Implementation of Large Scale Deep Learning Non-Destructive Methods for Characterizing 4H-SiC Materials
A whole wafer method for industrial high volume, non-destructive characterizing of extended defects is demonstrated for 150 mm and 200 mm 4H-SiC wafers. Deep learning (DL) coupled with non-destructive techniques (NDT, DL-NDT) involving high volume, fast optical microscopy methods correlates industry...
Gespeichert in:
Veröffentlicht in: | Diffusion and defect data. Solid state data. Pt. A, Defect and diffusion forum Defect and diffusion forum, 2023-06, Vol.426, p.3-9 |
---|---|
Hauptverfasser: | , , , , |
Format: | Artikel |
Sprache: | eng |
Schlagworte: | |
Online-Zugang: | Volltext |
Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
container_end_page | 9 |
---|---|
container_issue | |
container_start_page | 3 |
container_title | Diffusion and defect data. Solid state data. Pt. A, Defect and diffusion forum |
container_volume | 426 |
creator | Witry, Jason Balkas, Elif Conrad, Matthew van Brunt, Edward Leonard, Robert |
description | A whole wafer method for industrial high volume, non-destructive characterizing of extended defects is demonstrated for 150 mm and 200 mm 4H-SiC wafers. Deep learning (DL) coupled with non-destructive techniques (NDT, DL-NDT) involving high volume, fast optical microscopy methods correlates industry accepted chemistry and physics-based etch and diffraction techniques for defect characterization. The application of the DL-NDT method is shown to reproduce defect distributions achieved by accepted etch techniques for extended defects of threading dislocations (TD), basal plane dislocations (BPD), and threading screw dislocations (TSD). An example of algorithm development is described to show progress toward implementing the method, as well as DL-NDT defect density compared to etch density for multiple wafers. The development status for implementing this technique for large-scale industrial wafer production includes etch validation of the results to ensure the technique is consistent and reliable. The ability to use this non-destructive technique ultimately will result in better correlation with device behavior and provide feedback to crystal growth processes to improve substrate wafers, while reducing the need for etch methods. |
doi_str_mv | 10.4028/p-08c7e9 |
format | Article |
fullrecord | <record><control><sourceid>proquest_cross</sourceid><recordid>TN_cdi_proquest_journals_3091546711</recordid><sourceformat>XML</sourceformat><sourcesystem>PC</sourcesystem><sourcerecordid>3091546711</sourcerecordid><originalsourceid>FETCH-LOGICAL-c2009-db7f7a8e005529fffdf1828bdeea4c6eb75dcc66af3cea7148ac2e6f6e60edaf3</originalsourceid><addsrcrecordid>eNplkF9LwzAUxYMoOKfgRwj4IkI06Z80fZRuukGnD9PncpfebB1bU5NM0E9vRwUffDqHw49zL4eQa8HvEx6ph45xpTPMT8hISBmxPOXZae-5iBiPlTwnF95vOY-FEsmIrOf7bod7bAOExrbUGlqCWyNdatghnSB2tERwbdOu6Ytt2QR9cAcdmk-kCwwbW3tqrKPFBhzogK75PqLJjC2bgi7gmMDOX5Iz0wte_eqYvD9N34oZK1-f58VjyXTEec7qVWYyUMh5mka5MaY2QkVqVSNCoiWusrTWWkowsUbIRKJARyiNRMmx7tMxuRl6O2c_Dv2v1dYeXNufrGKeizSRmRA9dTtQ2lnvHZqqc80e3FcleHWcseqqYcYevRvQ4KD1AfXmr_Ef_AOAAnU6</addsrcrecordid><sourcetype>Aggregation Database</sourcetype><iscdi>true</iscdi><recordtype>article</recordtype><pqid>3091546711</pqid></control><display><type>article</type><title>Implementation of Large Scale Deep Learning Non-Destructive Methods for Characterizing 4H-SiC Materials</title><source>Scientific.net Journals</source><creator>Witry, Jason ; Balkas, Elif ; Conrad, Matthew ; van Brunt, Edward ; Leonard, Robert</creator><creatorcontrib>Witry, Jason ; Balkas, Elif ; Conrad, Matthew ; van Brunt, Edward ; Leonard, Robert</creatorcontrib><description>A whole wafer method for industrial high volume, non-destructive characterizing of extended defects is demonstrated for 150 mm and 200 mm 4H-SiC wafers. Deep learning (DL) coupled with non-destructive techniques (NDT, DL-NDT) involving high volume, fast optical microscopy methods correlates industry accepted chemistry and physics-based etch and diffraction techniques for defect characterization. The application of the DL-NDT method is shown to reproduce defect distributions achieved by accepted etch techniques for extended defects of threading dislocations (TD), basal plane dislocations (BPD), and threading screw dislocations (TSD). An example of algorithm development is described to show progress toward implementing the method, as well as DL-NDT defect density compared to etch density for multiple wafers. The development status for implementing this technique for large-scale industrial wafer production includes etch validation of the results to ensure the technique is consistent and reliable. The ability to use this non-destructive technique ultimately will result in better correlation with device behavior and provide feedback to crystal growth processes to improve substrate wafers, while reducing the need for etch methods.</description><identifier>ISSN: 1012-0386</identifier><identifier>ISSN: 1662-9507</identifier><identifier>EISSN: 1662-9507</identifier><identifier>DOI: 10.4028/p-08c7e9</identifier><language>eng</language><publisher>Zurich: Trans Tech Publications Ltd</publisher><subject>Algorithms ; Basal plane ; Crystal defects ; Crystal growth ; Deep learning ; Density ; Nondestructive testing ; Optical microscopy ; Screw dislocations ; Silicon carbide ; Substrates ; Threading dislocations ; Wafers</subject><ispartof>Diffusion and defect data. Solid state data. Pt. A, Defect and diffusion forum, 2023-06, Vol.426, p.3-9</ispartof><rights>2023 Leonard et al.</rights><rights>Copyright Trans Tech Publications Ltd. 2023</rights><lds50>peer_reviewed</lds50><oa>free_for_read</oa><woscitedreferencessubscribed>false</woscitedreferencessubscribed><cites>FETCH-LOGICAL-c2009-db7f7a8e005529fffdf1828bdeea4c6eb75dcc66af3cea7148ac2e6f6e60edaf3</cites></display><links><openurl>$$Topenurl_article</openurl><openurlfulltext>$$Topenurlfull_article</openurlfulltext><thumbnail>$$Uhttps://www.scientific.net/Image/TitleCover/6844?width=600</thumbnail><link.rule.ids>314,780,784,27924,27925</link.rule.ids></links><search><creatorcontrib>Witry, Jason</creatorcontrib><creatorcontrib>Balkas, Elif</creatorcontrib><creatorcontrib>Conrad, Matthew</creatorcontrib><creatorcontrib>van Brunt, Edward</creatorcontrib><creatorcontrib>Leonard, Robert</creatorcontrib><title>Implementation of Large Scale Deep Learning Non-Destructive Methods for Characterizing 4H-SiC Materials</title><title>Diffusion and defect data. Solid state data. Pt. A, Defect and diffusion forum</title><description>A whole wafer method for industrial high volume, non-destructive characterizing of extended defects is demonstrated for 150 mm and 200 mm 4H-SiC wafers. Deep learning (DL) coupled with non-destructive techniques (NDT, DL-NDT) involving high volume, fast optical microscopy methods correlates industry accepted chemistry and physics-based etch and diffraction techniques for defect characterization. The application of the DL-NDT method is shown to reproduce defect distributions achieved by accepted etch techniques