[go: up one dir, main page]

WO2019006222A1 - Systèmes et procédés de prédiction de défauts et de dimension critique utilisant un apprentissage profond dans un procédé de fabrication de semi-conducteurs - Google Patents

Systèmes et procédés de prédiction de défauts et de dimension critique utilisant un apprentissage profond dans un procédé de fabrication de semi-conducteurs Download PDF

Info

Publication number
WO2019006222A1
WO2019006222A1 PCT/US2018/040162 US2018040162W WO2019006222A1 WO 2019006222 A1 WO2019006222 A1 WO 2019006222A1 US 2018040162 W US2018040162 W US 2018040162W WO 2019006222 A1 WO2019006222 A1 WO 2019006222A1
Authority
WO
WIPO (PCT)
Prior art keywords
deep learning
learning model
wafer
sites
defect
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Ceased
Application number
PCT/US2018/040162
Other languages
English (en)
Inventor
Arpit YATI
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
KLA Corp
Original Assignee
KLA Tencor Corp
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Priority claimed from US15/814,561 external-priority patent/US11275361B2/en
Application filed by KLA Tencor Corp filed Critical KLA Tencor Corp
Priority to KR1020207002828A priority Critical patent/KR102568074B1/ko
Priority to CN201880041282.6A priority patent/CN110770886B/zh
Publication of WO2019006222A1 publication Critical patent/WO2019006222A1/fr
Anticipated expiration legal-status Critical
Ceased legal-status Critical Current

Links

Classifications

    • G—PHYSICS
    • G05—CONTROLLING; REGULATING
    • G05B—CONTROL OR REGULATING SYSTEMS IN GENERAL; FUNCTIONAL ELEMENTS OF SUCH SYSTEMS; MONITORING OR TESTING ARRANGEMENTS FOR SUCH SYSTEMS OR ELEMENTS
    • G05B19/00—Programme-control systems
    • G05B19/02—Programme-control systems electric
    • G05B19/418—Total factory control, i.e. centrally controlling a plurality of machines, e.g. direct or distributed numerical control [DNC], flexible manufacturing systems [FMS], integrated manufacturing systems [IMS] or computer integrated manufacturing [CIM]
    • G05B19/41875—Total factory control, i.e. centrally controlling a plurality of machines, e.g. direct or distributed numerical control [DNC], flexible manufacturing systems [FMS], integrated manufacturing systems [IMS] or computer integrated manufacturing [CIM] characterised by quality surveillance of production
    • G—PHYSICS
    • G05—CONTROLLING; REGULATING
    • G05B—CONTROL OR REGULATING SYSTEMS IN GENERAL; FUNCTIONAL ELEMENTS OF SUCH SYSTEMS; MONITORING OR TESTING ARRANGEMENTS FOR SUCH SYSTEMS OR ELEMENTS
    • G05B13/00—Adaptive control systems, i.e. systems automatically adjusting themselves to have a performance which is optimum according to some preassigned criterion
    • G05B13/02—Adaptive control systems, i.e. systems automatically adjusting themselves to have a performance which is optimum according to some preassigned criterion electric
    • G05B13/0265—Adaptive control systems, i.e. systems automatically adjusting themselves to have a performance which is optimum according to some preassigned criterion electric the criterion being a learning criterion
    • G05B13/027—Adaptive control systems, i.e. systems automatically adjusting themselves to have a performance which is optimum according to some preassigned criterion electric the criterion being a learning criterion using neural networks only
    • G—PHYSICS
    • G05—CONTROLLING; REGULATING
    • G05B—CONTROL OR REGULATING SYSTEMS IN GENERAL; FUNCTIONAL ELEMENTS OF SUCH SYSTEMS; MONITORING OR TESTING ARRANGEMENTS FOR SUCH SYSTEMS OR ELEMENTS
    • G05B2219/00—Program-control systems
    • G05B2219/30—Nc systems
    • G05B2219/32—Operator till task planning
    • G05B2219/32186—Teaching inspection data, pictures and criteria and apply them for inspection
    • G—PHYSICS
    • G05—CONTROLLING; REGULATING
    • G05B—CONTROL OR REGULATING SYSTEMS IN GENERAL; FUNCTIONAL ELEMENTS OF SUCH SYSTEMS; MONITORING OR TESTING ARRANGEMENTS FOR SUCH SYSTEMS OR ELEMENTS
    • G05B2219/00—Program-control systems
    • G05B2219/30—Nc systems
    • G05B2219/37—Measurements
    • G05B2219/37109—Photoelectric scanned raster, rule and photocell, microscope
    • G—PHYSICS
    • G05—CONTROLLING; REGULATING
    • G05B—CONTROL OR REGULATING SYSTEMS IN GENERAL; FUNCTIONAL ELEMENTS OF SUCH SYSTEMS; MONITORING OR TESTING ARRANGEMENTS FOR SUCH SYSTEMS OR ELEMENTS
    • G05B2219/00—Program-control systems
    • G05B2219/30—Nc systems
    • G05B2219/37—Measurements
    • G05B2219/37208—Vision, visual inspection of workpiece
    • G—PHYSICS
    • G05—CONTROLLING; REGULATING
    • G05B—CONTROL OR REGULATING SYSTEMS IN GENERAL; FUNCTIONAL ELEMENTS OF SUCH SYSTEMS; MONITORING OR TESTING ARRANGEMENTS FOR SUCH SYSTEMS OR ELEMENTS
    • G05B2219/00—Program-control systems
    • G05B2219/30—Nc systems
    • G05B2219/45—Nc applications
    • G05B2219/45031—Manufacturing semiconductor wafers

