Classifications

• • G—PHYSICS
  • G01—MEASURING; TESTING
  • G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
  • G01N21/00—Investigating or analysing materials by the use of optical means, i.e. using sub-millimetre waves, infrared, visible or ultraviolet light
  • G01N21/84—Systems specially adapted for particular applications
  • G01N21/88—Investigating the presence of flaws or contamination
  • G01N21/95—Investigating the presence of flaws or contamination characterised by the material or shape of the object to be examined
  • G01N21/9501—Semiconductor wafers
• • G—PHYSICS
  • G01—MEASURING; TESTING
  • G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
  • G01N21/00—Investigating or analysing materials by the use of optical means, i.e. using sub-millimetre waves, infrared, visible or ultraviolet light
  • G01N21/84—Systems specially adapted for particular applications
  • G01N21/88—Investigating the presence of flaws or contamination
  • G01N21/95—Investigating the presence of flaws or contamination characterised by the material or shape of the object to be examined
  • G01N21/9501—Semiconductor wafers
  • G01N21/9505—Wafer internal defects, e.g. microcracks
• • G—PHYSICS
  • G01—MEASURING; TESTING
  • G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
  • G01N21/00—Investigating or analysing materials by the use of optical means, i.e. using sub-millimetre waves, infrared, visible or ultraviolet light
  • G01N21/84—Systems specially adapted for particular applications
  • G01N21/88—Investigating the presence of flaws or contamination
  • G01N21/95—Investigating the presence of flaws or contamination characterised by the material or shape of the object to be examined
  • G01N21/956—Inspecting patterns on the surface of objects
• • G—PHYSICS
  • G01—MEASURING; TESTING
  • G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
  • G01N23/00—Investigating or analysing materials by the use of wave or particle radiation, e.g. X-rays or neutrons, not covered by groups G01N3/00 – G01N17/00, G01N21/00 or G01N22/00
  • G01N23/02—Investigating or analysing materials by the use of wave or particle radiation, e.g. X-rays or neutrons, not covered by groups G01N3/00 – G01N17/00, G01N21/00 or G01N22/00 by transmitting the radiation through the material
  • G01N23/04—Investigating or analysing materials by the use of wave or particle radiation, e.g. X-rays or neutrons, not covered by groups G01N3/00 – G01N17/00, G01N21/00 or G01N22/00 by transmitting the radiation through the material and forming images of the material
• • G—PHYSICS
  • G01—MEASURING; TESTING
  • G01R—MEASURING ELECTRIC VARIABLES; MEASURING MAGNETIC VARIABLES
  • G01R31/00—Arrangements for testing electric properties; Arrangements for locating electric faults; Arrangements for electrical testing characterised by what is being tested not provided for elsewhere
  • G01R31/28—Testing of electronic circuits, e.g. by signal tracer
  • G01R31/282—Testing of electronic circuits specially adapted for particular applications not provided for elsewhere
  • G01R31/2831—Testing of materials or semi-finished products, e.g. semiconductor wafers or substrates
• • G—PHYSICS
  • G01—MEASURING; TESTING
  • G01R—MEASURING ELECTRIC VARIABLES; MEASURING MAGNETIC VARIABLES
  • G01R31/00—Arrangements for testing electric properties; Arrangements for locating electric faults; Arrangements for electrical testing characterised by what is being tested not provided for elsewhere
  • G01R31/28—Testing of electronic circuits, e.g. by signal tracer
  • G01R31/302—Contactless testing
  • G01R31/308—Contactless testing using non-ionising electromagnetic radiation, e.g. optical radiation
  • G01R31/311—Contactless testing using non-ionising electromagnetic radiation, e.g. optical radiation of integrated circuits
• • G—PHYSICS
  • G06—COMPUTING OR CALCULATING; COUNTING
  • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
  • G06T7/00—Image analysis
  • G06T7/0002—Inspection of images, e.g. flaw detection
  • G06T7/0004—Industrial image inspection
