Kla Tencor Corporation

Christopher Maher

Lisheng Gao

Bjorn Brauer

Hong Xiao

Kenong Wu

Allen Park

Eliezer Rosengaus

Ellis Chang

Robert M Danen

Ajay Gupta

Santosh Bhattacharyya

Grace H Chen

Junqing Huang

Thanh Huy Ha

Michael John Van Riet

Laurent Karsenti

Khurram Zafar

Hedong Yang

Chien-Huei Adam Chen

Songnian Rong

John Raymond Jordan, Iii

Vijayakumar Ramachandran

Philip Measor

Ping Gu

Jianwei Wang

Bo Hua

Roni Miller

Liu-Ming Wu

Devashish Sharma

Adam Chien-Huei Chen

Michael Van Riet

Methods and systems for discovery of defects of interest (DOI) buried within three dimensional semiconductor structures and recipe optimization are described herein. The volume of a semiconductor wafer subject to defect discovery and verification is reduced by storing images associated with a subset of the total depth of the semiconductor structures under measurement. Image patches associated with defect locations at one or more focus planes or focus ranges are recorded. The number of optical modes under consideration is reduced based on any of a comparison of one or more measured wafer level defect signatures and one or more expected wafer level defect signatures, measured defect signal to noise ratio, and defects verified without de-processing. Furthermore, verified defects and recorded images are employed to train a nuisance filter and optimize the measurement recipe. The trained nuisance filter is applied to defect images to select the optimal optical mode for production.

Defect discovery and recipe optimization for inspection of three-dimensional semiconductor structures

A defect detection method includes acquiring a reference image; selecting a target region of the reference image; identifying, based on a matching metric, one or more comparative regions of the reference image corresponding to the target region; acquiring a test image; masking the test image with the target region of the reference image and the one or more comparative regions of the reference image; defining a defect threshold for the target region in the test image based on the one or more comparative regions in the test image; and determining whether the target region of the test image contains a defect based on the defect threshold.

Systems and methods for region-adaptive defect detection

An inspection tool includes a controller that is configured to generate a scan pattern for an electron beam to image areas of interest on the wafer. The scan pattern minimizes dwell time of the electron beam on the surface of the wafer between the areas of interest. At least one stage speed and at least one raster pattern can be selected based on the areas of interest. The controller sends instructions to electron beam optics to direct the electron beam at the areas of interest on the surface of the wafer using the scan pattern.

High-speed hotspot or defect imaging with a charged particle beam system

One embodiment relates to a method of inspecting an array of cells on a substrate. A reference image is generated using a cell image that was previously determined to be defect free. A reference contour image which includes contours of the reference image is also generated. The reference contour image is used to detect defects in the array of cells on the substrate. Another embodiment relates to a system for detecting defects in an array on a substrate. Other embodiments, aspects and features are also disclosed.

Contour-based array inspection of patterned defects

The present invention includes searching imagery data in order to identify one or more patterned regions on a semiconductor wafer, generating one or more virtual Fourier filter (VFF) working areas, acquiring an initial set of imagery data from the VFF working areas, defining VFF training blocks within the identified patterned regions of the VFF working areas utilizing the initial set of imagery data, wherein each VFF training block is defined to encompass a portion of the identified patterned region displaying a selected repeating pattern, calculating an initial spectrum for each VFF training block utilizing the initial set of imagery data from the VFF training blocks, and generating a VFF for each training block by identifying frequencies of the initial spectrum having maxima in the frequency domain, wherein the VFF is configured to null the magnitude of the initial spectrum at the frequencies identified to display spectral maxima.

Region based virtual fourier filter

Growing community of inventors

Campbell, CA, United States of America

Christopher Maher