Asml Netherlands B.v.

Jen-Shiang Wang

Guangqing Chen

Yen-Wen Lu

Mu Feng

Eric Richard Kent

Leiwu Zheng

Yongfa Fan

Qiang Zhang

Qian Zhao

Shufeng Bai

Jan Wouter Bijlsma

Maurits Van Der Schaar

Kaustuve Bhattacharyya

Youping Zhang

Peter Hanzen Wardenier

Leonardus Henricus Marie Verstappen

Omer Abubaker Omer Adam

Ya Luo

Jay Jianhui Chen

Panagiotis Pieter Bintevinos

Jiao Liang

Yunbo Guo

Paul Anthony Tuffy

Gertjan Zwartjes

Lotte Marloes Willems

Pieter Jacob Mathias Hendrik Knelissen

Johannes Catharinus Hubertus Mulkens

Yu Long Cao

Chen Zhang

Boris Menchtchikov

Feng Chen

Martin Ebert

Te-Chih Huang

Rafael C Howell

Henricus Johannes Lambertus Megens

Lingling Pu

Alvin Jianjiang Wang

Daimian Wang

Zhichao Chen

Wei Liu

Robert Kazinczi

Mir Farrokh Shayegan Salek

Stephen Peter Morgan

Yi-Yin Chen

Yu Cao

Ning Gu

Adrianus Fransiscus Petrus Engelen

Dianwen Zhu

Chang An Wang

Matteo Alessandro Francavilla

Ziyang Ma

Stephen Morgan

Xin Guo

Chenji Zhang

Danwu Chen

A method for training a machine learning model to generate a predicted measured image, the method including obtaining (a) an input target image associated with a reference design pattern, and (b) a reference measured image associated with a specified design pattern printed on a substrate, wherein the input target image and the reference measured image are non-aligned images; and training, by a hardware computer system and using the input target image, the machine learning model to generate a predicted measured image.

Machine learning based image generation for model base alignments

A method for training a patterning process model, the patterning process model configured to predict a pattern that will be formed by a patterning process. The method involves obtaining an image data associated with a desired pattern, a measured pattern of the substrate, a first model including a first set of parameters, and a machine learning model including a second set of parameters; and iteratively determining values of the first set of parameters and the second set of parameters to train the patterning process model. An iteration involves executing, using the image data, the first model and the machine learning model to cooperatively predict a printed pattern of the substrate; and modifying the values of the first set of parameters and the second set of parameters such that a difference between the measured pattern and the predicted pattern is reduced.

Method for training or using a process model for determining a pattern in a patterning process

A method involving determining an etch bias for a pattern to be etched using an etch step of a patterning process based on an etch bias model, the etch bias model including a formula having a variable associated with a spatial property of the pattern or with an etch plasma species concentration of the etch step, and including a mathematical term including a natural exponential function to the power of a parameter that is fitted or based on an etch time of the etch step; and adjusting the patterning process based on the determined etch bias.

Etch bias characterization and method of using the same

A method including performing a simulation to evaluate a plurality of metrology targets and/or a plurality of metrology recipes used to measure a metrology target, identifying one or more metrology targets and/or metrology recipes from the evaluated plurality of metrology targets and/or metrology recipes, receiving measurement data of the one or more identified metrology targets and/or metrology recipes, and using the measurement data to tune a metrology target parameter or metrology recipe parameter.

Method and apparatus for inspection and metrology

A method for evaluating images of a printed pattern. The method includes obtaining a first averaged image of the printed pattern, where the first averaged image is generated by averaging raw images of the printed pattern. The method also includes identifying one or more features of the first averaged image. The method further includes evaluating the first averaged image, using an image quality classification model and based at least on the one or more features. The evaluating includes determining, by the image quality classification model, whether the first averaged image satisfies a metric.

Utilize machine learning in selecting high quality averaged SEM images from raw images automatically

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 patterning process modeling method includes determining, with a front end of a process model, a function associated with process physics and/or chemistry of an operation within a patterning process flow; and determining, with a back end of the process model, a predicted wafer geometry. The back end includes a volumetric representation of a target area on the wafer. The predicted wafer geometry is determined by applying the function from the front end to manipulate the volumetric representation of the wafer. The volumetric representation of the wafer may be generated using volumetric dynamic B-trees. The volumetric representation of the wafer may be manipulated using a level set method. The function associated with the process physics and/or chemistry of the operation within the patterning process flow may be a velocity/speed function. Incoming flux on a modeled surface of the wafer may be determined using ray tracing.

Method for patterning process modelling

Methods of constructing a process model for simulating a characteristic of a product of lithography from patterns produced under different processing conditions. The methods use a deviation between the variation of the simulated characteristic and the variation of the measured characteristic to adjust a parameter of the process model.

Methods of tuning process models

A method for determining measurement data of a printed pattern on a substrate. The method involves obtaining (i) images of the substrate including a printed pattern corresponding to a reference pattern, (ii) an averaged image of the images, and (iii) a composite contour based on the averaged image. Further, the composite contour is aligned with respect to a reference contour of the reference pattern and contours are extracted from the images based on both the aligned composite contour and the output of die-to-database alignment of the composite contour. Further, the method determines a plurality of pattern measurements based on the contours and the measurement data corresponding to the printed patterns based on the plurality of the pattern measurements. Further, the method determines a one or more process variations such as stochastic variation, inter-die variation, intra-die variation and/or total variation.

Method for determining stochastic variation of printed patterns

A method including: obtaining a relationship between a performance indicator of a substrate measurement recipe and a parameter of the substrate measurement recipe; deriving a range of the parameter from the relationship, wherein absolute values of the performance indicator satisfy a first condition or a magnitude of variation of the performance indicator satisfies a second condition, when the first parameter is in the range; selecting a substrate measurement recipe that has the parameter in the range; and inspecting a substrate with the selected substrate measurement recipe.

Selection of substrate measurement recipes

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

A method including computing a multi-variable cost function, the multi-variable cost function representing a metric characterizing a degree of matching between a result when measuring a metrology target structure using a substrate measurement recipe and a behavior of a pattern of a functional device, the metric being a function of a plurality of design variables including a parameter of the metrology target structure, and adjusting the design variables and computing the cost function with the adjusted design variables, until a certain termination condition is satisfied.

Substrate measurement recipe configuration to improve device matching

Methods and systems for automatically generating robust metrology targets which can accommodate a variety of lithography processes and process perturbations. Individual steps of an overall lithography process are modeled into a single process sequence to simulate the physical substrate processing. That process sequence drives the creation of a three-dimensional device geometry as a whole, rather than 'building' the device geometry element-by-element.

Process based metrology target design

Growing community of inventors

Sunnyvale, CA, United States of America

Jen-Shiang Wang