Kla Tencor Corporation

Kla Corporation

Kla-tenor Corp.

Stilian Pandev

Andrei V Shchegrov

Alexander Kuznetsov

Noam Sapiens

Thaddeus Gerard Dziura

Dzmitry Sanko

Jonathan M Madsen

Min-Yeong Moon

Wei Guang Lu

Kevin A Peterlinz

Liran Yerushalmi

Dimitry Sanko

Ady Levy

Shankar Krishnan

Leonid Poslavsky

Gregory Brady

Jiyou Fu

Mahendra Dubey

Amnon Manassen

Daniel Kandel

John Josef Hench

Allen Park

Mark Davis Smith

Qiang Zhao

Zhengquan Tan

Houssam Chouaib

Torsten Rudolf Kaack

Liequan Lee

Lie-Quan Lee

Dawei Hu

Ming Di

Aaron J Rosenberg

Siddharth Srivastava

Antonio Mani

Tianhan Wang

Manh Dang Nguyen

John Lesoine

Choon Hoong Hoo

Fangren Ji

Jon Madsen

Sanjay Kapasi

Vladimir Levinski

Michael E Adel

Shifang Li

Walter Dean Mieher

Roie Volkovich

Yoel Feler

David Y Wang

Nadav Gutman

Nuriel Amir

John Charles Robinson

Andrei Veldman

Mohan Mahadevan

David W Price

Pradeep Subrahmanyan

Robert J Rathert

Oreste Donzella

Alexander Buettner

Bradley Ries

Chet V Lenox

Himanshu Vajaria

Myungjun Lee

Tommaso Torelli

Barry Blasenheim

Barry Saville

Shabnam Ghadar

Mike Von Den Hoff

Pablo Rovira

Meng-Fu Shih

Justin Lach

Emanuel Saerchen

Ori Tadmor

Thomas Groos

Dzimtry Sanko

Mohamed El Kodadi

Arvind Jayaraman

Phalguna Kumar Rachinayani

Jean-Christophe Perrin

Jonathan Madsen

Mohammed El Kodadi

Noam Sapien

Kevin Pterlinz

Methods and systems for improved monitoring of tool drift and tool-to-tool matching across large fleets of measurement systems employed to measure semiconductor structures are presented herein. One or more Quality Control (QC) wafers are measured by each of a fleet of measurement systems. Values of system variables are extracted from the QC measurement data associated with each measurement system using a trained QC encoder. The extracted values of the system variables are employed to condition the corresponding measurement model employed by each measurement tool to characterize structures under measurement having unknown values of one or more parameters of interest. Accurate tool-to-tool matching across a fleet of conditioned measurement systems is achieved by extracting values of system variables from measurement data collected from the same set of QC wafers. Tool health is monitored based on changes in values of system variables extracted from measurements performed at different times.

Semiconductor measurements with robust in-line tool matching

Methods and systems for training and implementing metrology recipes while dynamically controlling the convergence trajectories of multiple performance objectives are described herein. Performance metrics are employed to regularize the optimization process employed during measurement model training, model-based regression, or both. Weighting values associated with each of the performance objectives in the loss function of the model optimization are dynamically controlled during model training. In this manner, convergence of each performance objective and the tradeoff between multiple performance objectives of the loss function is controlled to arrive at a trained measurement model in a stable, balanced manner. A trained measurement model is employed to estimate values of parameters of interest based on measurements of structures having unknown values of one or more parameters of interest. In another aspect, weighting values associated with each of the performance objectives in a model-based regression on a measurement model are dynamically controlled.

Dynamic control of machine learning based measurement recipe optimization

A die screening system may receive die-resolved metrology data for a population of dies on one or more samples from the one or more in-line metrology tools after one or more fabrication steps, where the die-resolved metrology data includes images generated using one or more measurement configurations of the one or more in-line metrology tools. In this way, the die-resolved metrology data provides many measurement channels per die, where a particular measurement channel includes data from a particular pixel of a particular image. The controller may then generate screening data for the population of dies from the die-resolved metrology data, where the screening data includes a subset of the plurality of measurement channels of the die-resolved metrology data, and screen the plurality of dies into two or more disposition classes including at least outlier dies based on variability in the screening data.

