WO2002029392A2 - Defect source identifier - Google Patents

Defect source identifier Download PDF

Info

Publication number
WO2002029392A2
WO2002029392A2 PCT/US2001/031029 US0131029W WO0229392A2 WO 2002029392 A2 WO2002029392 A2 WO 2002029392A2 US 0131029 W US0131029 W US 0131029W WO 0229392 A2 WO0229392 A2 WO 0229392A2
Authority
WO
WIPO (PCT)
Prior art keywords
defect
source identifier
information
defect source
wafer
Prior art date
Application number
PCT/US2001/031029
Other languages
English (en)
French (fr)
Other versions
WO2002029392A3 (en
Inventor
Amos Dor
Maya Radzinski
Original Assignee
Applied Materials, Inc.
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 US09/905,607 external-priority patent/US6701259B2/en
Application filed by Applied Materials, Inc. filed Critical Applied Materials, Inc.
Priority to EP01977451A priority Critical patent/EP1322941A2/de
Priority to KR1020027007062A priority patent/KR20020063582A/ko
Publication of WO2002029392A2 publication Critical patent/WO2002029392A2/en
Publication of WO2002029392A3 publication Critical patent/WO2002029392A3/en

Links

Classifications

    • H—ELECTRICITY
    • H01—ELECTRIC ELEMENTS
    • H01L—SEMICONDUCTOR DEVICES NOT COVERED BY CLASS H10
    • H01L22/00—Testing or measuring during manufacture or treatment; Reliability measurements, i.e. testing of parts without further processing to modify the parts as such; Structural arrangements therefor
    • H01L22/20—Sequence of activities consisting of a plurality of measurements, corrections, marking or sorting steps