Definitions

  • the present disclosure relates to systems and methods for predicting defect sites and critical dimension measurements.
  • Fabricating semiconductor devices typically includes processing a substrate, such as a semiconductor wafer, using a large number of semiconductor fabrication processes to form various features and multiple levels of the semiconductor devices.
  • lithography is a semiconductor fabrication process that involves transferring a pattern from a reticle to a resist arranged on a semiconductor wafer. Additional examples of semiconductor fabrication processes include, but are not limited to, chemical-mechanical polishing (CMP), etch, deposition, and ion implantation.
  • CMP chemical-mechanical polishing
  • etch etch
  • deposition deposition
  • ion implantation ion implantation
  • process induced failures tend to fail at predetermined design patterns often repeated many times within the design. Elimination of spatially systematic, electrically relevant defects is important because eliminating such defects can have a significant overall impact on yield. Whether or not defects will affect device parameters and yield often cannot be determined from the inspection, review, and analysis processes described above since these processes may not be able to determine the position of the defect with respect to the electrical design.
  • Finding critical defect locations is crucial for the semiconductor manufacturing industry, especially to locate defects of interest (DOI) that are yield killers.
  • DOI defects of interest
  • a complex set of parameters are responsible for defect formation. Defects can be formed for many reasons and can be broadly categorized based on the root cause. These root causes include design-related issues, process-related issues, tool-related issues, and random defects.
  • Conventional optical or electron beam inspection tools have been used to scan the critical hot spots on the wafer to locate the real defect sites.
  • CD critical dimension
  • a method comprises scanning a wafer with an inspection tool. Presence of at least one defect is confirmed with a defect review tool. Parameters are inputted into a deep learning model. The parameters include one or more of: a location of the defect with respect to one or more of a design, a care area, or a design clip; focus; exposure; a type of the defect; neighboring design sites; and a layer type. In an instance, the parameters also include optical proximity correction. Defect sites are predicted, using a controller, based on the deep learning model. The defect sites are validated. The deep learning model is optionally retrained. During retraining, the predicting, the validating, and the retraining are repeated until the deep learning model meets a detection accuracy threshold.
  • the defect review tool may be a scanning electron microscope.
  • the scan may be a hot scan.
  • a heat map of the defect sites can be generated.
  • the deep learning model can be used to predict defects on a new wafer.
  • the deep learning model also can be used to predict defects for a new semiconductor manufacturing process on the wafer.
  • a non-transitory computer readable medium storing a program can be configured to instruct a processor to execute any of the variations of the method in the first embodiment.
  • a method is provided. The method comprises scanning a wafer with a scanning electron microscope. Using a controller, critical dimension variation is calculated across the wafer at sample sites. Parameters are inputted into a deep learning model. The parameters include one or more of: a location where critical dimension is measured with respect to one of a design, a care area, or a design clip; focus; exposure; neighboring design sites; and a layer type. In an instance, the parameters also include optical proximity correction. Critical dimension is predicted at sites across the wafer, using the controller, based on the deep learning model.
  • the sites with different critical dimensions are validated.
  • the deep learning model is optionally retrained. During retraining, the predicting, the validating, and the retraining are repeated until the deep learning model meets a detection accuracy threshold. [0016] A heat map of critical dimension variation can be generated.
  • the deep learning model can be used to predict critical dimension on a new wafer.
  • the deep learning model also can be used to predict critical dimension for a new semiconductor manufacturing process on the wafer.
  • a non-transitory computer readable medium storing a program can be configured to instruct a processor to execute any of the variations of the method in the second embodiment.
