Chenxi Lin

Guangqing Chen

Quan Zhang

Hong M Chen

Boris Menchtchikov

Feng Chen

Shauh-Teh Juang

Hua-Yu Liu

Te-Chih Huang

Yongfa Fan

Lih-Huah Yiin

Xi Chen

Min-Chun Tsai

Ya Luo

Roshni Biswas

Joseph Werner De Vocht

Wanyu Li

Frank Gang Chen

Yuelin Du

Jun Tao

Tatsuo Nishibe

Shufeng Bai

Jan Wouter Bijlsma

Paul Anthony Tuffy

Gertjan Zwartjes

Dong Mao

Taihui Liu

Mu Feng

Jay Jianhui Chen

Qian Zhao

Yang Lin

Qiang Zhang

George Q Chen

Yu Zhao

Wei Liu

Leiwu Zheng

Zhi-Pan Li

Yunbo Guo

Xiaorui Chen

Jin Cheng

Manman Ren

Dianwen Zhu

David S Alles

Yu Cao

Ziyang Ma

Methods of identifying a hot spot from a design layout or of predicting whether a pattern in a design layout is defective, using a machine learning model. An example method disclosed herein includes obtaining sets of one or more characteristics of performance of hot spots, respectively, under a plurality of process conditions, respectively, in a device manufacturing process; determining, for each of the process conditions, for each of the hot spots, based on the one or more characteristics under that process condition, whether that hot spot is defective; obtaining a characteristic of each of the process conditions; obtaining a characteristic of each of the hot spots; and training a machine learning model using a training set including the characteristic of one of the process conditions, the characteristic of one of the hot spots, and whether that hot spot is defective under that process condition.

Identification of hot spots or defects by machine learning

A method including: obtaining data based an optical proximity correction for a spatially shifted version of a training design pattern; and training a machine learning model configured to predict optical proximity corrections for design patterns using data regarding the training design pattern and the data based on the optical proximity correction for the spatially shifted version of the training design pattern.

Training methods for machine learning assisted optical proximity error

A method for improving a process model for a patterning process, the method including obtaining a) a measured contour from an image capture device, and b) a simulated contour generated from a simulation of the process model. The method also includes aligning the measured contour with the simulated contour by determining an offset between the measured contour and the simulated contour. The process model is calibrated to reduce a difference, computed based on the determined offset, between the simulated contour and the measured contour.

Method for improving a process for a patterning process

A method including: obtaining a thin-mask transmission function of a patterning device and a M3D model for a lithographic process, wherein the thin-mask transmission function is a continuous transmission mask (CTM) and the M3D model at least represents a portion of M3D attributable to multiple edges of structures on the patterning device; determining a M3D mask transmission function of the patterning device by using the thin-mask transmission function and the M3D model; and determining an aerial image produced by the patterning device and the lithographic process, by using the M3D mask transmission function.

Methods of determining scattering of radiation by structures of finite thicknesses on a patterning device

Training methods for machine learning assisted optical proximity error correction

A method to determine a curvilinear pattern of a patterning device that includes obtaining (i) an initial image of the patterning device corresponding to a target pattern to be printed on a substrate subjected to a patterning process, and (ii) a process model configured to predict a pattern on the substrate from the initial image, generating, by a hardware computer system, an enhanced image from the initial image, generating, by the hardware computer system, a level set image using the enhanced image, and iteratively determining, by the hardware computer system, a curvilinear pattern for the patterning device based on the level set image, the process model, and a cost function, where the cost function (e.g., EPE) determines a difference between a predicted pattern and the target pattern, where the difference is iteratively reduced.

Method for determining curvilinear patterns for patterning device

A method for determining an overlapping process window (OPW) of an area of interest on a portion of a design layout for a device manufacturing process for imaging the portion onto a substrate, the method including: obtaining a plurality of features in the area of interest; obtaining a plurality of values of one or more processing parameters of the device manufacturing process; determining existence of defects, probability of the existence of defects, or both in imaging the plurality of features by the device manufacturing process under each of the plurality of values; and determining the OPW of the area of interest from the existence of defects, the probability of the existence of defects, or both.

Process window identifier

A method of controlling a computer process for designing or verifying a photolithographic component includes building a source tree including nodes of the process, including dependency relationships among the nodes, defining, for some nodes, at least two different process conditions, expanding the source tree to form an expanded tree, including generating a separate node for each different defined process condition, and duplicating dependent nodes having an input relationship to each generated separate node, determining respective computing hardware requirements for processing the node, selecting computer hardware constraints based on capabilities of the host computing system, determining, based on the requirements and constraints and on dependency relations in the expanded tree, an execution sequence for the computer process, and performing the computer process on the computing system.

Method and apparatus for controlling a computing process

A method including: obtaining a thin-mask transmission function of a patterning device and a M3D model for a lithographic process, wherein the thin-mask transmission function represents a continuous transmission mask and the M3D model at least represents a portion of M3D attributable to multiple edges of structures on the patterning device; determining a M3D mask transmission function of the patterning device by using the thin-mask transmission function and the M3D model; and determining an aerial image produced by the patterning device and the lithographic process, by using the M3D mask transmission function.

A method including: obtaining a characteristic of a portion of a design layout; determining a characteristic of M3D of a patterning device including or forming the portion; and training, by a computer, a neural network using training data including a sample whose feature vector includes the characteristic of the portion and whose supervisory signal includes the characteristic of the M3D. Also disclosed is a method including: obtaining a characteristic of a portion of a design layout; obtaining a characteristic of a lithographic process that uses a patterning device including or forming the portion; determining a characteristic of a result of the lithographic process; training, by a computer, a neural network using training data including a sample whose feature vector includes the characteristic of the portion and the characteristic of the lithographic process, and whose supervisory signal includes the characteristic of the result.

Methods of determining, and using, a patterning process model that is a machine learning model. The process model is trained partially based on simulation or based on a non-machine learning model. The training data may include inputs obtained from a design layout, patterning process measurements, and image measurements.

Methods of determining process models by machine learning

Growing community of inventors

Los Altos, CA, United States of America

Yen-Wen Lu