Definitions

  • the invention relates to a method or associated apparatus for performing defect analysis in a semiconductor wafer processing system. More particularly, the invention relates to a method and apparatus the uses image analysis to analyze semiconductor wafers to determine defect causes and locations .
  • Many techniques are used involving, e.g., optical systems, electron microscopes, spatial signature analysis, and energy dispersive x-ray microanalysis, to identify and analyze defects on a semiconductor wafer.
  • wafers are intermittently selected from a lot of wafers that is being processed, i.e., one in every N wafers is selected.
  • the selected wafers are analyzed using one or more of the above-identified analysis techniques (these techniques are performed by tools that are commonly referred to as metrology tools) .
  • metrology tools tools that are commonly referred to as metrology tools
  • the source of the defect is generally identified through trial and error, i.e., changes are made in the process parameters in an attempt to eliminate the defect in a wafer selected from another lot.
  • Some types of defects occur for well-known reasons. These defects are cataloged in a searchable database of defect data and images . An operator can compare the test results to the defect database in an attempt to match the test results to defects contained in the defect database. If a match is found, the database may identify the source of that particular type of defect. The operator can then take corrective action to eliminate the defect.
  • a relatively large amount of information relating to wafer defects is necessary to provide an illustrative sample of the varied and multiple defects that may occur to any semiconductor wafer that is being processed through a series of processes.
  • a defect analysis system using a large amount of stored data can provide more effective defect comparisons than a defect analysis system using a small amount of stored data.
  • Even large volume semiconductor processes will require a certain amount of time until sufficient numbers of wafers have been processed and analyzed to provide reliable defect source information.
  • processing semiconductor wafers is very expensive, and many companies or groups can only afford to process a relatively small number of semiconductor wafers through any prescribed set of processes for testing purposes . Therefore, a need exists in the art for a system that can effectively analyze wafer defects and repeatedly utilize the defect source information through the use of a shared database of defect data that is accessible over a wide area network.
  • FIG. 1 shows one embodiment of a defect source identifier arranged in accordance with the present invention
  • FIG. 2 shows a block diagram illustrating the processes performed by the defect source identifier of FIG.
  • FIG. 3 shows an option screen to be displayed on a display of the defect source identifier
  • FIG. 4 shows a portion of a configuration screen to be displayed on a display of the defect source identifier
  • FIG. 5 shows another portion of a configuration screen to be displayed by a defect source identifier
  • FIG. 6 shows another portion of a configuration screen to be displayed by a defect source identifier
  • FIG. 7 shows another portion of a defect summary screen to be displayed by a defect source identifier
  • FIG. 8 shows a defect image screen to be displayed by a defect source identifier
  • FIG. 9 shows a portion of a defect cause selection screen to be displayed by a defect source identifier
  • FIG. 10 shows another portion of a defect cause selection screen to be displayed by a defect source identifier
  • FIG. 11 shows a case image screen to be displayed by a defect source identifier
  • FIG. 12 shows an image compare screen to be displayed by a defect source identifier
  • FIG. 13 shows a wafer search screen to be displayed by a defect source identifier
  • FIG. 14 shows a block diagram of block diagram of the defect source identifier, progressing through the screens shown in FIGs . 3 to 13 ;
  • FIG. 15 shows a defect detection method performed by the defect source identifier shown in FIG. 1;
  • FIG. 16 shows a multi-level client server architecture of one embodiment of defect source identifier.
  • defect source identifier 100 identifies defects in the wafers processed by a wafer processing system 102.
  • the wafer processing system 102 includes one or more process cells 103. Each one of the process cells is configured to perform such exemplary processes on wafers as chemical vapor deposition (CVD) , physical vapor deposition (PVD) , electro-chemical plating (ECP) , electroless deposition, other known deposition processes, or other known etching processes.
  • the defect source identifier 100 includes metrology tools that analyze defects that have occurred in wafers during processing within the wafer processing system 102.
  • the defect source identifier 100 transfers wafer data, images, and/or information relating to the wafer defects to a remote location for analysis. Certain embodiments of the defect source identifier 100 compare wafer images to case histories of wafer defects, performs spectral analysis on the wafer, and/or transfers defect sources and operational solutions to defects to the wafer processing system (or to an operator located at the wafer processing system) . The defect source identifier 100 analyzes undesired operation of and/or the states of one or more of the process cells as evidenced by defects in the wafers that have been processed within the process cells as well as the states of the process cells. Wafers that may undergo processing in process cells include semiconductor wafers or some other form of substrate upon which sequential process steps are performed.
  • defect source identifier 100 shown in FIG. 1 includes a wafer processing system 102, one or more defect source identifier clients 104, one or more defect source identifier servers 106, and a network 110.
  • the wafer processing system 102 includes a transfer cell 120 (also known as a factory interface) , a plurality of process cells 103, a wafer transfer system 121 (also referred to as a wafer transport robot or simply a robot) , and a factory interface 122.
  • the factory interface 122 includes a cassette load lock 123 and a metrology cell 124.
  • the cassette load lock 123 stores one or more wafer cassettes . The individual wafers are moved from the cassette 123 to the process cells 103 by the robot 121.
  • the metrology cell 124 includes metrology tools 180 that measure and test the wafer characteristics and wafer defects.
  • the metrology tools include, e.g., a scanning or transmission scanning electron microscope, an optical wafer defect inspection system, spatial signature analysis, or any metrology tool used to analyze defects of wafers, either in combination or individually.
  • a plurality of defect source identifier clients 104 are shown in the embodiment of FIG. 1 as defect source identifier clients A, B, and C. The following description references the defect source identifier client A, but is representative of all of the defect source identifier clients.
  • the defect source identifier client 104 includes the client computer 105 to control the operation of both the wafer processing system 102 and the individual process cells 103 in the wafer processing system 102.
  • the defect source identifier server 106 includes the server computer 107.
  • the client computer 105 interacts with the server computer 107 via the network 110 to receive data stored in the server computer 107 that relates to present and historical (i.e., case study) defects on wafers processed by the wafer processing system 102.
  • the client computer 105 and the server computer 107 interact with the metrology tools 180 of the metrology cell 124 and a variety of databases 186 that store wafer defect case histories to analyze defect generation in the wafer processing system 102.
  • the network 110 provides data communications between the client computer 105 and the server computer 107.
  • the network 110 may utilize the Internet, an intranet, a wide area network (WAN) , or any other form of a network.
  • WAN wide area network
  • the network 110 may utilize such computer languages utilized by, e.g., the Internet such as Hypertext Markup Language (HTML) or extensible Markup Language (XML) .
  • HTML is presently the predominant markup language utilized over the Internet.
  • XML is a markup language that is gaining greater acceptance in the Internet .
  • the use of HTML and/or XML requires the use of a respective HTML and/or XML browser installed at each client computer 105.
  • the client 104 and the defect source identifier server 106 interact to identify defects on processed semiconductor wafers and provide solutions to the wafer defects .
  • the operation of the wafer processing system 102 is controlled by a particular defect source identifier client 104.
  • the defect source identifier client 104 receives solutions from the defect source identifier server 106.
  • the solutions are applied to the wafer processing system 102 (either automatically or input from an operator) , and the solutions are used to control the operation of the wafer processing system.
  • the reference number of elements in the client computer 105 are appended with an additional reference character , a" .
  • the reference characters of the server computer 107 are appended with an additional reference character "b" .
  • the suitable respective reference character w a" or "b" is provided.
  • the appended letter following the reference character may be omitted.
  • the respective client computer 105 and server computer 107 comprise a respective central processing unit (CPU) 160a, 160b; a memory 162a, 162b; support circuits 165a, 165b; an input/output interface (I/O) 164a, 164b; and a bus 116a, 166b.
  • the client computer 105 and the server computer 107 may each be fashioned as a general-purpose computer, a workstation computer, a personal computer, a laptop computer, a microprocessor, a microcontroller, an analog computer, a digital computer, a microchip, a microcomputer, or any other known suitable type of computer.
  • the CPU 160a, 160b performs the processing and arithmetic operations for the respective client computer 105 and server computer 107.
  • the memory 162a, 162b includes random access memory (RAM) , read only memory (ROM) , removable storage, disk drive storage, that whether singly or in combination store the computer programs, operands, operators, dimensional values, wafer process recipes and configurations, and other parameters that control the defect source identification process and the wafer processing system operation.
  • RAM random access memory
  • ROM read only memory
  • Each bus 166a, 166bin the client computer 105 or the server computer 107 provides for digital information transmissions between respective CPU 160a, 160b; respective support circuits 165a, 165b; respective memory 162a, 162b; and respective I/O 164a, 164b.
  • the bus 166a, 166b in the client computer 105 or the server computer 107 also connects respective I/O 164a, 164b to other portions of the wafer processing system 102.
  • I/O 164a, 164b provides an interface to control the transmissions of digital information between each of the elements in the client computer 105 and/or the server computer 107. I/O 164a, 164b also provides an interface between the elements of the client computer 105 and/or the server computer 107 and different portions of the wafer processing system 102. Support circuits 165a, 165b comprise well-known circuits that are used in a computer such as clocks, cache, power supplies, other user interface circuits, such as a display and keyboard, system devices, and other accessories associated with the client computer 105 and/or the server computer 107. . To collect defect information, the client 104 is coupled to one or more metrology tools 180 within the wafer processing system 102.
  • the metrology tools that can perform a desired inspection on the wafer in the metrology cell 124 or cells include optical-based wafer defect inspection process, a scanning electron microscope process, and/or other wafer defect tools or processes.
  • the defect data collected by the client 104 is shared with the DSI server 106 via the I/O 164- , 164b and network 110. This defect data as well as process information is stored in various databases 186.
  • Client databases 188 are used to support various processes in the client 104.
  • the defect source identifier 100 utilizes an automated defect source identification software program 182, 184, portions of which are stored in the memory 162a or 162b to run respectively on the client computer 105 and the server computer 107.
  • the defect source identifier 100 automatically derives the source of a defect and either displays the possible causes with minimal user intervention and/or automatically remedies the process situation in the wafer processing system 102 that lead to the defect. Due to the automation of certain embodiments of defect source identifier 100 (and the production of possible solutions to certain defects by referencing historical defect case information) .
  • the defect source identifier 100 reduces problem solving cycle time, simplifies the defect source identifying process, and improves defect identification accuracy.
  • the defect source identifier 100 may be organized as a network-based application that generates an executive summary screen that is typically subdivided into a plurality of graphical user interface screen.
  • the graphical user interface screen displays its interfaces and defect sources at the defect source identifier client 104.
  • the users at the defect source identifier client 104 can thus interact with the defect knowledge database at the defect source identifier client to populate the executive summary screen.
  • the defect source identifier 100 can be configured as a stand-alone system contained in the defect source identifier client 104 that can operate without the network 110 and the defect source identifier server 106.
  • the selected configuration of the defect source identifier depends largely on the desired operation and performance characteristics of the system 100.
  • Defect Source Identifier Operation and Structure Different embodiments of the defect source identifier 100 receive data, text, images, defect case histories, etc. from one or more of a wide variety of databases, optical wafer inspection processes, and/or scanning electron microscope processes.
  • FIG. 2 shows one embodiment of the interrelated processes utilized by the defect source identifier 100. The varied processes included' in the embodiment of the defect source identifier 100 shown in FIG.
  • defect source identifier process 200 comprises a defect source identifier process 200, an (optical) wafer defect inspection process 204, a scanning electron microscope process 206, a defect management database process 208, a manufacturing execution database process 210 (that may be operationally and/or structurally subdivided into a distinct FAB manufacturing execution database process and/or a routing workstation manufacturing execution database process) , a defect source identifier database process 214, a defect knowledge database process 216 (also referred to as a defect knowledge library) , a customer knowledge database process 218, and a tool reference database process 220.
  • the defect knowledge database process 216 stores defect knowledge information while the customer knowledge database process 218 stores customer knowledge information.
  • the defect knowledge information and the customer knowledge information may together be defined as defect source information.
