[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"doc-detail-127202-en":3,"doc-seo-127202-105":30,"detail-sidebar-cat-0-en-105":92},{"code":4,"msg":5,"data":6},0,"success",{"doc_id":7,"user_id":8,"nickname":9,"user_avatar":10,"doc_module":4,"category_id":11,"category_name":12,"doc_title":13,"doc_description":14,"doc_content":15,"file_id":16,"file_url":17,"file_type":18,"file_size":19,"view_count":20,"is_deleted":4,"is_public":20,"is_downloadable":20,"audit_status":20,"page_count":21,"language":22,"language_code":23,"site_id":24,"html_lang":23,"table_of_contents":25,"faqs":26,"seo_title":27,"seo_description":14,"update_tm":28,"read_time":29},127202,549768702563,"Sage","https://ap-avatar.wpscdn.com/avatar/8000c4aa63b76e948b?x-image-process=image/resize,m_fixed,w_180,h_180&k=1786536092046926083",8,"Research & Report","Machine Learning Methods for Structure Loss Classification in Czochralski Silicon Ingots","A considerable fraction of Czochralski silicon ingots undergoes remelting due to dislocations generated during growth, commonly called structure loss. Identifying and categorizing failed ingots is critical to understanding root causes and improving production yield. The study applies machine learning to classify monocrystalline silicon ingots with structural loss from the Czochralski process using surface images and three CNN-based pipelines, comparing their accuracy and training stability. The best performance uses a pretrained model with incremental training.","This article is licensed under CC-BY 4.0   \n[pubs.acs.org/crystal](pubs.acs.org/crystal)  Article   \nMachine Learning Methods for Structure Loss Classification in Czochralski Silicon Ingots  \nAlfredo Sanchez Garcia, * Rania Hendawi, and Marisa Di Sabatino  \n Cite This: Cryst. Growth Des. 2024, 24, 7132−7140  \nRead Online  \nACCESS  \n Metrics & More  \n Article Recommendations  \n*sı   \nSupporting Information  \nDownloaded via SINTEF on November 22, 2024 at 12:31:37 (UTC) . See [https://pubs.acs.org/sharingguidelines](https://pubs.acs.org/sharingguidelines) for options on how to legitimately share published articles.  \nABSTRACT: A considerable fraction of Czochralski silicon ingots undergoes remelting mainly because of dislocation generated during the growth, commonly termed structure loss. Being able to identify and categorize these failed ingots is a key step for understanding the root causes of structure loss and achieving a higher production yield. This work reports the utilization of machine learning (ML) to classify monocrystalline silicon ingots  \nthat have experienced structural loss during the Czochralski process. Three ML pipelines are implemented using different convolutional neural network architectures to analyze the surface images of the ingots. The accuracy and stability of the three ML pipelines are assessed. The results indicate that the pipeline that combines a pretrained model with an incremental training strategy obtains the highest accuracies and stable trainings of all tested pipelines, thereby making it the most suitable classification of structure loss in Czochralski-grown ingots.  \n1. INTRODUCTION  \nThe solar energy sector continues to be predominantly led by silicon-based PV, commanding a 95% market share. Notably, over 85% of silicon wafers have a monocrystalline structure due to its higher efficiency and fewer defects compared to multicrystalline silicon wafers.1 Industrially, monocrystalline silicon wafers are cut from single-crystal silicon ingots grown using the Czochralski method,2 which has seen significant advances in recent years. However, the growing demand for silicon has led to an ongoing need for larger ingot diametersand crucible sizes, posing challenges in the Czochralski silicon growth process.1  \nCzochralski (Cz) silicon ingots have a dislocation-free structure because of Dash’s procedure, which aims at eliminating the dislocations resulting from the thermal stress during seed dipping by growing a thin and long necking.3 Losing the dislocation-free structure during growth􀀁which is known as structure loss􀀁is one of the major issues facing themonocrystalline Si industry. The causes of structure loss are diverse due to the complexity of the growth process and sensitivity to disturbances. It has been reported in previous studies that foreign particle pinholes or the so-called gas bubbles and thermal shock can lead to dislocation generation.4−8  \nBeing able to identify and classify the different types of structure loss is an essential step toward understanding the root causes and, ultimately, preventing their occurrence. Currently, all methodologies that address this task involve human intervention. The process is repetitive and consists of carefully analyzing the images of ingots obtained with a digital microscope to find common features between the affected ingots.9 Because of this, machine learning (ML) and, most  \n© 2024 The Authors. Published by American Chemical Society  \nspecifically, deep learning (DL) seem to be the appropriate frameworks to automate this process. The reason is that MLand DL are often employed in finding nonlinear complex patterns in data. Therefore, applying ML may make the process of classifying the structure loss fast and accurate. This work aims to be a first step toward the automation of off-line classification of grown Cz-ingots. To accomplish this, three potential ML pipelines for the classification of structure loss will be studied. All the three pipelines are ","cbCailk5jJTmMSgB","https://ap.wps.com/l/cbCailk5jJTmMSgB","pdf",12703432,1,9,"English","en",105,"# Introduction\n## Solar and manufacturing context\n## Structure loss causes and current identification methods\n## Motivation and paper structure\n# Structure Loss and Ingot Categorization\n## Growth ridges and detection\n## Challenges of attributing causes in the Czochralski process","[{\"question\":\"What is the main problem addressed in the paper?