  • a system in a third embodiment, comprises a controller in electronic communication with a scanning electron microscope or with the scanning electron microscope and an optical inspection tool.
  • the controller includes a processor and an electronic data storage unit in electronic communication with the processor.
  • the controller is configured to: receive results of a review of a wafer; input parameters in a deep learning model; predict additional results at sites across the wafer; receive validation of the results at the sites across the wafer; and optionally retrain the deep learning model.
  • the results are one of defect locations or critical dimension.
  • the parameters include one or more of: a location where the results are measured with respect to one of a design, a care area, or a design clip; focus; exposure; neighboring design sites; and a layer type.
  • the additional results are the same as the one of the defect locations or the critical dimension.
  • the predict step, the validate step, and the retrain step are repeated until the deep learning model meets a detection accuracy threshold.
  • the system can further include a scanning electron microscope in electronic communication with the controller.
  • the system also can further include an optical inspection tool and a scanning electron microscope in electronic communication with the controller.
  • the controller can be configured to generate a heat map of results variation. [0023]
  • the controller can be configured to use the deep learning model to predict results on a new wafer.
  • the controller also can be configured to use the deep learning model to predict results for a new semiconductor manufacturing process on the wafer.
  • FIG. 1 is a flowchart illustrating an embodiment of a method in accordance with the present disclosure
  • FIG. 2 is a flowchart illustrating another embodiment of a method in accordance with the present disclosure
  • FIG. 3 is a flowchart of an example of the methods disclosed herein.
  • FIG. 4 is a system drawing of a system used in conjunction with the embodiments of the present disclosure.
  • Techniques disclosed herein collect wafer data, such as with an initial hot spot inspection.
  • an initial inspection or CD measurement can be made at various sites on the wafer.
  • the wafer data can include defects or CD variation.
  • the location, design clips, process tool parameters, or other parameters can be used to train a deep learning model, which is then used to predict across the wafer or a subsequent wafer. Visits to the predicted sites using inspection and review tools can validate the deep learning model and these results can be used to retrain the deep learning model. This process can be repeated until the predictions have an acceptable degree of accuracy.
  • the deep learning model can be used to predict new probable defect location or CD failure sites.
  • FIG. 1 is a flowchart illustrating a method 100.
  • a wafer is scanned with an inspection tool, such as an optical inspection tool, at 101.
  • Detecting defects on the wafer may involve using one or more optics modes including performing a hot scan on the wafer using the one or more optics modes.
  • a "hot scan” generally refers to a measurement/inspection of a wafer performed to detect defects or take measurements on the wafer by applying relatively aggressive detection settings (e.g., thresholds substantially close to the noise floor).
  • the hot scan may be performed to collect inspection or measurement data about the wafer that will be used for other steps disclosed herein.
  • the goal of the hot scan is to measure a representative sample of all CD measurements or detect a representative sample of all defect and nuisance types on the wafer in the selected mode(s).
  • the presence of at least one defect is confirmed at 102 with a defect review tool, which may be an SEM or another device. Sampling may be performed to confirm the presence of the at least one defect.
  • Parameters are inputted into a deep learning model at 103.
  • the parameters can include one or more of a location of the defect with respect to at least one of a design, a care area, or a design clip; tool parameters (e.g., focus, exposure); a type of the defect; neighboring design sites; a layer type; or optical proximity correction.
  • the neighboring design sites can be used to determine what kind of design files can lead to defects.
  • defect location may be with respect to a die corner.
  • the defect location also can be referenced to a wafer coordinate system or to a die index with respect to a wafer coordinate system.