  • process is used to describe processes 200, 204, 206, 208, 210, 216, 218, and 220, it is envisioned that certain ones of these processes may be fashioned using software, hardware, databases, metrology equipment, and/or any suitable component, as described to perform the function of the process.
  • the wafer defect inspection process 204 and the scanning electron microscope process 206 are characterized as metrology tools (180 in FIG. 1),.
  • the metrology tools may further comprise a variety of processes such as wafer defect analyzers, transmission electron microscope, spatial signature analysis, ion beam analyzers, etc.
  • Other types of optical wafer defect inspection equipment may be utilized by the defect source identifier 100 in a similar manner as described.
  • KLA-TENCOR® of San Jose, CA is a producer of optical wafer defect inspection equipment such as shown at 204 and/or 206.
  • the defect images from the wafer defect inspection process 204 and/or the scanning electron microscope process 206 outputs detect information in the form of, e.g., a KLA file or KLA resource files (KLARF) .
  • KLARF KLA resource files
  • the wafer defect inspection process 204 generates defect inspection information as a KLA file that can be utilized in, stored by, or displayed to users located at the defect source identifier client 104 or the defect source identifier server 106.
  • the wafer defect inspection process produces a high resolution image of the wafer.
  • the scanning electron microscope process 206 is used to inspect the surface or subsurface of the wafer.
  • One embodiment of the scanning electron microscope process automatically classifies general defect types as the defects are identified by the microscope.
  • One embodiment of the scanning electron microscope process 206 generates defect inspection information as a "KLA file" that can be analyzed, stored, or displayed by the defect source identifier server 106 or the defect source identifier client 104.
  • An embodiment of the defect source identifier client 104 includes a display to view defect images referenced by the KLA files produced by such metrology tools as the wafer defect inspection process 204 or the scanning electron process 206.
  • One embodiment of the defect source identifier 100 allows wafer defect case histories to be displayed on a display of the defect source identifier client 104. An image from a current defect may be displayed on the display beside the image of a case study defect (reference image) for comparison purposes.
  • the defect source identifier system 100 creates and displays a wafer map image for each wafer that will visually indicate the location of defects on the wafer.
  • the defect management database process 208 stores and accesses defect images, data, and information. Such images, data, and other recently, collected information may be utilized during repetitive wafer defect analysis of one or more wafers. Such repetitive wafer defect analysis may be utilized to provide defect repeater information (e.g., where a similar defect occurs at the same location of subsequent processed wafers) and adder information (where a similar defect has not occurred in a similar location in another wafer) . The data, images, or other information may also provide cluster information, where multiple instances of a defect occur within a region.
  • the defect source identifier process 200 is coupled to the manufacturing execution database process 210.
  • manufacturing execution database process 210 includes a WORKSTREAM * manufacturing execution system, manufactured by CONSILIUM ® of Mountain View, Calif.
  • the manufacturing execution database process 210 is a database application that controls the flow routes of the wafer lots utilized during the manufacturing process. As such, the manufacturing execution database process contains routing information about which processes have been applied to each wafer or wafer lot. Such lot routing information is useful in determining those processes (or series of processes) that wafers having defects have undergone .
  • the manufacturing execution database process 210 may also include an equipment interface and a recipe management system.
  • the manufacturing execution database process 210 therefore contains considerable information about each process and condition used by the process cells 103 to process each wafer.
  • the manufacturing execution database process 210 thus forms "context information" and forms a message to send to a recipe management system that is used to set the recipe for processing each wafer.
  • the context information can be used to uniquely identify the process that is going to occur in a recipe in a specific process cell, and includes such information as lot number, entity, product, route, etc.
  • the recipe management system produces a "recipe" based on the message provided by the manufacturing execution database process 210.
  • the recipe is essentially the process instructions, such as the pressure, temperature, gas flow, etc. for that product in that step.
  • the manufacturing process steps are then performed by the respective processing tool in accordance with the setup and the recipe.
  • Some data collection is performed by the wafer processing system 102 such as reports on when the processing began, ended, etc. This information is sent to the manufacturing execution database process 210 and stored, e.g., in a lot and entity record in the memory.
  • defect source identifier process 200 is configured to allow transfer of data between the customer knowledge database process 218 and a defect knowledge database process 216 (a class cross-reference file is used to make this transfer) .
  • the defect knowledge database process 216 is typically stored in the memory 162b of the server computer 107.
  • the defect knowledge database process 216 stores case history defect data, images, and information obtained from a variety of sources, e.g., defect source identifier clients 104.
  • the customer knowledge database process 218 is typically stored in the memory 162a of the client computer 105.
  • the customer knowledge database process 218 stores and accesses case histories defect data, images, and information obtained from a single defect source identifier client 104. If a user has access to both the defect knowledge database process 216 and the customer knowledge database process 218, it will be important for the user to access databases in both the client 104 and the server 106 during a defect case history search.
  • defect source identifier 100 if any one specific customer knowledge database process 218 supports one specific defect knowledge database process
  • defect knowledge database process 216 At least some of the contents handled by the customer knowledge database process 218 (defect data, images, and information) will be allowed to be accessed by the defect knowledge database process 216.
  • One embodiment of the automated embodiments of defect source identifier 100 utilizes a software program 182, 184 that includes image processing and data analysis technology.
  • the automated defect source identifier 100 matches current defects occurring in the wafer processing system 102 with previously collected defect inspection information.
  • the defect knowledge database process 216 and a customer knowledge database process 218 cooperate to accumulate historical defect source information. Information for both the defect knowledge database process 216 and the customer knowledge database process 218 may be stored in either memory 162a and/or memory 162b.
  • the defect source identifier database process 214 stores and accesses data relating to the sources of defects. For each defect source, a list of defect solutions (e.g., possible corrective actions, that can be taken to correct certain defects) is stored.
  • the defect source identifier database 214 contains specific data from the KLA files produced by the optical wafer defect inspection process 204 and the scanning electron microscope process 206.
  • the defect source identifier database 214 also contains file references to the inspection image files .
  • defect source identifier clients 104 may utilize historic defect data, images or other information stored by both the customer reference database process 218 and/or the defect knowledge database process 216.
  • the defect knowledge database process 216 stores and accesses images, data, or other information relating to the historical defect cases of the defect source identifier 100.
  • the images, data, or other •information in the defect knowledge database process 216 is preferably compiled by interaction, over time, with a plurality of individual defect source identifier clients 104.
  • Each defect source identifier client 104 may be operated by a different company or group.
  • the customer knowledge database process 218 utilizes data, images, or other information relating to the defect case for a particular defect source identifier client 104.
  • the historical defects relating to multiple defect source identifier clients 104 that are in communication with the defect source identifier server 106 may be stored as data in the memory 162b of the defect source identifier server 106. Only certain defect source identifier clients 104 may access the data, images, or other information contained in the defect knowledge database process 216.
  • One embodiment of the defect source identifier 100 is configured to allow access to historic case information stored by the defect knowledge database process 216 only if the customer knowledge database process 216 of that particular defect source identifier client supports the defect knowledge database process 216 of the defect source identifier server 106. If the customer knowledge database process 218 of a particular defect source identification client 104 supports the defect knowledge database process 216, then the individual customer knowledge database process 218 provides access to the historical defect cases in the defect source identifier client. Therefore, the defect source identifier client 104 can obtain historic data, images, or other information from only these defect knowledge database process if that defect source identifier client 104 supports (by allowing the defect source identifier server 106 to access the data, images, and other information contained in the customer knowledge database process) .
  • Allowing a defect knowledge database process 216 to access data, images, or other information from a plurality of customer knowledge database processes 218 allows the defect knowledge database process to obtain historic defect data, images, or other information from a large variety of different defect source identifier clients 104.
  • the defect knowledge database process 216 becomes a repository of wafer defect data, images, or other information from a potential vast array of different defect source identifier clients 104.
  • the different defect source identifier clients may or may not be operated by a variety of different companies or groups that process wafers differently so the wafers are exposed to a vast array of different wafer processing techniques and wafer defects.
  • the data images, data, or other information relating to defects initially detected by a first defect source identifier client 104 operated by a first company or group may be later utilized for analysis purposes by a second defect source identifier client 104 operated by a different company or group.
  • the identity of the company or group operating the first defect source identifier client 104 may not be available to the operators of the second defect source identifier clients.
  • certain aspects of the process cell conditions, recipes, operating temperatures, and/or one or more solutions to the defect may be provided to the operators of the second defect source identifier client.
  • Individual defect source identifier clients 104 might, or might not, individually process a sufficient number of wafers to compile sufficient data, images, or other information to make their individual customer knowledge systems reliable.
  • the number historical wafer data, images, and information relating to most processes can be increased by utilizing the vast defect knowledge database process 216 that includes information from other defect source identifier clients 104.
  • An embodiment of the defect source identifier process 200 gathers such defect attributes as adders, repeaters, spatial signature analysis, and cluster information from the defect management database process 208 in near real-time.
  • the defect source identifier process 200 gathers lot routing information from the manufacturing execution database process 210 in near real-time.
  • the defect source identifier process 200 of selected system users may access the defect knowledge database process 216 and/or the customer knowledge database process 218. If the databases processes 216, 218 are available to a specific user, the users defect source identifier performs optimally when it utilizes the stored images, data, and other case history information from both the defect knowledge database process 216 and the customer knowledge database process 218.
  • the defect knowledge database process and the customer knowledge database process may be each accessed through known database access programs and techniques such as ADO.
  • the images produced by the wafer defect inspection process 204 and the scanning electron microscope process 206 are typically in the form of TIFF files. Images, data, and other information in database processes 208, 210, 216, 218, 220, and 214 can also be stored in TIFF format.
  • Multiple images may be contained in a single TIFF file in which the image file directory in the TIFF file contains multiple entries, one entry for each image.
  • the file includes not only the multiple images, but also alignment data indicating the alignment of the different images in the file. Both alignment and defect image data are thus contained in TIFF file referenced in the KLA File. Storing multiple images in a single TIFF file avoids requiring a separate TIFF file for each image.
  • the multiple mages associated with a single defect may be contained in a single, or multiple, TIFF file.
  • Multiple TIFF files are defined by multiple TiffFileName records in the KLA File.
  • the defect source identifier system 100 is configured to convert TIFF defect image files to JPEG-compressed or MPEG-compressed image files because the compressed image files are readily transported between any one of the client computers 105 and the server computer 107.
  • the defect source identifier process 200 connects to the scanning electron microscope process 206 and the optical wafer defect inspection process 204. This connection between processes 204 and 206 allows the user to access the process's 204, 206 historic KLA files and/or other image files.
  • An embodiment of the defect source identifier process 200 supports retrieving a processing tool list from a flat file within the tool reference database process 220.
  • the defect source identification server 106 runs a defect source identifier database 186 using the CPU 160b to access memory 162b.
  • One embodiment of the defect source identifier may use a database that is accessed by the defect source identifier servers 106, e.g. an SQL server database.
  • the defect source identification client 104 may contain well-known network client software that is designed to support interaction with the network server.
  • the network client software includes an operating system such as WINDOWS NT ⁇ , SOLARIS ® (a registered trademark of Sun Microsystems, Inc. of Palo Alto, CA) , or IRIX ® (a registered trademark of SGI of Mountain View, CA) .
  • the defect source identification client 104 runs a browser such as INTERNET EXPLORER ® (a registered trademark of the Microsoft Corporation of Redmond, WA) or NETSCAPE NAVIGATOR * (a registered trademark of Netscape Communications Corporation of Mountain View, CA) .