  • G06T7/001—Industrial image inspection using an image reference approach
• • G—PHYSICS
  • G01—MEASURING; TESTING
  • G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
  • G01N21/00—Investigating or analysing materials by the use of optical means, i.e. using sub-millimetre waves, infrared, visible or ultraviolet light
  • G01N21/01—Arrangements or apparatus for facilitating the optical investigation
  • G01N2021/0106—General arrangement of respective parts
  • G01N2021/0118—Apparatus with remote processing
  • G01N2021/0137—Apparatus with remote processing with PC or the like
• • G—PHYSICS
  • G01—MEASURING; TESTING
  • G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
  • G01N21/00—Investigating or analysing materials by the use of optical means, i.e. using sub-millimetre waves, infrared, visible or ultraviolet light
  • G01N21/84—Systems specially adapted for particular applications
  • G01N21/88—Investigating the presence of flaws or contamination
  • G01N21/8851—Scan or image signal processing specially adapted therefor, e.g. for scan signal adjustment, for detecting different kinds of defects, for compensating for structures, markings, edges
  • G01N2021/8883—Scan or image signal processing specially adapted therefor, e.g. for scan signal adjustment, for detecting different kinds of defects, for compensating for structures, markings, edges involving the calculation of gauges, generating models
• • G—PHYSICS
  • G01—MEASURING; TESTING
  • G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
  • G01N21/00—Investigating or analysing materials by the use of optical means, i.e. using sub-millimetre waves, infrared, visible or ultraviolet light
  • G01N21/84—Systems specially adapted for particular applications
  • G01N21/88—Investigating the presence of flaws or contamination
  • G01N21/8851—Scan or image signal processing specially adapted therefor, e.g. for scan signal adjustment, for detecting different kinds of defects, for compensating for structures, markings, edges
  • G01N2021/8887—Scan or image signal processing specially adapted therefor, e.g. for scan signal adjustment, for detecting different kinds of defects, for compensating for structures, markings, edges based on image processing techniques
• • G—PHYSICS
  • G01—MEASURING; TESTING
  • G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
  • G01N2223/00—Investigating materials by wave or particle radiation
  • G01N2223/60—Specific applications or type of materials
  • G01N2223/646—Specific applications or type of materials flaws, defects
  • G01N2223/6462—Specific applications or type of materials flaws, defects microdefects
• • G—PHYSICS
  • G06—COMPUTING OR CALCULATING; COUNTING
  • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
  • G06T2207/00—Indexing scheme for image analysis or image enhancement
  • G06T2207/10—Image acquisition modality
  • G06T2207/10056—Microscopic image
  • G06T2207/10061—Microscopic image from scanning electron microscope
• • G—PHYSICS
  • G06—COMPUTING OR CALCULATING; COUNTING
  • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
  • G06T2207/00—Indexing scheme for image analysis or image enhancement
  • G06T2207/20—Special algorithmic details
  • G06T2207/20081—Training; Learning
• • G—PHYSICS
  • G06—COMPUTING OR CALCULATING; COUNTING
  • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
  • G06T2207/00—Indexing scheme for image analysis or image enhancement
  • G06T2207/20—Special algorithmic details
  • G06T2207/20084—Artificial neural networks [ANN]
• • G—PHYSICS
  • G06—COMPUTING OR CALCULATING; COUNTING
  • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
  • G06T2207/00—Indexing scheme for image analysis or image enhancement
  • G06T2207/30—Subject of image; Context of image processing
  • G06T2207/30108—Industrial image inspection
  • G06T2207/30148—Semiconductor; IC; Wafer

Definitions

• the described embodiments relate to systems for specimen inspection, and more particularly to semiconductor wafer inspection modalities.