\",\"answer\":\"The paper focuses on classifying Czochralski-grown monocrystalline silicon ingots that have experienced structure loss caused by dislocations.\"},{\"question\":\"How is structure loss detection automated in this work?\",\"answer\":\"Three machine-learning pipelines using different convolutional neural network architectures analyze ingot surface images to automate classification.\"},{\"question\":\"Which pipeline performed best and why?\",\"answer\":\"The pipeline combining a pretrained model with an incremental training strategy achieved the highest accuracies and the most stable training among the tested approaches.\"}]","Machine Learning Methods for Structure Loss Classification in Czochralski Silicon Ingots | PDF",1785937479,23,{"code":4,"msg":31,"data":32},"ok",{"site_id":24,"language":23,"slug":33,"title":13,"keywords":34,"description":14,"schema_data":35,"social_meta":87,"head_meta":89,"extra_data":91,"updated_unix":28},"machine-learning-methods-for-structure-loss-classification-in-czochralski-silicon-ingots","",{"@graph":36,"@context":86},[37,54,69],{"@type":38,"itemListElement":39},"BreadcrumbList",[40,44,48,51],{"item":41,"name":42,"@type":43,"position":20},"https://docshare.wps.com","Home","ListItem",{"item":45,"name":46,"@type":43,"position":47},"https://docshare.wps.com/document/","Document",2,{"item":49,"name":12,"@type":43,"position":50},"https://docshare.wps.com/document/research-report/",3,{"item":52,"name":13,"@type":43,"position":53},"https://docshare.wps.com/document/machine-learning-methods-for-structure-loss-classification-in-czochralski-silicon-ingots/127202/",4,{"url":52,"name":13,"@type":55,"author":56,"headline":13,"publisher":58,"fileFormat":61,"inLanguage":23,"description":14,"dateModified":62,"datePublished":63,"encodingFormat":61,"isAccessibleForFree":64,"interactionStatistic":65},"DigitalDocument",{"name":9,"@type":57},"Person",{"url":41,"name":59,"@type":60},"DocShare","Organization","application/pdf","2026-08-22","2026-08-05",true,{"@type":66,"interactionType":67,"userInteractionCount":20},"InteractionCounter",{"@type":68},"ViewAction",{"@type":70,"mainEntity":71},"FAQPage",[72,78,82],{"name":73,"@type":74,"acceptedAnswer":75},"What is the main problem addressed in the paper?","Question",{"text":76,"@type":77},"The paper focuses on classifying Czochralski-grown monocrystalline silicon ingots that have experienced structure loss caused by dislocations.","Answer",{"name":79,"@type":74,"acceptedAnswer":80},"How is structure loss detection automated in this work?",{"text":81,"@type":77},"Three machine-learning pipelines using different convolutional neural network architectures analyze ingot surface images to automate classification.",{"name":83,"@type":74,"acceptedAnswer":84},"Which pipeline performed best and why?",{"text":85,"@type":77},"The pipeline combining a pretrained model with an incremental training strategy achieved the highest accuracies and the most stable training among the tested approaches.","https://schema.org",{"og:url":52,"og:type":88,"og:title":13,"og:site_name":59,"og:description":14},"article",{"robots":90,"canonical":52},"index,follow",{"doc_id":7,"site_id":24},{"code":4,"msg":5,"data":93},[94,98,102,106,111,116,121,124,128,131,135],{"id":20,"doc_module":4,"doc_module_name":46,"category_name":95,"show_sort_weight":96,"slug":97},"Story & Novel",90,"story-novel",{"id":47,"doc_module":4,"doc_module_name":46,"category_name":99,"show_sort_weight":100,"slug":101},"Literature",80,"literature",{"id":53,"doc_module":4,"doc_module_name":46,"category_name":103,"show_sort_weight":104,"slug":105},"Exam",70,"exam",{"id":107,"doc_module":4,"doc_module_name":46,"category_name":108,"show_sort_weight":109,"slug":110},5,"Comic",60,"comic",{"id":112,"doc_module":4,"doc_module_name":46,"category_name":113,"show_sort_weight":114,"slug":115},6,"Technology",50,"technology",{"id":117,"doc_module":4,"doc_module_name":46,"category_name":118,"show_sort_weight":119,"slug":120},7,"Healthcare",40,"healthcare",{"id":11,"doc_module":4,"doc_module_name":46,"category_name":12,"show_sort_weight":122,"slug":123},30,"research-report",{"id":21,"doc_module":4,"doc_module_name":46,"category_name":125,"show_sort_weight":126,"slug":127},"Religion & Spirituality",20,"religion-spirituality",{"id":126,"doc_module":4,"doc_module_name":46,"category_name":129,"show_sort_weight":126,"slug":130},"World Cup","world-cup",{"id":132,"doc_module":4,"doc_module_name":46,"category_name":133,"show_sort_weight":132,"slug":134},10,"Lifestyle","lifestyle",{"id":136,"doc_module":4,"doc_module_name":46,"category_name":137,"show_sort_weight":107,"slug":138},19,"General","general"]