  • the defect type may be the type of defect, such as a particle, void, short, bridge or any other general types of defect.
  • the neighboring design sites may be any specific shape or pattern of the design (e.g., T-shape or other critical shape), and may be or may be related to density.
  • the layer type may be a CMP layer, an etch layer, or other layers on the wafer.
  • Defect sites are predicted based on the deep learning model, which has been trained, at 104.
  • the predicting 104 may be performed by a controller.
  • the defect sites are validated at 105, such as with a defect review tool like an SEM or another device.
  • the validating can be performed to confirm presence of a defect at the defect site.
  • Wafer data can be collected at the defect sites. For example, SEM images are collected at the defect sites.
  • the deep learning model can use TensorFlow, which was originally developed by the Google Brain team.
  • TensorFlow is a software library for numerical
  • Graph nodes represent mathematical operations, while the graph edges represent the multidimensional data arrays (tensors) that flow between them.
  • Other mathematical libraries for general machine learning applications and neural networks can be used besides TensorFlow and TensorFlow is merely one example.
  • the deep learning model is only retrained with positive data and negative data is removed.
  • the deep learning model is retrained at 106.
  • the predicting 104, validating 105, and retraining 106 can be repeated until the deep learning model meets a detection accuracy threshold. Wafer data from the validating at 104 can be inputted into the deep learning model. In an instance, no retraining is necessary because the deep learning model meets the detection accuracy threshold based on results of the initial predicting 104.
  • the detection accuracy threshold can be set as per semiconductor manufacturer requirements. The detection accuracy threshold may be determined by a nuisance rate desired by a semiconductor
  • the deep learning model can be used to predict defects on a new wafer and/or predict defects for a new semiconductor manufacturing process on the wafer.
  • the input to the deep learning model may be the same as used for training or retraining.
  • the deep learning model should be able to predict the defects and/or CD measurements. If any changes are made to any of the parameters, those parameters may be inputted to the deep learning model to provide accurate predictions.
  • the method 100 can further include generating a heat map of the defect sites.
  • the heat map can illustrate possible defects and the type of defect on the wafer map.
  • FIG. 2 is a flowchart illustrating a method 200.
  • a wafer is scanned with an SEM at 101.
  • the CD variation across the wafer at sample sites is calculated at 202. This may be performed by a controller.
  • Parameters are inputted into a deep learning model at 203.
  • the parameters can include one or more of a location where CD is measured with respect to at least one of a design, a care area, or a design clip; tool parameters (e.g., focus, exposure); a type of the defect;
  • the neighboring design sites can be used to determine what kind of design files can lead to defects.
  • CD at sites across the wafer is predicted at 204 based on the deep learning model, which has been trained.
  • the predicting 204 may be performed by a controller.
  • the sites with different CDs are validated at 205, such as with an SEM or another device.
  • the validating can be performed to confirm CD at the sites.
  • Wafer data can be collected at the sites. In an instance, SEM images needs to be collected at those sites to confirm the presence of defects. Wafer data can refer to CD measurement data using SEM images, which can be collected at those sites to confirm the quantum of CD variation.
  • the deep learning model is retrained at 206.
  • the predicting 204, validating 205, and retraining 206 can be repeated until the deep learning model meets a detection accuracy threshold. Wafer data from the validating at 204 can be inputted into the deep learning model. In an instance, no retraining is necessary because the deep learning model meets the detection accuracy threshold based on results of the initial predicting 204.
  • the deep learning model can be used to predict CD on a new wafer and/or predict CD for a new semiconductor manufacturing process on the wafer.
  • the method 100 can further include generating a heat map of the defect sites.
  • the heat map can illustrate CD variation on the wafer map.
