Asml Netherlands B.v.

Mark John Maslow

Wim Tjibbo Tel

Frank Staals

Marinus Jochemsen

Abraham Slachter

Roy Anunciado

Paul Christiaan Hinnen

Patrick Warnaar

Simon Hendrik Celine Van Gorp

Koenraad Van Ingen Schenau

Scott Anderson Middlebrooks

Antoine Gaston Marie Kiers

Adrianus Cornelis Matheus Koopman

Johannes Catharinus Hubertus Mulkens

Hugo Augustinus Joseph Cramer

Thomas Theeuwes

Maxim Pisarenco

Daan Maurits Slotboom

Hermanus Adrianus Dillen

Marc Jurian Kea

Peter David Engblom

Marleen Kooiman

Ruben Cornelis Maas

Ralph Timotheus Huijgen

Koen Thuijs

Bernardo Andres Oyarzun Rivera

Described herein is a method of training a model configured to predict whether a feature associated with an imaged substrate will be defective after etching of the imaged substrate and determining etch conditions based on the trained model. The method includes obtaining, via a metrology tool, (i) an after development image of the imaged substrate at a given location, the after development image including a plurality of features, and (ii) an after etch image of the imaged substrate at the given location; and training, using the after development image and the after etch image, the model configured to determine defectiveness of a given feature of the plurality of features in the after development image. In an embodiment, the determining of defectiveness is based on comparing the given feature in the after development image with a corresponding etch feature in the after etch image.

Method for determining defectiveness of pattern based on after development image

A method including: obtaining an image of at least part of a substrate, wherein the image includes at least one feature manufactured on the substrate by a manufacturing process including a lithographic process and one or more further processes; determining one or more image-related metrics in dependence on a contour determined from the image, wherein one of the one or more image-related metrics is an edge placement error, EPE, of the at least one feature; and determining one or more control parameters of the lithographic process and/or the one or more further processes in dependence on the edge placement error, wherein at least one control parameter is determined so as to minimize the edge placement error of the at least one feature.

Method of determining control parameters of a device manufacturing process

A method of determining a characteristic of one or more processes for manufacturing features on a substrate, the method including: obtaining image data of a plurality of features on a least part of at least one region on a substrate; using the image data to obtain measured data of one or more dimensions of each of at least some of the plurality of features; determining a statistical parameter that is dependent on the variation of the measured data of one or more dimensions of each of at least some of the plurality of features; determining a probability of defective manufacture of features in dependence on a determined number of defective features in the image data; and determining the characteristic of the one or more processes to have the probability of defective manufacture of features and the statistical parameter.

Method for characterizing a manufacturing process of semiconductor devices

A method for training a deep learning model of a patterning process. The method includes obtaining (i) training data comprising an input image of at least a part of a substrate having a plurality of features and a truth image, (ii) a set of classes, each class corresponding to a feature of the plurality of features of the substrate within the input image, and (iii) a deep learning model configured to receive the training data and the set of classes, generating a predicted image, by modeling and/or simulation of the deep learning model using the input image, assigning a class of the set of classes to a feature within the predicted image based on matching of the feature with a corresponding feature within the truth image, and generating, by modeling and/or simulation, a trained deep learning model by iteratively assigning weights using a loss function.

Deep learning for semantic segmentation of pattern

A method including: obtaining an image of at least part of a substrate, wherein the image includes at least one feature of a device being manufactured in a layer on the substrate; obtaining a layout of features associated with a previous layer adjacent to the layer on the substrate; calculating one or more image-related metrics in dependence on: 1) a contour determined from the image including the at least one feature and 2) the layout; and determining one or more control parameters of a lithographic apparatus and/or one or more further processes in a manufacturing process of the device in dependence on the one or more image-related metrics, wherein at least one of the control parameters is determined to modify the geometry of the contour in order to improve the one or more image-related metrics.

A method involving determining a contribution that one or more process apparatuses make to a characteristic of a substrate after the substrate has been processed according to a patterning process by the one or more process apparatuses by removing from values of the characteristic of the substrate a contribution of a lithography apparatus to the characteristic and a contribution of one or more pre-lithography process apparatuses to the characteristic.

Method and system to monitor a process apparatus

A method including: computing a value of a first variable of a pattern of, or for, a substrate processed by a patterning process by combining a fingerprint of the first variable on the substrate and a certain value of the first variable; and determining a value of a second variable of the pattern based at least in part on the computed value of the first variable.

Separation of contributions to metrology data

A method for determining a metric of a feature on a substrate obtained by a semiconductor manufacturing process involving a lithographic process, the method including: obtaining an image of at least part of the substrate, wherein the image includes at least the feature; determining a contour of the feature from the image; determining a plurality of segments of the contour; determining respective weights for each of the plurality of segments; determining, for each of the segments, an image-related metric; and determining the metric of the feature in dependence on the weights and the calculated image-related metric of each of the segments.

A method for training a deep learning model of a patterning process. The method includes obtaining (i) training data including an input image of at least a part of a substrate having a plurality of features and including a truth image, (ii) a set of classes, each class corresponding to a feature of the plurality of features of the substrate within the input image, and (iii) a deep learning model configured to receive the training data and the set of classes, generating a predicted image, by modeling and/or simulation with the deep learning model using the input image, assigning a class of the set of classes to a feature within the predicted image based on matching of the feature with a corresponding feature within the truth image, and generating, by modeling and/or simulation, a trained deep learning model by iteratively assigning weights using a loss function.

A method for adjusting a lithography process, wherein processing parameters of the lithography process include a first group of processing parameters and a second group of processing parameters, the method including: obtaining a change of the second group of processing parameters; determining a change of a sub-process window (sub-PW) as a result of the change of the second group of processing parameters, wherein the sub-PW is spanned by only the first group of processing parameters; and adjusting the first group of processing parameters based on the change of the sub-PW.

Growing community of inventors

Eindhoven, Netherlands

Mark John Maslow