  • the defect source identifier could be developed in such languages as VISUAL BASIC (hereinafter referred to as VB) (a trademark of the Microsoft Corporation of Redmond, WA) , C, C++, or other, object-oriented or traditional computer programming languages .
  • the defect source identifier server 106 executes the defect knowledge database process 216 and the customer knowledge database process 218 through the communication process of the defect knowledge database process 216, that is compatible with VB.
  • the defect source identifier server 106 executes the manufacturing execution database process through communication processes that are compatible with VB.
  • Certain embodiments of database software support enterprise networks, including ORACLE ⁇ i from Oracle, QUEST Quest Software of Irvine, CA, and KNIGHT TM , through COM processes that are compatible with VB.
  • a KLA result file (KLARF) , or KLA file, is a flat ASCII file produced by computer equipment.
  • the defect source identifier supports the KLA or KLARF files produced by the optical wafer defect inspection process 204 to capture specific parameters from the wafer defect inspection process.
  • the KLARF and image files from the optical wafer defect inspection process 204 and the scanning electron microscope process 206 be exported by the tools onto a directory local to the tool .
  • Each tool connected to the defect source identifier makes available their export directory as a Network File System (NFS) mountable file system.
  • NFS Network File System
  • a series of graphical user interface may be displayed on, e.g., a screen, monitor, or other display associated with the respective I/O 164a, 164b on either the respective defect source identifier client 104, the defect source identifier server 106, or at a location in the network 110 in a manner to provide user interaction.
  • the GUI display is typically located at the defect source identifier client 104 to provide user interactivity.
  • the GUI of the defect source identifier 100 may display a series of interface screens within a browser window such as a login screen, a configuration screen that can contain multiple segments as shown in FIGs. 4 to 6, a defect summary screen one embodiment of which is one embodiment of which is shown in FIG.
  • a defect image screen that may contain multiple screens one embodiment of which is in FIGs. 9 and 10, a case image screen one embodiment of which is shown in FIG. 11, an image compare screen one embodiment of which is shown in FIG. 12, and a wafer compare screen one embodiment of which is shown in FIG. 13.
  • These GUI screens provide interactivity for a use with the defect source identifier 100 so the defect source identifier 100 can analyze surface features of a desired wafer using prescribed tools and techniques.
  • the GUI screens displayed in FIGs . 3 to 13 may be considered to represent different "states" to allow input of different information, and display different information, relative to the defect source identifier 100.
  • the user of the defect source identifier can navigate between the different GUI screen states as indicated by the embodiment of interaction state diagram 1400 shown in FIG. 14.
  • the interaction state diagram 1400 of FIG. 14 should be viewed in conjunction with the GUI screens described relative to FIGs. 3 to 13. Though the term “screen” is used in many cases in this disclosure to describe the various GUIs, the terms “screens”, “GUIs”, or “displays” are used interchangeably.
  • a user logs onto the defect source identifier client in step 1402 with a unique user identification and password by entering the information into a login screen (not shown) . The user must be authorized to log in before they can access the defect source identifier client 104.
  • the defect source identifier 100 is preferably provided with its own log in that is distinct from the operating system log in.
  • the distinct defect source identifier relates to the variety of different users having different job requirements and thereby requiring different levels of interactivity.
  • Each user is assigned an account that characterizes a prescribed user authorization level . Differing access levels are provided to different users such as wafer defect inspection process operator, scanning electron microscope process operator, FAB engineer, FAB defect source identifier administrator, yield expert, etc.
  • the embodiment of mode selection screen 300 shown in FIG. 3 is displayed at step 1404 in FIG. 14.
  • the mode selection screen 300 includes the mode fields, in which a user selects either start defect source identifier option or a configuration option as the desired mode.
  • the mode selection screen 300 allows the user to select a configuration option shown in decision step 1406.
  • the user can select a defect source identifier button 302 to start the operation of the defect source identifier 100 in a prescribed system configuration that is typically selected by all new users prior to using the defect source identifier 100.
  • the user selects the configuration button 304 to edit the configuration of the defect source identifier by starting from an existing, e.g., saved, configuration.
  • the method 1400 continues to decision step 1408 in which the defect source identifier 100 determines whether the user has proper authorization to receive an existing saved configuration information that can be displayed or edited by the user.
  • the user can enter a configuration screen (portions of which are shown in FIGs. 4, 5, and 6) by pressing the configuration button 304 in the embodiment of mode selection screen 300 shown in FIG. 3.
  • the configuration button 304 of the mode selection screen will be enabled in step 1408 if the user has the required authorization.
  • step 1412 If the answer to decision step 1408 is no, the method continues to step 1412. If the answer to decision step 1408 is yes, the method continues to step 1410 in which the saved configuration screen is displayed on the display screen of the defect source identifier. Following step 1410, the method 1400 continues to step 1414.
  • a new configuration screen (s), one embodiment shown in FIGs. 4, 5, or 6 will be displayed.
  • the initialization function of the configuration screen is performed.
  • Portions of the configuration screen are shown respectively at 410, 510, and 610 of FIGs. 4, 5, 6.
  • the configuration screen allows the user to select options, in step 1414, that affect the data from the defect source identifier 100 that is displayed to the users.
  • the configuration screen is divided into multiple configuration screen portions in the embodiment shown in FIGs. 4, 5, and 6 as, respectively, configuration screen portions 410, 510, and 610.
  • Each configuration screen portion 410, 510, and 610 displays one or more of the configuration options. If the user wants to change the settings of the configuration options, the user can access and edit the appropriate configuration screen in step 1414 from the embodiment of defect summary screen 702 shown in FIG. 7, or at startup of the defect source identifier client 104.
  • a station type button 401, a classification criteria button 403, and a display classes button 404 are each positioned on each configuration screen portion 410, 510, and 610 in FIGs. 4 to 6.
  • the buttons allow the user to enter the desired configuration screen portion, when the user is at the state indicated by step 1414 in FIG. 14. Accessing the desired configuration screen portion 410, 510, and 610 allows the user to set the desired respective parameters.
  • a done button 406 is positioned on each configuration screen portion 410, 5-10, and 610 to close that particular configuration screen portion while saving the updated configuration.
  • the defect source identifier 100 stores the configuration information for each accessed and saved defect source identifier server.
  • a cancel button 408 is positioned on each configuration screen portion 410, 510, and 610 to close that configuration screen portion and cancel any configuration updates. Access to any one of the specific configuration buttons 401, 403, 404, 406, or 408 may be either enabled or disabled depending on the user's level of access. Selecting the station type button 404, when the user is at the state indicated by step 1414, in each respective configuration screen portion 410, 510, and 610, allows the user to be transferred to the station type configuration screen portion 410 of FIG. 4.
  • the user can select the station type to which the tool (such as the optical wafer defect inspection process 204 or the scanning electron microscope process 206) is connected as the station is to run in near real- time.
  • the select station type display configuration screen portion 410 contains three buttons (e.g., radio buttons) 420, 422, 424 that, when selected, respectively allows the user to select from three respective modes : the wafer defect inspection process mode, the scanning electron microscope process mode, or the off-line (i.e., search) mode. User entry to each of these modes may be regulated according to the access level authorization.
  • the wafer defect inspection process mode and scanning electron microscope process mode allow the user to monitor that specific inspection process being conducted.
  • the defect summary screen of FIG. 7 will update in real-time.
  • the defect summary screen is updated after each wafer is inspected by the scanning electron microscope process.
  • the defect summary screen is updated after each lot is inspected. The last wafer inspected in the lot is automatically displayed on the screen and if the user wishes to view any other wafers from the lot, the global search button on the defect summary screen may be used as described below. Since all stations do not require real-time analysis to be their default mode of operation, those stations can be configured as off-line (search) stations by selecting button 424. Configuring the defect source enables the user to view specific previously inspected wafers.
  • the search mode station and search function on the real-time stations display information from both the wafer defect inspection process and scanning electron microscope process.
  • Selecting the classification criteria button 403 from any of the configuration screen portions 410, 510, or 610 in respective FIGs. 4, 5, or 6 causes the classification criteria screen portion 510, shown in FIG. 5, to be displayed when the method 1400 shown in FIG. 14 is in step 1414.
  • the user can set parameters that determine which defects or wafers are displayed (e.g., all wafers or only wafers that were "flagged" as in excursion cases) .
  • classification criteria configuration screen portion 510 is used to determined the types of defect, the tools to view the defects, and the wafers will be displayed on the defects summary screen (FIG. 7) and also what information will be displayed for the wafers .
  • the user sets the parameters pertaining to adders/repeaters, clusters, spatial signature analysis and excursion in the classification criteria configuration screen portion 510.
  • Adders/repeaters can be viewed by selecting box 524 and then selecting either the adders button 520, the repeaters button 522 or both.
  • the adders option selected by selecting radio button 520 causes the system 100 to calculate and display the defects that were detected by the tool that most recently inspected the wafer.
  • the repeaters option selected by selecting radio button 522 causes the system to calculate and display the defects that are repeated throughout a plurality of wafers. If neither the adders nor repeaters options are selected, then a default all defects are displayed on the defect summary screen.
  • the cluster option activated by selecting check box 525, identifies clusters of defects on a wafer.
  • the clusters of defects are highlighted on the wafer map and cluster IDs (CIDs) are displayed in the defects table 706 of the embodiment of defect summary screen 702 shown in FIG. 7.
  • Clustering may be performed by the wafer defect inspection process tool if a wafer defect inspection process radio button 526 is selected or by the defect management database process 208 if a defect management database process radio button 528 is selected.
  • the user can configure the defect source identifier 100 to use either one of the clustering methods. Selecting the cluster configuration option on the embodiment of defect summary screen shown in FIG.
  • the spatial signature analysis option is selected if a spatial signature analysis check box 530 is selected.
  • the spatial signature analysis calculation is performed by the defect management database process 208.
  • the spatial signature analysis calculates and displays the spatial signature analysis result in the defects table 706 of the defect summary screen 702. If the defect management database process 208 is not available, the spatial signature analysis information will not be displayed.
  • the display options When the embodiment of defect summary screen 702 shown in FIG. 7 is displayed, the display options will read 'calculated' if the spatial signature option was selected. When the spatial signature analysis option is not selected, the display options will provide a suitable indication. If an error occurs calculating the spatial signature analysis result, an error icon will be displayed on the defects summary screen 702 and the error will be logged on an error message page. Selecting the excursion check box 532 of the classification criteria configuration screen 510 executes an excursion option that gives the user the ability to display wafers that exceed the excursion criteria. Selecting the excursion option means that not all wafers will be displayed in real-time but only, e.g., problematic wafers .
  • Selecting the display classes button 404 on any configuration screen portion 410, 510, or 610 causes the display classes screen portion 610 (one embodiment of which is shown in FIG. 6) to be displayed, when the user is at the state indicated by step 1414 in FIG. 14. Selecting the options on the display classes screen allows the user to select which classes are to be displayed. Selecting the done button 406 saves the configuration and performs an initialization function. Selecting the cancel button 408 cancels the configuration changes and returns the display of the defect source identifier to the previously displayed screen, e.g., 'the mode selection screen or the defect summary screen.
  • the on-the-fly (OTF) classes of the wafer defect inspection process are pre-defined. Therefore, the user may select classes to view by checking one or more of the class boxes 612. Specific re-visit classes 614 as well as specific scanning electron microscope-automated defect classification classes 616 are configurable by each customer. The user selects the classes of interest (highlights them) and then selects or deselects the classes using list shift buttons 618 or 620.
  • the defect source identifier 100 only displays the user-selected classes, i.e., those classes appearing in the "selected" lists 622 and 624 as opposed to the "unselected" lsits 626 and 628.
  • the user has completed altering the information contained in any of the configuration screen portions 410, 510, and/or 610, he/she selects the done button 406 of the configuration screen shown in Figs. 4 to 6 to save the edited configuration as shown in decision step 1416 in FIG. 14.
  • the initialize function for the defect summary screen 702 shown in FIG. 7 is displayed.
  • the defect summary screen 702 of the embodiment shown in FIG. 7 is subdivided into four tables including: a general information table 704, a defects table 706, a causes table 708, and a processing tools selection list table 710.
  • the top portion 700 of the screen comprises conventional browser control menus. The user inputs to the defect summary screen 702 are described relative to step 1418 in FIG. 14.