for extended defects of threading dislocations (TD), basal plane dislocations (BPD), and threading screw dislocations (TSD). An example of algorithm development is described to show progress toward implementing the method, as well as DL-NDT defect density compared to etch density for multiple wafers. The development status for implementing this technique for large-scale industrial wafer production includes etch validation of the results to ensure the technique is consistent and reliable. The ability to use this non-destructive technique ultimately will result in better correlation with device behavior and provide feedback to crystal growth processes to improve substrate wafers, while reducing the need for etch methods.</description><subject>Algorithms</subject><subject>Basal plane</subject><subject>Crystal defects</subject><subject>Crystal growth</subject><subject>Deep learning</subject><subject>Density</subject><subject>Nondestructive testing</subject><subject>Optical microscopy</subject><subject>Screw dislocations</subject><subject>Silicon carbide</subject><subject>Substrates</subject><subject>Threading dislocations</subject><subject>Wafers</subject><issn>1012-0386</issn><issn>1662-9507</issn><issn>1662-9507</issn><fulltext>true</fulltext><rsrctype>article</rsrctype><creationdate>2023</creationdate><recordtype>article</recordtype><recordid>eNplkF9LwzAUxYMoOKfgRwj4IkI06Z80fZRuukGnD9PncpfebB1bU5NM0E9vRwUffDqHw49zL4eQa8HvEx6ph45xpTPMT8hISBmxPOXZae-5iBiPlTwnF95vOY-FEsmIrOf7bod7bAOExrbUGlqCWyNdatghnSB2tERwbdOu6Ytt2QR9cAcdmk-kCwwbW3tqrKPFBhzogK75PqLJjC2bgi7gmMDOX5Iz0wte_eqYvD9N34oZK1-f58VjyXTEec7qVWYyUMh5mka5MaY2QkVqVSNCoiWusrTWWkowsUbIRKJARyiNRMmx7tMxuRl6O2c_Dv2v1dYeXNufrGKeizSRmRA9dTtQ2lnvHZqqc80e3FcleHWcseqqYcYevRvQ4KD1AfXmr_Ef_AOAAnU6</recordid><startdate>20230606</startdate><enddate>20230606</enddate><creator>Witry, Jason</creator><creator>Balkas, Elif</creator><creator>Conrad, Matthew</creator><creator>van Brunt, Edward</creator><creator>Leonard, Robert</creator><general>Trans Tech Publications Ltd</general><scope>AAYXX</scope><scope>CITATION</scope></search><sort><creationdate>20230606</creationdate><title>Implementation of Large Scale Deep Learning Non-Destructive Methods for Characterizing 4H-SiC Materials</title><author>Witry, Jason ; Balkas, Elif ; Conrad, Matthew ; van Brunt, Edward ; Leonard, Robert</author></sort><facets><frbrtype>5</frbrtype><frbrgroupid>cdi_FETCH-LOGICAL-c2009-db7f7a8e005529fffdf1828bdeea4c6eb75dcc66af3cea7148ac2e6f6e60edaf3</frbrgroupid><rsrctype>articles</rsrctype><prefilter>articles</prefilter><language>eng</language><creationdate>2023</creationdate><topic>Algorithms</topic><topic>Basal plane</topic><topic>Crystal defects</topic><topic>Crystal growth</topic><topic>Deep learning</topic><topic>Density</topic><topic>Nondestructive testing</topic><topic>Optical microscopy</topic><topic>Screw dislocations</topic><topic>Silicon carbide</topic><topic>Substrates</topic><topic>Threading dislocations</topic><topic>Wafers</topic><toplevel>peer_reviewed</toplevel><toplevel>online_resources</toplevel><creatorcontrib>Witry, Jason</creatorcontrib><creatorcontrib>Balkas, Elif</creatorcontrib><creatorcontrib>Conrad, Matthew</creatorcontrib><creatorcontrib>van Brunt, Edward</creatorcontrib><creatorcontrib>Leonard, Robert</creatorcontrib><collection>CrossRef</collection><jtitle>Diffusion and defect data. Solid state data. Pt. A, Defect and diffusion forum</jtitle></facets><delivery><delcategory>Remote Search