Imaging reflectometry for inline screening

Methods and systems for measurement of wafer tilt and overlay are described herein. In some embodiments, the measurements are based on the value of an asymmetry response metric and known wafer statistics. Spectral measurements are performed at two different azimuth angles, preferably separated by one hundred eighty degrees. A sub-range of wavelengths is selected with significant signal sensitivity to wafer tilt or overlay. An asymmetry response metric is determined based on a difference between the spectral signals measured at the two different azimuth angles within the selected sub-range of wavelengths. The value of the asymmetry response metric is mapped to an estimated value of wafer tilt or overlay. In some other embodiments, the measurement of wafer tilt or overlay is based on a trained measurement model. Training data may be programmed or determined based on one or more asymmetry response metrics at two different azimuth angles.

Methods and systems for measurement of tilt and overlay of a structure

A metrology module includes an estimation model that is configured to provide an estimation of independent overlay with tool induced shift on received wafers based on only one azimuth angle spectra. The estimation model can use at least one machine learning algorithm. The estimation model can be derived by the machine learning algorithm applied to calculated training data based on a first training sample set from initial metrology measurements and an additional tool induced shift training sample.

Estimating in-die overlay with tool induced shift correction

Methods and systems for determining information for a specimen are provided. One system includes an output acquisition subsystem configured to generate output for a specimen at one or more target locations on the specimen and one or more temperature sensors configured to measure one or more temperatures within the system. The system also includes a deep learning model configured for predicting error in at least one of the one or more target locations based on at least one of the one or more measured temperatures input to the deep learning model by the computer subsystem. The computer subsystem is configured for determining a corrected target location for the at least one of the one or more target locations by applying the predicted error to the at least one of the one or more target locations.

Correcting target locations for temperature in semiconductor applications

A self-calibrating overlay metrology system may receive device overlay data for a device targets on a sample from an overlay metrology tool, determine preliminary device overlay measurements for the device targets including device-scale features using an overlay recipe with the device overlay data as inputs, receive assist overlay data for one or more assist targets on the sample including device-scale features from the overlay metrology tool, where at least one of the one or more assist targets has a programmed overlay offset of a selected value along a particular measurement direction, determine self-calibrating assist overlay measurements for the one or more assist targets based on the assist overlay data, where the self-calibrating assist overlay measurements are linearly proportional to overlay on the sample, and generate corrected overlay measurements for the device targets by adjusting the preliminary device overlay measurements based on the self-calibrating assist overlay measurements.

Self-calibrating overlay metrology

A spectroscopic metrology system includes a spectroscopic metrology tool and a controller. The controller generates a model of a multilayer grating including two or more layers, the model including geometric parameters indicative of a geometry of a test layer of the multilayer grating and dispersion parameters indicative of a dispersion of the test layer. The controller further receives a spectroscopic signal of a fabricated multilayer grating corresponding to the modeled multilayer grating from the spectroscopic metrology tool. The controller further determines values of the one or more parameters of the modeled multilayer grating providing a simulated spectroscopic signal corresponding to the measured spectroscopic signal within a selected tolerance. The controller further predicts a bandgap of the test layer of the fabricated multilayer grating based on the determined values of the one or more parameters of the test layer of the fabricated structure.

Bandgap measurements of patterned film stacks using spectroscopic metrology

A method and system for measuring overlay in a semiconductor manufacturing process comprise capturing an image of a feature in an article at a predetermined manufacturing stage, deriving a quantity of an image parameter from the image and converting the quantity into an overlay measurement. The conversion is by reference to an image parameter quantity derived from a reference image of a feature at the same predetermined manufacturing stage with known overlay ('OVL'). There is also disclosed a method of determining a device inspection recipe for use by an inspection tool comprising identifying device patterns as candidate device care areas that may be sensitive to OVL, deriving an OVL response for each identified pattern, correlating the OVL response with measured OVL, and selecting some or all of the device patterns as device care areas based on the correlation.