  • the left side of the general information table 704 of the defect summary screen 702 contains such information as the layer, lot and wafer identification and the number of defects of the wafer currently viewed.
  • a defect wafer map graphic 712 occupies the center of this section and shows the location of the defects 713.
  • the right side of the general information section contains the status of the classification options 714 as selected in the configuration screen, as well as a configuration button 720 and a search button 722.
  • the configuration button allows the user to execute decision step 1420 in FIG. 14 to access the configuration screen portions 410, 510, 610 shown respectively in FIGs. 4, 5, and 6 to make alterations to the status of the defect summary screen 702.
  • a global search function is selected by pressing the search button 722.
  • the global search function causes the method 1400 to proceed step 1422.
  • the method invokes a search method and the user is promoted to enter search criteria. See FIG. 13 and the associated description below for a discussion of the search screen.
  • the defect source identifier 100 proceeds to step 1424 wherein the method 1400 searches for wafers that match the search query provided by the user.
  • the defect source identifier displays the defect summary screen for a previously processed wafer that meets the search criteria.
  • the defects table 706 contains information on the classification of the wafer defects.
  • the user selecting the defects details button 726 invokes the defect detail function of step 1426 in FIG. 14. Step 1426 extends the number of rows included in the defects table 706 to include more information such as the precise location and size of the defects.
  • step 1426 the method 1400 continues to step 1428 in which the displayed defect table 706 is expanded.
  • the method 1400 then returns to step 1418.
  • the fields contained in the non-expanded defects table 706 are shown in TABLE 1, and depend on the configured station type.
  • the expanded information contained in TABLE 2 is displayed in the defects table 706: TABLE 2 : Expanded Fields In Defects Table Of Defect Summary Screen
  • the causes table 708 in the embodiment of defect summary table 702 of FIG. 7 reflects case study information from the defect source identifier.
  • a user can select a refresh causes button 709 to invoke step 1430. Pressing the refresh causes button 709 will automatically update the causes table 708 at step 1432 and the causes column 730 of the defects table 706. Following step 1432, the method 1400 returns to step 1418.
  • the refresh causes button 709 helps identify which tools may be responsible for any particular defects.
  • the causes table 708 displays the various classes of defects in a class column 732 shown in TABLE 4 with the number of defects of each class on the current wafer, and the possible causes for the defects.
  • Causes can be ordered in the causes table 708 in alphabetical order, or in any other desired order. For example spatial signature analysis classes can be displayed first then scanning electron microscope-automated defect classification, re-visit, and on-the-fly classes are displayed.
  • the on-the-fly classes may be pre-configured.
  • the scanning electron microscope-automated defect classification and revisit classes are configurable. Each user will have their own set of scanning electron microscope-automatic defect classification and revisit classes.
  • the defect source identifier 100 uses a map file to translate the customer's scanning electron microscope- automatic defect classification and revisit classes that the defect knowledge database process will recognize. A customer's classes will exist in their particular customer knowledge database process 218 if one exists.
  • KLA files that are generated from on-the-fly are displayed distinctly from KLA files that are generated by revisit classes.
  • the KLA files have separate columns in the defect section for on-the-fly and re-visit classes.
  • the wafer defect inspection process will generate the file once it has finished all testing on the lot so that both on-the-fly and revisit results can be stored within one file.
  • the processing tools table 710 in the defect summary screen 702 includes a list of processing tools that the wafer identified in portion 704 was processed with since the last inspection.
  • the tools can be selected to view case studies that apply to a specific defect or a class of defects caused by the tools selected. This helps the user identify which tools may be responsible for the defects.
  • the tools are listed in reverse processing order, from the last tool to process the wafer to the first. Selecting a small tools arrow button 736, located next to the processing tools table 710 title, reverses the order of the tools and displays them in processing order. Any case studies that are relevant for each of the defects on the wafer indicate one specific responsible processing tool that will be highlighted on the screen.
  • the processing tool table fields are shown in TABLE 5.
  • the defect source identifier 100 will execute the defect source identifier database process 214 to determine the last wafer processed by the tool.
  • the defect source identifier 100 executes the manufacturing execution database process 210 to retrieve a sequential list of processing tools that processed the wafer. If any of the " adders, repeaters, spatial signature analysis or clusters from the defect management database process 208 have been configured, the defect source identifier 100 accesses the defect management database process 208 to retrieve the configured information.
  • the defect source identifier 100 then executes the defect knowledge database process 216 to obtain a list of causes for each detected defect. This list of causes is based on either all of the defect's classifications or the configured classifications for a specific defect.
  • the defect source identifier 100 also executes the defect knowledge database process 216 to get a list of causes for each selected defect classification and populates the fields of the causes table 708 of the defect summary screen 702.
  • the defect summary screen 702 is displayed. If the defect source identifier 100 cannot access data for any one of the data collection points described above, an error message is displayed to the user. In addition to modifications of the displayed defect summary screen 702 using, e.g., the configuration function or the search function.
  • the user can cause various defect images to be displayed.
  • the user can select any defects on the wafer map 712 shown in FIG. 7 by, e.g., a mouse select at a suitable location on the wafer map or an alphanumeric selection where each defect is provided a referencing number or letter, to highlight the corresponding defect record on the defects table.
  • the user can select on a particular defect number in a defect # column 740 on the defects table of the defect summary screen 702 to highlight the corresponding defect on the defect map.
  • the row corresponding to the selected defect displays a gallery of images for that defect.
  • an arrow is located beside the titles of some of the fields in the defect summary screen 702. The user selects the arrow beside the field stating the criterion used to sort the table.
  • the defects details associated with the selected defect field are displayed.
  • the user can display additional detail for each defect on the defects table 706 by selecting the defects details button 726.
  • the size and location of certain defects are derived if the defect was reviewed, e.g., from information generated by the scanning electron microscope process.
  • the user can select any class in the cause column 730 of the defects table 706 to select that class and the rows in the defects table 706 and defects on the wafer map 712 that relate to the selected class will be displayed. If an error occurs during the execution of any of the above functions, the execution of the function will be terminated and the original defect data will be displayed. An error icon will appear on the defect summary screen and the error will be logged on the error message page.
  • the defect summary screen 702 displays data from various data sources .
  • the inspection tools of the defect source identifier creates a KLA file after they are done inspecting a lot or wafer and store wafer identification and defect inspection information in this file.
  • the general information section 704 of the defect summary screen 702 is populated with layer, wafer, and lot IDs gathered from the KLA result file.
  • the display options table 714 is populated with the information from the configuration settings.
  • the defects table section 706 contains data that was collected by the metrology tools, e.g. the scanning electron microscope process 206 and the wafer defect inspection process 204. Depending on the station the user is running the defect source identifier from, data from the wafer defect inspection process 204 only, the scanning electron microscope process only if data from the wafer defect inspection process is unavailable, or both processes 204, 206 will be displayed in the defects table section
  • the KLA file produced by each wafer defect inspection tool contains the information to populate the fields shown in TABLE 6 : TABLE 6: Defect Table Fields Of The Defect Summary Screen
  • the cause column 730 in the defects table 706 is based on data retrieved from the defect knowledge database process and the customer knowledge database process with the data from the metrology tools 204, 206 for each defect detected on the wafer.
  • the defect source identifier finds the relevant case histories for each defect by correlating the following search criteria that was input into these systems using on-the-fly, spatial signature analysis, revisit and scanning electron microscope-automated defect classification classes, and one or more processing tools.
  • the spatial signature analysis data is retrieved from the defect management database process 208.
  • the spatial signature analysis data is gathered if the spatial signature analysis option 530 is selected in the select defect classification criteria configuration screen portion 510.
  • the cluster data (CID) is gathered from the defect management database process 208 or KLA file depending on the configuration. If the spatial signature analysis or cluster information cannot be retrieved from the defect management database process 208, then the defects summary screen will be displayed without the information.
  • the causes table 708 of the defect summary screen 702 contains data retrieved from the KLA file, the defect knowledge database process 216 and the customer knowledge database process 218. If the defect source identifier 100 cannot access the defect knowledge database process 216, the defects summary screen is displayed without any causes information.
  • Each defect classification listed in the defects table (which includes input from one or many KLA files) is used to look up the case studies in defect knowledge database process 216.
  • the inputs into the defect knowledge database process 216 to retrieve the cases are tool type and on the fly, spatial signature analysis, revisit and scanning electron microscope-automated defect classifications.
  • Accessing the manufacturing execution database process 210 populates the fields of the processing tools table as shown in TABLE 7. If the tool (e.g., scanning electron microscope process 206 or wafer defect inspection process 204) information cannot be retrieved for a wafer identification or a lot identification, the defect inspection information for the wafer will not be displayed.
  • the tool e.g., scanning electron microscope process 206 or wafer defect inspection process 204
  • defect source identifier accesses the defect source identifier database to retrieve the image file name(s) for the specific defect with respective input and output fields as shown respectively in TABLES 8 and 9.
  • the image file(s) are exported by the defect source identifier along with the KLA file and stored in the defect source identifier file system.
  • the wafer defect inspection process 204 and the scanning electron microscope process 206 generate an image file for each KLA file.
  • the image files may contain all the images for the lot or wafer. If an error occurs locating or retrieving the images, on the error message page. The error may also be stored in an error log file database.
  • the user can select the wafer defect to display as illustrated in step 1440 of FIG. 14.
  • the cause column 730 of the defects table 706 the name of a cause of a defect is displayed only if a matching cause is found. If more than one matching cause is found, the number of causes found is displayed.
  • the user can click on the number to open a new browser window displaying a defect cause selection screen 900 (one embodiment of which is shown in FIG. 9) listing the various cause names in field column 902, cause descriptions in filed column 904 and case images in field column 906 that have historically been found to apply to that specific defect.
  • the user can then click on any name in the cause column 902 to display a detailed case description 1000 of FIG. 10.
  • case image screen 1100 corresponding to images 1102, 1104, 1106 associated with the selected case study.
  • the embodiment of case image screen 1100 shown in FIG. 11 is based on the assumption that the case studies in defect knowledge database process are organized so the defect source identifier 100 can retrieve images generated by the scanning electron microscope process 206, the wafer defect inspection process 204 and other such processes.
  • the fields associated to the expanded defect cause selection screen of FIG. 9 are shown in TABLE 10.
  • the case image screen 1100 of FIG. 11 may access the different screens shown in TABLE 11. If an error occurs displaying any of the case image screens 1100, the defects summary screen will be redisplayed and an error message will be displayed on the error message page and logged in the error lot file.
  • the user can select to compare an image displayed in the case image screen 1100 with a wafer defect case history retrieved by the customer knowledge database process 218 and/or the defect knowledge database process 216 as indicated by steps 1444 and 1446 of FIG. 14.
  • the user selects whether they wish to utilize an image compare screen 1200 as shown in Fig. 12 in decision step 1444. If the answer to decision step 1444 is no, then the method 1400 returns to step 1418. If the answer to decision step
  • step 1444 the method 1400 continues to step 1446 wherein the user selects the particular case history and metrology tool process 204 or 206 to display in the image compare screen 1200.
  • the method 1400 displays an image compare screen 1200 of FIG. 12 (step 1448) , that allows the user to compare a defect image selected from a defect image screen 800 including fields shown in TABLE 12.
  • the embodiment of defect image screen 800 shown in FIG. 8 displays a plurality of case images including images sets 802, 804, 806 from such tools as the scanning electron microscope process 206 and/or the optical wafer defect inspection database process 204 similar to as displayed on the case image screen 1100 shown in FIG. 11.
  • the user may selects one image from each of the types of processes 204 and/or 206 or other.
  • the user selects an image in either the case image screen 1100 or the defect image screen 800 and a new browser window displaying the image compare screen 1200 is launched with an enlarged version of the image 1202 selected.
  • the user can then select an image from the case image screen 1100 and an enlarged version of the image 1204 will be displayed adjacent to the previous image selected".
