Model-Based Single Parameter Measurement

Patent metadata

Field	Value
Publication number	US-2016282105-A1
Application number	US-201615076530-A
Country	US
Kind code	A1
Filing date	Mar 21, 2016
Priority date	Mar 24, 2015
Publication date	Sep 29, 2016
Grant date	—

Abstract

Methods and systems for building and using a parameter isolation model to isolate measurement signal information associated with a parameter of interest from measurement signal information associated with incidental model parameters are presented herein. The parameter isolation model is trained by mapping measurement signals associated with a first set of instances of a metrology target having known values of a plurality of incidental model parameters and known values of a parameter of interest to measurement signals associated with a second set of instances of the metrology target having nominal values of the plurality of incidental model parameters and the known values of the parameter of interest. The trained parameter isolation model receives raw measurement signals and isolates measurement signal information associated with a specific parameter of interest for model-based parameter estimation. The number of floating parameters of the measurement model is reduced, resulting in a significant reduction of computational effort.

First claim

What is claimed is:

1 . A measurement system comprising:
a metrology system configured to perform a measurement of a metrology target on a wafer and generate an amount of raw measurement signals in response to the measurement; and a computing system configured to:
receive the amount of raw measurement signals associated with the measurement of the metrology target by a metrology system;
determine a reduced set of raw measurement signals by applying a trained parameter isolation model to the amount of raw measurement signals, wherein the trained parameter isolation model is trained based on measurements of a first set of instances of the metrology target having known values of a plurality of incidental model parameters and known values of a parameter of interest and a second set of instances of the metrology target having nominal values of the plurality of incidental parameters and the known values of the parameter of interest; and
estimate a value of a parameter of interest of the metrology target based on the reduced set of raw measurement signals and a measurement model, wherein the plurality of incidental model parameters of the measurement model are fixed to the nominal values.

2 . The measurement system of claim 1 , wherein the computing system is further configured to:
receive an amount of reference measurement signals associated with the measurements of the first set of instances of the metrology target having known values of the plurality of incidental model parameters and the known values of the parameter of interest; receive an amount of reduced reference measurement signals associated with the measurements of the second set of instances of the metrology target having nominal values of the plurality of incidental model parameters and the known values of the parameter of interest; and train the parameter isolation model, wherein the training of the parameter isolation model is based on mapping the reference measurement signals to the reduced reference measurement signals.

3 . The measurement system of claim 2 , wherein the computing system is further configured to:
transform the reference measurement signals into a set of reference signal components, wherein the transforming of the reference measurement signals involves an input signal transformation model; transform the reduced reference measurement signals into a set of reduced reference signal components, wherein the transforming of the reduced reference measurement signals involves a reference signal transformation model; determine an inverse transformation model, wherein the inverse transformation model transforms the reduced reference signal components into the reduced reference measurement signals; transform the amount of raw measurement signals into a first set of signal components, wherein the transforming of the amount of raw measurement signals involves the input signal transformation model, and wherein a second set of signal components is determined by applying the trained parameter isolation model to the first set of signal components; and transform the second set of signal components into the reduced set of raw measurement signals by applying the inverse transformation model to the second set of signal components.

4 . The measurement system of claim 2 , wherein the computing system is further configured to:
generate the reference measurement signals and the reduced reference measurement signals by simulation of the measurement model.

5 . The measurement system of claim 3 , wherein the input signal transformation model, the inverse transformation model, and the reference signal transformation model are determined based on a principle components analysis.

6 . The measurement system of claim 3 , wherein the input signal transformation model, the inverse transformation model, and the reference signal transformation model are model-based transformations trained on the reference measurement signals and the reduced reference measurement signals.

7 . The measurement system of claim 1 , wherein the parameter isolation model is any of a linear model, a non-linear model, a neural network model, a polynomial model, a response surface model, and a support vector machines model.

8 . The measurement system of claim 1 , wherein the estimating of the value of the parameter of interest involves any of a model-based regression, a model-based library search, a model-based library regression, image-based analysis, and a signal response metrology model.

9 . The measurement system of claim 1 , wherein the parameter of interest is any of a lithography focus parameter, a lithography dosage parameter, a critical dimension parameter, an overlay parameter, a film thickness parameter, and a material composition parameter.

10 . A method comprising:
receiving an amount of raw measurement signals associated with a measurement of a metrology target by a measurement system; determining a reduced set of raw measurement signals by applying a trained parameter isolation model to the amount of raw measurement signals, wherein the trained parameter isolation model is trained based on measurements of a first set of instances of the metrology target having known values of a plurality of incidental model parameters and known values of a parameter of interest and a second set of instances of the metrology target having nominal values of the plurality of incidental parameters and the known values of the parameter of interest; and estimating a value of a parameter of interest of the metrology target based on the reduced set of raw measurement signals and a measurement model, wherein the plurality of incidental model parameters of the measurement model are fixed to the nominal values.

11 . The method of claim 10 , further comprising:
receiving an amount of reference measurement signals associated with the measurements of the first set of instances of the metrology target having known values of the plurality of incidental model parameters and the known values of the parameter of interest; receiving an amount of reduced reference measurement signals associated with the measurements of the second set of instances of the metrology target having nominal values of the plurality of incidental model parameters and the known values of the parameter of interest; and training the parameter isolation model, wherein the training of the parameter isolation model is based on mapping the reference measurement signals to the reduced reference measurement signals.

12 . The method of claim 11 , further comprising:
transforming the reference measurement signals into a set of reference signal components, wherein the transforming of the reference measurement signals involves an input signal transformation model; transforming the reduced reference measurement signals into a set of reduced reference signal components, wherein the transforming of the reduced reference measurement signals involves a reference signal transformation model; determining an inverse transformation model, wherein the inverse transformation model transforms the reduced reference signal components into the reduced reference measurement signals; transforming the amount of raw measurement signals into a first set of signal components, wherein the transforming of the amount of raw measurement signals involves the input signal transformation model, and wherein a second set of signal components is determined by applying the trained parameter isolation model to the first set of signal components; and transforming the second set of signal components into the reduced set of raw measurement signals by applying the inverse transformation model to the second set of

Assignees

• Kla Tencor Corp

Inventors

• Pandev Stilian Ivanov

Classifications

• H10P74/203 Primary

  Structural properties, e.g. testing or measuring thicknesses, line widths, warpage, bond strengths or physical defects · CPC title

• G01B11/24

  for measuring contours or curvatures · CPC title

• G01B2210/56

  Measuring geometric parameters of semiconductor structures, e.g. profile, critical dimensions or trench depth · CPC title

• G01B11/0616 Primary

  of coating · CPC title

• G01N21/00

  Investigating or analysing materials by the use of optical means, i.e. using sub-millimetre waves, infrared, visible or ultraviolet light (G01N3/00 - G01N19/00 take precedence) · CPC title