• processing steps applied to a substrate or wafer The various features and multiple structural levels of the semiconductor devices are formed by these processing steps. For example, lithography among others is one semiconductor fabrication process that involves generating a pattern on a semiconductor wafer. Additional examples of semiconductor fabrication processes include, but are not limited to, chemical-mechanical polishing, etch, deposition, and ion implantation. Multiple semiconductor devices may be
• Inspection processes are used at various steps during a semiconductor manufacturing process to detect defects on wafers to promote higher yield. As design rules and process windows continue to shrink in size, inspection systems are required to capture a wider range of physical defects while maintaining high throughput.
• Inspection systems such as unpatterned wafer inspection systems and patterned wafer inspection systems illuminate and inspect a wafer for undesired defects.
• the minimum defect size that must be detected continues to shrink in size .
• memory architectures are transitioning from two dimensional floating-gate architectures to fully three dimensional geometries.
• film stacks and etched structures are very deep (e.g., up to six micrometers in depth, or more) . Such high aspect ratio structures create challenges for patterned wafer
• electronic tests are employed to detect defects buried within three dimensional structures.
• multiple device layers must be fabricated before electronic tests are performed.
• defects cannot be detected early in the production cycle.
• electronic tests are prohibitively expensive to perform, particularly during research and development and ramp phases of the production process, where rapid assessment of defects is critical.
• defects buried within three dimensional structures can be detected based on x-ray based measurement techniques.
• an x-ray diftractive measurement system or a coherent x-ray imaging system may be employed to detect buried defects.
• X-ray based measurement techniques For example, an x-ray diftractive measurement system or a coherent x-ray imaging system may be employed to detect buried defects.
• X-ray based measurement system or a coherent x-ray imaging system may be employed to detect buried defects.
• EBI electron beam inspection
• EBI electron beam inspection
• EBI is employed directly to detect defects buried within three dimensional structures.
• EBI is extremely limited in its ability to detect defects beyond a depth of approximately one micrometer.
• EBI is limited to depths that are far less than one micrometer (e.g., less than fifty nanometers) . This limitation is due to practical limits on electron dosage before sample distortion or destruction occurs.
• EBI is limited in its effectiveness as a defect detection tool for thick, three dimensional structures.
• confocal optical inspection is employed at different depths of focus. Confocal imaging eliminates spurious or nuisance optical signals from structures above and below the focal plane. The confocal optical inspection technique is described in further detail in U.S. Patent Publication No. 2014/0300890, which is incorporated herein by reference in its entirety.
• a rotating illumination beam is employed to detect buried defects in relatively thick layers. Optical inspection utilizing a rotating illumination beam is described in further detail in U.S. Patent Publication No. 2014/0268117, which is incorporated herein by reference in its entirety.
• Patent No. 8,912,495 which is incorporated herein by reference it its entirety.
• defect discovery and inspection recipe optimization are based on the inspection of planar, two-dimensional structures.
• An optical inspection tool is employed to measure a large number of defects of interest
• DOI associated with two dimensional structures (e.g., less than one micrometer thick) located at the surface of the wafer.
• the DOI detected by the optical inspection tool are verified by inspecting the identified DOI with a scanning electron microscopy (SEM) tool. This is commonly referred to as SEM review.
• SEM scanning electron microscopy
• an inspection recipe for the optical inspection tool is formulated that maximizes the capture rate of real DOI and minimizes the capture rate of nuisance defects.
• SEM has a very limited penetration depth. Thus, SEM review is only effective for the measurement of defects at or very near the surface of the structure under inspection. To verify defects buried within three dimensional
• the wafer must be de-processed to uncover the buried defects. Wafer de-processing is time consuming and destroys the wafer by removing layers to reveal DOI
• optical inspection of three dimensional structures is much larger than for optical inspection of two dimensional structures because detecting defects of buried structures is much more difficult and depends heavily on the structure itself.
• optical inspection of three dimensional structures is much larger than for optical inspection of two dimensional structures because detecting defects of buried structures is much more difficult and depends heavily on the structure itself.