  • the user can select any of the possible causes included in the causes table 708 of the defect summary screen 702 to display a new possible cause selection browser window containing detailed information about the specific case from the defect source identifier.
  • the inputs to the possible cause selection screen is shown in TABLE 13.
  • the defects summary screen 700 will be redisplayed and an error message will be displayed on an error message page and logged in the error log file database .
  • Each wafer undergoing processing is sequentially processed by a plurality of processing tools. Processing of the wafer by any of these processing tools may cause certain defects on the wafer.
  • a list of the processing tools that have processed a wafer is provided in the processing tool table 710 located at the bottom of one embodiment of the defect summary screen 702 in FIG. 7. The user can select one or more of these processing tools and click the refresh causes button 709 to refresh the defects table 706 and the causes table 708.
  • the refreshed defects table 706 summarize the causes that may apply to each processing tool used to process the particular wafer. By default, the causes table 708 shows all possible causes for the wafer's defects.
  • the inputs to the processing tool includes the fields shown in TABLE 15.
  • the outputs from the processing tool includes those fields shown in TABLE 16.
  • a global search button 722 is located to the right of the wafer map 712 in the embodiment of defect summary screen 702 shown in FIG. 7.
  • This global search button 722 allows the users to display defect inspection information for a specific wafer using a wafer search screen 1300 shown in FIG. 13.
  • the global search button 722 allows the users of search stations in the wafer search screen 1300 to display the defect information for a specific wafer on the existing browser as indicated in steps 1422 and 1424 of FIG. 14, which is not updated in real-time.
  • the user can specify various parameters in the wafer search screen 1300 to narrow the search. If multiple wafers match the search criteria, a list of matching wafers will be displayed for the user.
  • the user selects a specific wafer for display in the defect summary screen 702.
  • the user will enter in the wafer search screen 1300 the date/time ranges of wafers in fields 1302A, 1302B, 1302C and 1302D.
  • fields 1304A and 1304B the inspection/review tool of interest and a processing tool of interest in fields 1306A and 1306B.
  • the user also has the option of retrieving a list of wafers that are in excursion by selecting excursion check field 1308 or all wafers that match the above filters by not selecting excursion check field 1308.
  • the user selects a find wafers button 1310 to retrieve the list of wafer lots matching the selected filter criteria.
  • the user selects a specific lot of interest by selected a lot identification option in a lot identification pull-down menu 1314.
  • the list of wafers is then populated with wafer IDs in a wafer identification field that are contained in the selected lot in a wafer identification pull-down menu 1312.
  • the user selects the wafer of interest.
  • the list of layer IDs is then populated by selecting one wafer identification option in the wafer identification pull-down menu 1312, then the layer identification pull down menu 1316 is populated. Once all the selections are complete, the user selects a complete button 1320 to retrieve the analysis data and to display the results in the defect summary screen 702 shown in FIG. 7.
  • the global " wafer search table includes those fields shown in TABLE 17.
  • the defect source identifier has error handling capabilities that display errors and informational messages to the user using a message page.
  • the errors are displayed to the user in an error message screen and the error details are logged in a log file database.
  • Each error message in the log file database includes a date/timestamp.
  • the defect source identifier can't access tool information using the manufacturing execution database process 210 for a specific lot due to errors accessing them, the lot of wafers will not be displayed. The errors will be displayed on the error message page and logged in the error log file database. The wafer search screen 1300 may then be displayed instead of the defect summary screen 702.
  • the defect summary screen will be displayed without the case information.
  • the error will be displayed on the error message page and logged in the error log file database.
  • the defect source identifier 100 is unable to access adders, repeaters, spatial signature analysis or cluster information from the defect management database process 208, the defect summary screen is displayed without this information.
  • the error will be displayed on the error message page and logged in the error log file database.
  • the defect source identifier is unable to access the defect source identifier historical database, this error will be displayed on the error message page and logged in the error log file database.
  • the defect source identifier will thereupon exit .
  • defect source identifier 100 shown in FIG. 16 has a three tiered client-server architecture 1600 including a client ' tier 1602, a middle tier 1604, and a data tier 1606.
  • the client tier 1602 provides user interfaces for defect source identifier.
  • the client tier consists of browser software commonly utilized to provide an Internet connection to client computers 105 located at the defect source identifier client 104.
  • the middle tier comprises MICROSOFT ® Internet Information Server and MICROSOFT ® Transaction Server.
  • the middle tier implements the business rules for the client application, manages transactions with the database processes 210, 212, 214, 216, 218, and 220, and serves web pages to the browser clients.
  • the middle tier processes resides on the defect source identifier server 106.
  • One embodiment of the middle tier processes comprise, e.g., a WINDOWS NT ® server, a SQL server database, and the defect knowledge database process 216.
  • the middle tier processes may interact with databases and other data source products that reside on other servers that are typically operated outside the scope of the defect source identifier 100.
  • FIG. 15 is a flow diagram of one embodiment of a method 1500 performed by the defect source identifier 100 of Fig. 1 to identify defects on wafers. To best understand the operation of this method 1500, the reader should refer to FIGS. 1 and 2 while reading the following description of FIG. 15.
  • the method 1500 starts with step 1502 in which the wafer processing system 102 processes a wafer in one of the plurality of process cells 103.
  • the wafer is then displaced, typically using robots 121, to a metrology cell 124 as shown in step 1504.
  • the wafer is inspected for defects using a metrology tool such as a scanning electron microscope process 206 or a wafer defect inspection process 204 while the wafer is in the metrology cell 124.
  • the wafer defect inspection process typically stores the defect inspection information as a KLA file.
  • the scanning electron microscope process shown as 206 in the embodiment in FIG. 2, typically stores its defect inspection information as a KLA file.
  • the method 1500 transmits the defect inspection information from the wafer processing system 102 to the client computer 105 in step 1508.
  • the defect inspection information is originally stored in the defect management database process 208 shown in FIG. 2.
  • the method 1500 continues to decision step 1510 that determines whether the customer knowledge database process 218 exists. If the answer to the decision step 1510 is yes, then the method 1500 continues to step 1514.
  • step 1514 defect inspection information obtained from the metrology tools (e.g. the scanning electron microscope process 206 or the wafer defect inspection process 204) is compared to the defect source information representing the case histories stored in the customer knowledge database process 218.
  • a source of the wafer defect is determined by comparing the defect to the case histories to the defect source information corresponding to prior defects stored in the customer knowledge system.
  • the stored contents of the customer knowledge database process 218 is typically smaller, and includes fewer defect history cases, than the defect knowledge system database process 216. However, if any particular defect source identifier client 104 is operated for a considerable time, the size and utility of the customer knowledge database process will increase.
  • the defect source information and the defect inspection information are both displayed on a graphical user interface (GUI) associated with a client computer 105 in an image compare screen step 1516.
  • the defect inspection information in step 1516 is similar to the image compare screen shown in FIG. 12.
  • the method 1500 continues to step 1518, in which the defect source identifier client 104 determines if the displayed defect source information is acceptable in that the cause of the case history defect in the defect source information is actually the cause of the present defect in the defect inspection information. Such determination of • acceptability is determined either based upon a skilled user determining that the displayed defect is similar to the case history defect, or a correlation program in the client computer 105 providing the same determination. If the query in step 1518 is answered yes, then method 1500 continues to decision step 1519.
  • the defect source identifier 100 determines whether the customer knowledge database process supports the defect knowledge database process. If a particular customer knowledge database process 218 does not support the defect knowledge database process 216, as determined by step 1519, then the customer knowledge database process 218 will not be granted access to the defect knowledge database process 216, and the method 1500 continues to step 1521.
  • step 1521 the operating of the wafer processing system 102 is modified to remedy the processing situation that created the defects on the wafer according to the recommended operation changes to the wafer processing system included in the case history. This modification of the operation of the wafer processing system can be performed manually by a user viewing the defect source information, or automatically by the client computer 105 altering the operation of the wafer processing system. Following step 1521, the method 1500 terminates.
  • step 1520 the user accesses new source defect information. This may be provided by a user selecting new defect source information that contains different defect sources and solutions, one of the multiple causes shown in 708 in FIG. 7. If the user wants to select another cause for the defect that includes further defect source information, then the method 1500 continues to loop back to decision step 1512. If the answer to decision step 1519 is yes, the method continues to step 1522. Also, if the answer in decision step 1510 is yes, then the method 1500 continues to step 1522. In step 1522, the selected defect for each defect on the wafer is sent to the server computer 107. There may be a plurality of selected defects associated with each wafer. The method 1500 continues to step 1524, in which the defect inspection information is compared to defect source information stored in the defect knowledge database process 216 that is typically stored in the memory 162B of the defect source identifier server 106b.
  • the method 1500 continues to step 1526 wherein the defect knowledge database process is accessed by the defect source identifier server 106 to derive potential causes based on defect inspection information for each defect located on the wafer.
  • a compilation of selected defect cause information is transmitted back to the defect source identifier client 104.
  • the method 1500 continues to step 1528 to utilize the selected defect cause information.
  • the utilization of the defect cause information includes displaying selected causes from the defect cause information on the defect summary screen 702 shown in FIG 7.
  • the method 1500 then continues to step 1530 in which a user, proximate the defect source identifier client 104, interfaces with the client computer 105 to select one defect cause from the selected defect cause information for each particular wafer defect.
  • the defect causes are listed, e.g., in the cause section 708 in the embodiment of defect summary screen shown in Fig. 7.
  • the user selects one of the defect sources listed in the cause section by, e.g. "clicking on" that particular cause.
  • the method 1500 then continues to step 1534 wherein the selected defect cause is transmitted from the defect source identifier client 104 to the defect source identifier server 106.
  • the method 1500 continues to step 1536 in which the defects source information, including the selected defect cause, is generated in the server computer in response to the selected defect cause transmitted in step 1534.
  • the defect source information generated in step 1536 is included as part of the defect knowledge database process 216 included in the embodiment of defect source identifier server 106 shown in FIG. 1.
  • the method 1500 continues to step 1538 in which the defect source information is transmitted to the defect source identifier client 104.
  • step 1540 in which defect source information is displayed as an image or over an image compare screen, one embodiment shown in FIG. 12, along with the defect inspection information derived in step 1506.
  • the image compare screen is displayed at the defect source identifier client 104.
  • the user can view the displayed images over the image compare screen, and determine if he/she is satisfied with the correlation between the defect shown in the defect source information and the defect shown in the defect inspection information contained in the image compare screen.
  • the client computer 105 running a correlation program can provide a similar determination.
  • the method 1500 continues to step 1542 wherein the defect cause is accepted, or not accepted, by the user or the defect source identifier client 104 indicating whether the defect source information is sufficiently closely correlated to the defect inspection information. If the answer to the decision step 1542 is yes, then the method 1500 continues to step 1544. By comparison, if the defect cause is not accepted by the user at step 1542, then the method continues looping back to step 1530.
  • step 1544 the user interfaces with the defect source identifier client 104 to manually correct the operation of the wafer processing system according to the displayed defect source information.
  • the defect source identifier client 104 automatically applies a solution displayed by the defect source information over the display at the defect source identifier client 104 by altering the operation of the wafer processing system 102 to remedy the source of the defect. For example, if the source of the defect (the defect cause) is that the process cell is dirty, the process cell will be operated in a clean mode for a prescribed duration.
  • the conditions in the wafer processing system 102 e.g., the process cell 103, will be altered.
  • These corrections to the conditions in the wafer processing system 102 to limit the defect occurrences can be performed automatically, or they can be input by a system operator altering the settings or conditions at the wafer processing system 102.
  • a larger defect knowledge database process indicated as 216 in FIG. 2, that is stored in the defect source identifier server 106, may be accessed by the user. Following step 1544, the method 1500 terminates at step 1545.