Resource</delcategory><fulltext>fulltext</fulltext></delivery><addata><au>Witry, Jason</au><au>Balkas, Elif</au><au>Conrad, Matthew</au><au>van Brunt, Edward</au><au>Leonard, Robert</au><format>journal</format><genre>article</genre><ristype>JOUR</ristype><atitle>Implementation of Large Scale Deep Learning Non-Destructive Methods for Characterizing 4H-SiC Materials</atitle><jtitle>Diffusion and defect data. Solid state data. Pt. A, Defect and diffusion forum</jtitle><date>2023-06-06</date><risdate>2023</risdate><volume>426</volume><spage>3</spage><epage>9</epage><pages>3-9</pages><issn>1012-0386</issn><issn>1662-9507</issn><eissn>1662-9507</eissn><abstract>A whole wafer method for industrial high volume, non-destructive characterizing of extended defects is demonstrated for 150 mm and 200 mm 4H-SiC wafers. Deep learning (DL) coupled with non-destructive techniques (NDT, DL-NDT) involving high volume, fast optical microscopy methods correlates industry accepted chemistry and physics-based etch and diffraction techniques for defect characterization. The application of the DL-NDT method is shown to reproduce defect distributions achieved by accepted etch techniques for extended defects of threading dislocations (TD), basal plane dislocations (BPD), and threading screw dislocations (TSD). An example of algorithm development is described to show progress toward implementing the method, as well as DL-NDT defect density compared to etch density for multiple wafers. The development status for implementing this technique for large-scale industrial wafer production includes etch validation of the results to ensure the technique is consistent and reliable. The ability to use this non-destructive technique ultimately will result in better correlation with device behavior and provide feedback to crystal growth processes to improve substrate wafers, while reducing the need for etch methods.</abstract><cop>Zurich</cop><pub>Trans Tech Publications Ltd</pub><doi>10.4028/p-08c7e9</doi><tpages>7</tpages><oa>free_for_read</oa></addata></record> |
fulltext | fulltext |
identifier | ISSN: 1012-0386 |
ispartof | Diffusion and defect data. Solid state data. Pt. A, Defect and diffusion forum, 2023-06, Vol.426, p.3-9 |
issn | 1012-0386 1662-9507 1662-9507 |
language | eng |
recordid | cdi_proquest_journals_3091546711 |
source | Scientific.net Journals |
subjects | Algorithms Basal plane Crystal defects Crystal growth Deep learning Density Nondestructive testing Optical microscopy Screw dislocations Silicon carbide Substrates Threading dislocations Wafers |
title | Implementation of Large Scale Deep Learning Non-Destructive Methods for Characterizing 4H-SiC Materials |
url | https://sfx.bib-bvb.de/sfx_tum?ctx_ver=Z39.88-2004&ctx_enc=info:ofi/enc:UTF-8&ctx_tim=2025-01-03T02%3A20%3A58IST&url_ver=Z39.88-2004&url_ctx_fmt=infofi/fmt:kev:mtx:ctx&rfr_id=info:sid/primo.exlibrisgroup.com:primo3-Article-proquest_cross&rft_val_fmt=info:ofi/fmt:kev:mtx:journal&rft.genre=article&rft.atitle=Implementation%20of%20Large%20Scale%20Deep%20Learning%20Non-Destructive%20Methods%20for%20Characterizing%204H-SiC%20Materials&rft.jtitle=Diffusion%20and%20defect%20data.%20Solid%20state%20data.%20Pt.%20A,%20Defect%20and%20diffusion%20forum&rft.au=Witry,%20Jason&rft.date=2023-06-06&rft.volume=426&rft.spage=3&rft.epage=9&rft.pages=3-9&rft.issn=1012-0386&rft.eissn=1662-9507&rft_id=info:doi/10.4028/p-08c7e9&rft_dat=%3Cproquest_cross%3E3091546711%3C/proquest_cross%3E%3Curl%3E%3C/url%3E&disable_directlink=true&sfx.directlink=off&sfx.report_link=0&rft_id=info:oai/&rft_pqid=3091546711&rft_id=info:pmid/&rfr_iscdi=true |