Measurement of overlay error using device inspection system

Methods and systems for improved regularization associated with tomographically resolved image based measurements of semiconductor structures are presented herein. The regularizations described herein are based on measurement data and parameterization of a constrained voxel model that captures known process variations. The constrained voxel model is determined based on simplified geometric models, process models, or both, characterizing the structure under measurement. A constrained voxel model has dramatically fewer degrees of freedom compared to an unconstrained voxel model. The value associated with each voxel of the constrained voxel model depends on a relatively small number of independent variables. Selection of the independent variables is informed by knowledge of the structure and the underlying fabrication process. Regularization based on a constrained voxel model enables faster convergence and a more accurate reconstruction of the measured structure with less computational effort. This enables semiconductor measurements with reduced data acquisition requirements, and reduced measurement time.

Tomography based semiconductor measurements using simplified models

An overlay metrology system may receive overlay data for in-die overlay targets within various fields on a skew training sample from one or more overlay metrology tools, wherein the in-die overlay targets within the fields have a range programmed overlay offsets, wherein the fields are fabricated with a range of programmed skew offsets. The system may further generate asymmetric target signals for the in-die overlay targets using an asymmetric function providing a value of zero when physical overlay is zero and a sign indicative of a direction of physical overlay. The system may further generate corrected overlay offsets for the in-die overlay targets on the asymmetric target signals, generate self-calibrated overlay offsets for the in-die overlay targets based on the programmed overlay offsets and the corrected overlay offsets, generate a trained overlay recipe, and generate overlay measurements for in-die overlay targets on additional samples using the trained overlay recipe.

Self-calibrated overlay metrology using a skew training sample

A self-calibrating overlay metrology system may receive device overlay data from device targets on a sample, determine preliminary device overlay measurements for the device targets including device-scale features using an overlay recipe with the device overlay data as inputs, receive assist overlay data from sets of assist targets on the sample including device-scale features, where a particular set of assist targets includes one or more target pairs formed with two overlay targets having programmed overlay offsets of a selected value with opposite signs along a particular measurement direction. The system may further determine self-calibrating assist overlay measurements for the sets of assist targets based on the assist overlay data, where the self-calibrating assist overlay measurements are linearly proportional to overlay on the sample, and generate corrected overlay measurements for the device targets by adjusting the preliminary device overlay measurements based on the self-calibrating assist overlay measurements.

Methods and systems for estimating a value of a quality metric indicative of one or more performance characteristics of a semiconductor measurement are presented herein. The value of the quality metric is normalized to ensure applicability across a broad range of measurement scenarios. In some embodiments, a value of a quality metric is determined for each measurement sample during measurement inference. In some embodiments, a trained quality metric model is employed to determine the uncertainty of defect classification. In some embodiments, a trained quality metric model is employed to determine the uncertainty of estimated parameters of interest, such as geometric, dispersion, process, and electrical parameters. In some examples, a quality metric is employed as a filter to detect measurement outliers. In some other examples, a quality metric is employed as a trigger to adjust a semiconductor process.

Methods and systems for determining quality of semiconductor measurements

Methods and systems for training and implementing metrology recipes based on performance metrics employed to quantitatively characterize the measurement performance of a metrology system in a particular measurement application. Performance metrics are employed to regularize the optimization process employed during measurement model training, model-based regression, or both. For example, the known distributions associated with important measurement performance metrics such as measurement precision, wafer mean, etc., are specifically employed to regularize the optimization that drives measurement model training. In a further aspect, a trained measurement model is employed to estimate values of parameters of interest based on measurements of structures having unknown values of one or more parameters of interest. In a further aspect, trained measurement model performance is validated with test data using error budget analysis. In another aspect, a model-based regression on a measurement model is physically regularized by on one or more measurement performance metrics.

Measurement recipe optimization based on probabilistic domain knowledge and physical realization

First and second metrology data are used to train a machine-learning model to predict metrology data for a metrology target based on metrology data for a device area. The first metrology data are for a plurality of instances of a device area on semiconductor die fabricated using a fabrication process. The second metrology data are for a plurality of instances of a metrology target that contains structures distinct from structures in the device area. Using the trained machine-learning model, fourth metrology data are predicted for the metrology target based on third metrology data for an instance of the device area. Using a recipe for the metrology target, one or more parameters of the metrology target are determined based on the fourth metrology data. The fabrication process is monitored and controlled based at least in part on the one or more parameters.

Signal-domain adaptation for metrology

Growing community of inventors

Santa Clara, CA, United States of America

Stilian Pandev