Landscapes

  • Engineering & Computer Science (AREA)
  • Manufacturing & Machinery (AREA)
  • Computer Hardware Design (AREA)
  • Microelectronics & Electronic Packaging (AREA)
  • Power Engineering (AREA)
  • Testing Or Measuring Of Semiconductors Or The Like (AREA)
PCT/US2001/031029 2000-10-02 2001-10-02 Defect source identifier WO2002029392A2 (en)

Priority Applications (2)

Application Number Priority Date Filing Date Title
EP01977451A EP1322941A2 (de) 2000-10-02 2001-10-02 Identifizierer für defektquellen
KR1020027007062A KR20020063582A (ko) 2000-10-02 2001-10-02 결함 소오스 탐지기

Applications Claiming Priority (4)

Application Number Priority Date Filing Date Title
US23729700P 2000-10-02 2000-10-02
US60/237,297 2000-10-02
US09/905,607 2001-07-13
US09/905,607 US6701259B2 (en) 2000-10-02 2001-07-13 Defect source identifier

Publications (2)

Publication Number Publication Date
WO2002029392A2 true WO2002029392A2 (en) 2002-04-11
WO2002029392A3 WO2002029392A3 (en) 2002-06-06

Family

ID=26930528

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/US2001/031029 WO2002029392A2 (en) 2000-10-02 2001-10-02 Defect source identifier

Country Status (5)

Country Link
EP (1) EP1322941A2 (de)
KR (1) KR20020063582A (de)
CN (1) CN1398348A (de)
TW (1) TWI256468B (de)
WO (1) WO2002029392A2 (de)

Cited By (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2004059723A2 (en) * 2002-12-20 2004-07-15 Applied Materials, Inc. System to identify a wafer manufacturing problem and method therefor
EP1376681A3 (de) * 2002-06-26 2005-09-28 Agilent Technologies Datenanalysegerät
WO2018134158A1 (en) * 2017-01-18 2018-07-26 Asml Netherlands B.V. Knowledge recommendation for defect review