• inspection of three dimensional structures is based on through focus measurements (i.e., three dimensional images collected at multiple depths through the structure) that involve large amounts of data. Storing and processing excessively large numbers of three dimensional images is impractical .
• the three dimensional volume of a semiconductor wafer subject to defect discovery
• an inspection system After determining one or more focus planes or a focus range, an inspection system records image patches associated with defect locations at the one or more focus planes or focus range, rather than throughout the entire depth of the structure. In this manner, the amount of recorded data associated with defect discovery is limited to a subset of depths. The recorded data are employed during subsequent defect verification and recipe
• a wafer level defect signature is a wafer defect map that illustrates regions of the wafer area where defects are more highly or less highly concentrated.
• a wafer level defect signature includes any indication of wafer defects expressed across the entire wafer area under inspection.
• the number of optical modes under consideration is reduced based on measured defect signal to noise ratio.
• the signal to noise ratio associated with each selected optical mode is analyzed at one or more focus planes or focus levels. The optical modes with the highest signal to noise ratio are selected for further consideration, and the other optical modes are discarded.
• the number of optical modes under consideration is reduced based on SEM review of defects without de-processing.
• SEM review measurements are performed to verify defects of interest at the surface or even slightly below the surface.
• high energy SEM is utilized to review defects buried within a structure under consideration. Defects verified by SEM review are compared to optical inspection results and the optical modes with the highest capture rates of verified defects, and the fewest capture rate of nuisance defects, are selected for further consideration.
• verified defect images/features are mapped to corresponding defects identified by optical inspection.
• the verified defects and the recorded three dimensional images are employed to train a nuisance filter and optimize the measurement recipe.
• defect information associated with defects identified by optical inspection is sampled to generate a diversity set of defects of interest (DOIs) .
• DOIs defects of interest
• the defects are binned and a few defects are selected from each bin to generate the diversity set of DOIs.
• Defect verification measurements are performed on the diversity set of DOIs.
• Defect verification data from the diversity set of DOIs, any other set of verified defects, or a combination thereof are mapped to the saved through focus defect image patches and corresponding feature vectors. The defect verification data and the corresponding through focus defect image patches and corresponding feature vectors are employed to train a nuisance elimination filter.
• the trained nuisance elimination filter is applied to defect images associated with each optics mode under consideration. In this manner, defect detection is
• Detection threshold values associated with each optical mode are adjusted to achieve a desirable nuisance rate.
• the optical mode that achieves the best defect signature matching and real defect capture rate is selected for implementation as the production measurement recipe for the measurement application under consideration.
• FIG. 1 depicts an illustration of a 3D NAND
• FIG. 2 is a simplified schematic view of one
• an optical inspection system 100 configured to perform discovery of defects of interest (DOI) on semiconductor wafers based on three-dimensional images.
• DOE defects of interest
• FIG. 3 is a simplified schematic view of one
• FIG. 4 depicts a plot 190 of a cross-sectional view of a measured three dimensional image illustrating a peak signal near a focus offset of -0.5 micrometers.
• FIG. 5 depicts a plot 191 of another cross-sectional view of the measured three dimensional image also
• FIG. 6 illustrates a flowchart of an exemplary method 200 useful for detecting defects based on 3-D images of thick structures.
• DOI defects of interest
• FIG. 1 depicts a simplified illustration of a 3D NAND structure 160 at the silicon nitride (e.g., SiN or Si3N4) removal step of the wafer production process.
• FIG. 1 is depicted for illustration purposes.
• a manufactured 3D NAND structure includes additional features and elements.
• a manufactured 3D NAND structure includes many additional layers and some of the depicted structures
• structure 182 include additional materials.
• structures 181 extend vertically (e.g., normal to the surface of substrate 186) in the multi-layer 3D NAND structure.
• Layers of Silicon oxide 180 are spaced apart from one another by layers of Silicon nitride (not shown) that are subsequently etched away. Silicon nitride layer 183 is not etched away for purposes of illustration in FIG. 1.