  • the method 100 and the method 200 use deep learning models to predict probable defect locations or CD measurements on a wafer. This can help find probable defect locations during semiconductor manufacturing. A reduction in nuisance rate is one of the biggest challenges in defect inspection because it reduces the inspection review cycle for finding critical DOI locations. Furthermore, heat maps of defects and/or CDs can be provided. This can improve process monitoring.
  • the method 100 and the method 200 use actual wafer data points and additional parameters.
  • a small sample set may be used for the initial prediction.
  • FIG. 3 is a flowchart 300 of an example of the methods disclosed herein.
  • Wafers are inspected with an inspection tool, such as an optical inspection tool, at 301.
  • Defect or CD sites can be found at 302, such as using an SEM.
  • the defect or CD type also can be determined at 302.
  • the deep learning model is trained at 303, such as with the location, defect type, and manufacturing tool parameters as inputs.
  • Sites with defects or CD failures are predicted with the deep learning model at 304 and confirmed at 305. Confirming the sites 305 can include revisiting the sites with the SEM, such as at 302, which can be used to further train the deep learning model.
  • the deep learning model can optionally be retrained as shown in 306.
  • FIG. 4 is a system drawing of a system 400.
  • the system 400 includes an optical inspection tool 401, a defect review tool 402, and a controller 403 with a processor 404 and an electronic data storage unit 405 in electronic communication with the processor 404.
  • the defect review tool 402 may be an SEM.
  • the wafer inspection tool 401 may be a BBP inspection tool, which can be configured to perform a hot scan to capture inspection results.
  • the wafer inspection tool 401 also may be an electron beam tool or a laser scanning tool.
  • the controller 403 is in electronic communication with the optical inspection tool 401 and the defect review tool 402.
  • the controller 403 may be part of the optical inspection tool 401 or the defect review tool 402, or another device.
  • the controller 403 may be a standalone control unit or in a centralized quality control unit. Multiple controllers 403 may be used.
  • the controller 403 may be implemented in practice by any combination of hardware, software, and firmware. Also, its functions as described herein may be performed by one unit, or divided up among different components, each of which may be implemented in turn by any combination of hardware, software and firmware. Program code or instructions for the controller 403 to implement various methods and functions may be stored in controller readable storage media, such as a memory in the electronic data storage unit 405 or other memory.
  • the controller 403 can be configured to receive wafer data.
  • the wafer data are one of defect locations or CD and may be generated by, for example, the defect review tool 402 or the optical inspection tool 401.
  • Parameters are inputted in a deep learning model.
  • the parameters include one or more of a location where the results are measured with respect to one of a design, a care area, or a design clip; tool parameters (e.g., focus, exposure); a type of the defect; neighboring design sites; a layer type; or optical proximity correction. Additional results are predicted at sites across the wafer. The additional results are the same as the one of the defect locations or the CD, meaning that if defect locations are received then defect locations will be predicted and if CD results are received then CD results will be predicted.
  • the controller 403 receives validation of results at the sites across the wafer. Thus, the additional results may be confirmed. If necessary, the deep learning model is retrained. Predicting, receiving validation, and retraining can be repeated until the deep learning model meets a detection accuracy threshold.
  • the controller can be configured to use the deep learning model to predict results on a new wafer or for a new semiconductor manufacturing process on the wafer. With new semiconductor processes, a new set of parameters can be inputted to the deep learning model to determine if predictions are accurate.
  • the controller 403 also can be configured to generate a heat map of results variation.
  • the system 400 can include an SEM 402 in electronic communication with the controller 403 or both an optical inspection tool 401 and an SEM 402 in electronic
  • the system 400 may only include an SEM 402 in electronic communication with the controller 403.
  • the controller 403 may be part of the optical inspection tool 401 or the SEM 402. [0054]