Families Citing this family (27)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US7208328B2 (en) * 2004-03-16 2007-04-24 Macronix International Co., Ltd. Method and system for analyzing defects of an integrated circuit wafer
US8532949B2 (en) * 2004-10-12 2013-09-10 Kla-Tencor Technologies Corp. Computer-implemented methods and systems for classifying defects on a specimen
US7570796B2 (en) 2005-11-18 2009-08-04 Kla-Tencor Technologies Corp. Methods and systems for utilizing design data in combination with inspection data
JP4882505B2 (ja) * 2006-05-19 2012-02-22 東京エレクトロン株式会社 異物分布パターンの照合方法及びその装置
US9659670B2 (en) 2008-07-28 2017-05-23 Kla-Tencor Corp. Computer-implemented methods, computer-readable media, and systems for classifying defects detected in a memory device area on a wafer
KR101674698B1 (ko) * 2009-02-13 2016-11-09 케이엘에이-텐코어 코오포레이션 웨이퍼 상의 결함들 검출
US8775101B2 (en) 2009-02-13 2014-07-08 Kla-Tencor Corp. Detecting defects on a wafer
TWI386643B (zh) * 2009-04-17 2013-02-21 Chipmos Technologies Inc 晶圓缺陷標示系統
US8781781B2 (en) 2010-07-30 2014-07-15 Kla-Tencor Corp. Dynamic care areas
US9170211B2 (en) 2011-03-25 2015-10-27 Kla-Tencor Corp. Design-based inspection using repeating structures
US9087367B2 (en) 2011-09-13 2015-07-21 Kla-Tencor Corp. Determining design coordinates for wafer defects
US8831334B2 (en) 2012-01-20 2014-09-09 Kla-Tencor Corp. Segmentation for wafer inspection
US8826200B2 (en) 2012-05-25 2014-09-02 Kla-Tencor Corp. Alteration for wafer inspection
US9189844B2 (en) 2012-10-15 2015-11-17 Kla-Tencor Corp. Detecting defects on a wafer using defect-specific information
US9053527B2 (en) 2013-01-02 2015-06-09 Kla-Tencor Corp. Detecting defects on a wafer
US9134254B2 (en) 2013-01-07 2015-09-15 Kla-Tencor Corp. Determining a position of inspection system output in design data space
US9311698B2 (en) 2013-01-09 2016-04-12 Kla-Tencor Corp. Detecting defects on a wafer using template image matching
KR102019534B1 (ko) 2013-02-01 2019-09-09 케이엘에이 코포레이션 결함 특유의, 다중 채널 정보를 이용한 웨이퍼 상의 결함 검출
US9865512B2 (en) 2013-04-08 2018-01-09 Kla-Tencor Corp. Dynamic design attributes for wafer inspection
US9310320B2 (en) 2013-04-15 2016-04-12 Kla-Tencor Corp. Based sampling and binning for yield critical defects
US10114368B2 (en) * 2013-07-22 2018-10-30 Applied Materials Israel Ltd. Closed-loop automatic defect inspection and classification
CN103606529B (zh) * 2013-10-23 2016-08-24 上海华力微电子有限公司 一种提升缺陷分类准确度的方法及装置
US9286675B1 (en) 2014-10-23 2016-03-15 Applied Materials Israel Ltd. Iterative defect filtering process
WO2018213487A1 (en) * 2017-05-17 2018-11-22 Applied Materials Israel Ltd. Method, computer program product and system for detecting manufacturing process defects
US10620135B2 (en) * 2017-07-19 2020-04-14 Kla-Tencor Corp. Identifying a source of nuisance defects on a wafer
TWI757907B (zh) * 2020-10-07 2022-03-11 財團法人國家實驗研究院 一種在真空環境中叢集式即時線上製程暨分析傳輸系統
TWI815195B (zh) * 2021-10-13 2023-09-11 瑞霸生技股份有限公司 電化學檢測之起始時間判斷方法

Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US5761064A (en) * 1995-10-06 1998-06-02 Advanced Micro Devices, Inc. Defect management system for productivity and yield improvement
EP0910123A1 (de) * 1996-03-19 1999-04-21 Hitachi, Ltd. Prozesssteuerungssystem
WO1999059200A1 (en) * 1998-05-11 1999-11-18 Applied Materials, Inc. Fab yield enhancement system

Patent Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US5761064A (en) * 1995-10-06 1998-06-02 Advanced Micro Devices, Inc. Defect management system for productivity and yield improvement
EP0910123A1 (de) * 1996-03-19 1999-04-21 Hitachi, Ltd. Prozesssteuerungssystem
WO1999059200A1 (en) * 1998-05-11 1999-11-18 Applied Materials, Inc. Fab yield enhancement system

Cited By (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
EP1376681A3 (de) * 2002-06-26 2005-09-28 Agilent Technologies Datenanalysegerät
WO2004059723A2 (en) * 2002-12-20 2004-07-15 Applied Materials, Inc. System to identify a wafer manufacturing problem and method therefor
WO2004059723A3 (en) * 2002-12-20 2005-04-07 Applied Materials Inc System to identify a wafer manufacturing problem and method therefor
WO2018134158A1 (en) * 2017-01-18 2018-07-26 Asml Netherlands B.V. Knowledge recommendation for defect review
TWI709182B (zh) * 2017-01-18 2020-11-01 荷蘭商Ａｓｍｌ荷蘭公司 用於知識推薦及缺陷分類之伺服器及其方法
US11650576B2 (en) 2017-01-18 2023-05-16 Asml Netherlands B.V. Knowledge recommendation for defect review

Also Published As

Publication number Publication date
WO2002029392A3 (en) 2002-06-06
TWI256468B (en) 2006-06-11
CN1398348A (zh) 2003-02-19
EP1322941A2 (de) 2003-07-02
KR20020063582A (ko) 2002-08-03

Similar Documents

Publication Publication Date Title
US6701259B2 (en) Defect source identifier
WO2002029392A2 (en) Defect source identifier
US6744266B2 (en) Defect knowledge library
JP5107506B2 (ja) 欠陥分析器
KR100858861B1 (ko) 데이터 분석 방법 및 장치
KR100873114B1 (ko) 자동 센서 설치 방법 및 장치
US6965895B2 (en) Method and apparatus for analyzing manufacturing data
US7401066B2 (en) Correlation of end-of-line data mining with process tool data mining
KR100986789B1 (ko) 툴 성능을 모니터링하는 장치 및 방법
CN100375086C (zh) 用于动态传感器配置和运行时间执行的方法和设备
CN100407215C (zh) 用于监视和控制半导体生产过程的方法和装置
KR100682641B1 (ko) 반도체 디바이스의 제조 공정에 대한 특성화 데이터를 발생시키기 위한 명령어를 수록한 컴퓨터 판독가능한 매체 및 상기 특성화 데이터를 발생시키기 위한 시스템
US7577486B2 (en) Quality improvement system
US8295969B2 (en) Combinatorial processing management system
US6446021B1 (en) Method and apparatus to display processing parameter
US20030135295A1 (en) Defect source identifier with static manufacturing execution system
Lee et al. Yield analysis and data management using Yield Manager/sup TM
Nag et al. Simulation of yield/cost learning curves with Y4
US20080189245A1 (en) Accessing data from diverse semiconductor manufacturing applications
JP2001155979A (ja) 製造情報管理システム
Hodges et al. Computer integrated manufacturing (semiconductor processing)
US7203707B2 (en) System and method for knowledge asset acquisition and management
Merino et al. SmartBit/sup TM: bitmap to defect correlation software for yield improvement
Lee Advanced yield enhancement: integrated yield analysis
US20140188264A1 (en) Workflow Manager and Bar Coding System for Processing of Samples/Substrates in HPC (High Productivity Combinatorial) R&D Environment

Legal Events

Date Code Title Description
AK Designated states

Kind code of ref document: A2

Designated state(s): CN JP KR SG

AL Designated countries for regional patents

Kind code of ref document: A2

Designated state(s): AT BE CH CY DE DK ES FI FR GB GR IE IT LU MC NL PT SE TR

WWE Wipo information: entry into national phase

Ref document number: 2001977451

Country of ref document: EP

WWE Wipo information: entry into national phase

Ref document number: 018029922

Country of ref document: CN

WWE Wipo information: entry into national phase

Ref document number: 1020027007062

Country of ref document: KR

AK Designated states

Kind code of ref document: A3

Designated state(s): CN JP KR SG

AL Designated countries for regional patents

Kind code of ref document: A3

Designated state(s): AT BE CH CY DE DK ES FI FR GB GR IE IT LU MC NL PT SE TR

WWP Wipo information: published in national office

Ref document number: 1020027007062

Country of ref document: KR

121 Ep: the epo has been informed by wipo that ep was designated in this application
WWP Wipo information: published in national office

Ref document number: 2001977451

Country of ref document: EP

NENP Non-entry into the national phase

Ref country code: JP