  • the controller 403 may be coupled to the components of the system 400 in any suitable manner (e.g., via one or more transmission media, which may include wired and/or wireless transmission media) such that the controller 403 can receive the output generated by the system 400.
  • the controller 403 may be configured to perform a number of functions using the output.
  • the controller 403, other system(s), or other subsystem(s) described herein may be part of various systems, including a personal computer system, image computer, mainframe computer system, workstation, network appliance, internet appliance, or other device.
  • the subsystem(s) or system(s) may also include any suitable processor known in the art, such as a parallel processor.
  • the subsystem(s) or system(s) may include a platform with high speed processing and software, either as a standalone or a networked tool.
  • the system includes more than one subsystem, then the different subsystems may be coupled to each other such that images, data, information, instructions, etc. can be sent between the subsystems.
  • one subsystem may be coupled to additional subsystem(s) by any suitable transmission media, which may include any suitable wired and/or wireless transmission media known in the art.
  • Two or more of such subsystems may also be effectively coupled by a shared computer-readable storage medium (not shown).
  • An additional embodiment relates to a non-transitory computer-readable medium storing program instructions executable on a controller for performing a computer-implemented defect detection or CD analysis, as disclosed herein.
  • the processor 404 can be coupled to a memory in the electronic data storage unit 405 or other electronic data storage medium with non-transitory computer-readable medium that includes program instructions executable on the processor 404.
  • the computer-implemented method may include any step(s) of any method(s) described herein.
  • the controller 403 may be programmed to perform some or all of the steps of FIG. 1, FIG. 2, or FIG. 3, which can be executed by the processor 404.
  • the memory in the electronic data storage unit 405 or other electronic data storage medium may be a storage medium such as a magnetic or optical disk, a magnetic tape, or any other suitable non-transitory computer-readable medium known in the art.
  • the program instructions may be implemented in any of various ways, including procedure-based techniques, component-based techniques, and/or object-oriented techniques, among others.
  • the program instructions may be implemented using ActiveX controls, C++ objects, JavaBeans, Microsoft Foundation Classes (MFC), SSE (Streaming SIMD Extension), or other technologies or methodologies, as desired.
  • MFC Microsoft Foundation Classes
  • SSE Streaming SIMD Extension
  • wafer generally refers to substrates formed of a semiconductor or non-semiconductor material.
  • a semiconductor or non- semiconductor material include, but are not limited to, monocrystalline silicon, gallium arsenide, and indium phosphide. Such substrates may be commonly found and/or processed in
  • a wafer may include one or more layers formed upon a substrate.
  • such layers may include, but are not limited to, a resist, a dielectric material, and a conductive material.
  • a wafer as used herein is intended to encompass a wafer including all types of such layers.
  • One or more layers formed on a wafer may be patterned or unpatterned.
  • a wafer may include a plurality of dies, each having repeatable patterned features. Formation and processing of such layers of material may ultimately result in completed devices.
  • ICs may be formed on a wafer, and the term wafer as used herein is intended to encompass a wafer on which any type of device known in the art is being fabricated.
  • the term "chip” may comprise a collection of ICs designed for a particular purpose.
  • reticle which may also be commonly referred to as a mask or a photomask.
  • reticle which may also be commonly referred to as a mask or a photomask.
  • mask which may also be commonly referred to as a mask or a photomask.
  • photomask many different types of reticles are known in the art, and the terms "reticle,” “mask,” and “photomask” as used herein are intended to encompass all types of reticles known in the art.
  • each of the steps of the method may be performed as described herein.
  • the methods also may include any other step(s) that can be performed by the controller and/or computer subsystem(s) or system(s) described herein.
  • the steps can be performed by one or more computer systems, which may be configured according to any of the embodiments described herein.
  • the methods described above may be performed by any of the system embodiments described herein.

Landscapes

  • Engineering & Computer Science (AREA)
  • General Physics & Mathematics (AREA)
  • Artificial Intelligence (AREA)
  • Automation & Control Theory (AREA)
  • Physics & Mathematics (AREA)
  • Quality & Reliability (AREA)
  • Manufacturing & Machinery (AREA)
  • General Engineering & Computer Science (AREA)
  • Evolutionary Computation (AREA)
  • Health & Medical Sciences (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Medical Informatics (AREA)
  • Software Systems (AREA)
  • Testing Or Measuring Of Semiconductors Or The Like (AREA)

Abstract

Selon la présente invention, une inspection initiale ou une mesure de dimension critique peut être réalisée au niveau de divers sites sur une plaquette. L'emplacement, les clips de conception, les paramètres d'outil de procédé ou d'autres paramètres peuvent être utilisés pour entraîner un modèle d'apprentissage profond. Le modèle d'apprentissage profond peut être validé et ces résultats peuvent être utilisés pour reformer le modèle d'apprentissage profond. Ce procédé peut être répété jusqu'à ce que les prédictions respectent un seuil de précision de détection. Le modèle d'apprentissage profond peut être utilisé pour prédire un nouvel emplacement de défaut probable ou des sites de défaillance de dimension critique.
PCT/US2018/040162 2017-06-30 2018-06-29 Systèmes et procédés de prédiction de défauts et de dimension critique utilisant un apprentissage profond dans un procédé de fabrication de semi-conducteurs Ceased WO2019006222A1 (fr)

Priority Applications (2)

Application Number Priority Date Filing Date Title
KR1020207002828A KR102568074B1 (ko) 2017-06-30 2018-06-29 반도체 제조 프로세스에서 딥 러닝을 사용하여 결함 및 임계 치수를 예측하기 위한 시스템 및 방법
CN201880041282.6A CN110770886B (zh) 2017-06-30 2018-06-29 用于使用半导体制造工艺中的深度学习预测缺陷及临界尺寸的系统及方法

Applications Claiming Priority (6)

Application Number Priority Date Filing Date Title
IN201741023043 2017-06-30
IN201741023043 2017-06-30
US201762545918P 2017-08-15 2017-08-15
US62/545,918 2017-08-15
US15/814,561 US11275361B2 (en) 2017-06-30 2017-11-16 Systems and methods for predicting defects and critical dimension using deep learning in the semiconductor manufacturing process
US15/814,561 2017-11-16

Publications (1)

Publication Number Publication Date
WO2019006222A1 true WO2019006222A1 (fr) 2019-01-03

Family

ID=64734414

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/US2018/040162 Ceased WO2019006222A1 (fr) 2017-06-30 2018-06-29 Systèmes et procédés de prédiction de défauts et de dimension critique utilisant un apprentissage profond dans un procédé de fabrication de semi-conducteurs

Country Status (1)

Country Link
WO (1) WO2019006222A1 (fr)

Cited By (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN113405493A (zh) * 2020-03-17 2021-09-17 应用材料以色列公司 确定三维信息
CN114154386A (zh) * 2020-09-07 2022-03-08 联华电子股份有限公司 半导体制作工艺的良率预测方法、系统与模型训练装置
CN116245846A (zh) * 2023-03-08 2023-06-09 华院计算技术(上海)股份有限公司 带钢的缺陷检测方法及装置、存储介质、计算设备
US12055904B2 (en) 2018-12-03 2024-08-06 Asml Netherlands B.V. Method to predict yield of a device manufacturing process

Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20080077907A1 (en) * 2006-09-21 2008-03-27 Kulkami Anand P Neural network-based system and methods for performing optical proximity correction
US20090063378A1 (en) * 2007-08-31 2009-03-05 Kla-Tencor Technologies Corporation Apparatus and methods for predicting a semiconductor parameter across an area of a wafer
US20110320149A1 (en) * 2009-02-06 2011-12-29 Kla-Tencor Corporation Selecting One or More Parameters for Inspection of a Wafer
US20130064442A1 (en) * 2011-09-13 2013-03-14 Kla-Tencor Corporation Determining Design Coordinates for Wafer Defects
US20160335753A1 (en) * 2013-05-14 2016-11-17 Kla-Tencor Corporation Machine learning method and apparatus for inspecting reticles

Patent Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20080077907A1 (en) * 2006-09-21 2008-03-27 Kulkami Anand P Neural network-based system and methods for performing optical proximity correction
US20090063378A1 (en) * 2007-08-31 2009-03-05 Kla-Tencor Technologies Corporation Apparatus and methods for predicting a semiconductor parameter across an area of a wafer
US20110320149A1 (en) * 2009-02-06 2011-12-29 Kla-Tencor Corporation Selecting One or More Parameters for Inspection of a Wafer
US20130064442A1 (en) * 2011-09-13 2013-03-14 Kla-Tencor Corporation Determining Design Coordinates for Wafer Defects
US20160335753A1 (en) * 2013-05-14 2016-11-17 Kla-Tencor Corporation Machine learning method and apparatus for inspecting reticles

Cited By (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US12055904B2 (en) 2018-12-03 2024-08-06 Asml Netherlands B.V. Method to predict yield of a device manufacturing process
CN113405493A (zh) * 2020-03-17 2021-09-17 应用材料以色列公司 确定三维信息
CN113405493B (zh) * 2020-03-17 2023-10-24 应用材料以色列公司 确定三维信息
CN114154386A (zh) * 2020-09-07 2022-03-08 联华电子股份有限公司 半导体制作工艺的良率预测方法、系统与模型训练装置
CN116245846A (zh) * 2023-03-08 2023-06-09 华院计算技术(上海)股份有限公司 带钢的缺陷检测方法及装置、存储介质、计算设备
CN116245846B (zh) * 2023-03-08 2023-11-21 华院计算技术(上海)股份有限公司 带钢的缺陷检测方法及装置、存储介质、计算设备

Similar Documents

Publication Publication Date Title
US11275361B2 (en) Systems and methods for predicting defects and critical dimension using deep learning in the semiconductor manufacturing process
US8139844B2 (en) Methods and systems for determining a defect criticality index for defects on wafers
US7681159B2 (en) System and method for detecting defects in a semiconductor during manufacturing thereof
US8194968B2 (en) Methods and systems for using electrical information for a device being fabricated on a wafer to perform one or more defect-related functions
TWI468958B (zh) 用於以設計為基礎之裝置風險評估之方法
US10127653B2 (en) Determining coordinates for an area of interest on a specimen
US20080163140A1 (en) Methods, designs, defect review tools, and systems for determining locations on a wafer to be reviewed during defect review
WO2017123888A1 (fr) Amélioration de la sensibilité aux défauts des inspecteurs de tranche de semi-conducteur en utilisant des données de conception incluant des données d'image de tranche
US10692690B2 (en) Care areas for improved electron beam defect detection
CN110582842B (zh) 依据光学检验结果进行计量导引检验样品成形
KR20180088485A (ko) 형상 기반 그루핑
WO2019006222A1 (fr) Systèmes et procédés de prédiction de défauts et de dimension critique utilisant un apprentissage profond dans un procédé de fabrication de semi-conducteurs
CN116583788A (zh) 半导体试样的电属性的预测
US10503078B2 (en) Criticality analysis augmented process window qualification sampling
KR102688381B1 (ko) 반도체 디바이스 제조 방법 및 반도체 제조 어셈블리를 위한 공정 제어 시스템
KR102589631B1 (ko) 뉴슨스 맵에 기반한 광대역 플라즈마 검사
CN120917381A (zh) 在没有地面实况信息的情况下监测cgi模型性能的方法
CN120871557A (zh) 一种确定光刻工艺窗口的方法

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 18824081

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

ENP Entry into the national phase

Ref document number: 20207002828

Country of ref document: KR

Kind code of ref document: A

122 Ep: pct application non-entry in european phase

Ref document number: 18824081

Country of ref document: EP

Kind code of ref document: A1