• the next step in the process is to grow tungsten in the space between the silicon oxide layers.
• incomplete etching has left behind silicon nitride defects 184 and 185.
• the electronic device will not function with defects 184 and 185. Thus, it is important to measure this defect as early as possible in the fabrication process to prevent loss of time and
• FIG. 2 is a simplified schematic view of one embodiment of an optical inspection system 100 configured to detect and classify defects of interest (DOI) of
• Optical inspection system 100 includes a computing system, a wafer positioning system, and an optical inspection subsystem including an
• the illumination subsystem includes an illumination source 101 and all optical elements in the illumination optical path from the illumination source to the wafer.
• the collection subsystem includes all optical elements in the collection optical path from the specimen to each detector. For simplification, some optical
• a wafer 103 is illuminated by a normal incidence beam 104 generated by one or more illumination sources 101.
• the illumination subsystem may be configured to direct the beam of light to the specimen at an oblique angle of incidence.
• system 100 may be configured to direct
• multiple beams of light to the specimen such as an oblique incidence beam of light and a normal incidence beam of light.
• the multiple beams of light may be directed to the specimen substantially simultaneously or sequentially.
• Illumination source 101 may include, by way of example, a broad band laser sustained plasma light source, a laser, a supercontinuum laser, a diode laser, a helium neon laser, an argon laser, a solid state laser, a diode pumped solid state (DPSS) laser, a xenon arc lamp, a gas discharging lamp, an LED array, and an incandescent lamp.
• the light source may be configured to emit near
• the illumination subsystem may also include one or more spectral filters that may limit the wavelength of the light directed to the specimen.
• the one or more spectral filters may be bandpass filters and/or edge filters and/or notch filters. Illumination may be provided to the specimen over any suitable range of wavelengths.
• the illumination light includes wavelengths ranging from 260 nanometers to 950 nanometers. In some examples, illumination light includes wavelengths greater than 950 nanometers (e.g., extending to 2,500 nanometers) to capture defects in high aspect ratio structures.
• the illumination subsystem may also include one or more polarization optics to control the polarization of illumination light directed to the
• Beam 104 generated by illumination source 101 is directed to a beam splitter 105.
• Beam splitter 105 directs the beam to objective lens 109.
• Objective lens 109 focuses the beam 111 onto wafer 103 at incident spot 119.
• Incident spot 119 is defined (i.e., shaped and sized) by the
• the inspection system 100 includes illumination aperture 124. As depicted in FIG. 2, computing system 130 communicates command signal 122C to illumination aperture 124. In response, illumination aperture 124 adjusts the illumination direction and beam shape provided onto the surface of the wafer 103. In one embodiment the
• illumination aperture 124 is an assembly that provides varied aperture shapes controlled by command signal 122C communicated from computing system 130.
• computing system 130 As depicted in FIG. 2, computing system 130
• illumination source 101 communicates command signal 122A to illumination source 101.
• illumination source 101 adjusts the spectral range (s) of the illumination beam 111.
• the beam 111 that is incident on wafer 103 may differ from the light emitted by illumination source 101 in one or more ways, including polarization, intensity, size and shape, etc.
• inspection system 100 includes selectable illumination polarization elements 180.
• computing system 130 communicates command signal 122E to illumination polarization elements 180.
• illumination polarization elements 180 adjust the polarization of the illumination light provided onto the surface of the wafer 103.
• inspection system 100 As depicted in FIG. 2, inspection system 100
• the illumination power density includes an illumination power attenuator 102 that controls the illumination power delivered to wafer 103.
• the illumination power density includes an illumination power attenuator 102 that controls the illumination power delivered to wafer 103.
• Attenuator is a beam shaping element that resizes the illumination spot 119 to reduce the illumination power density delivered to wafer 103.
• a combination of illumination power reduction and beam sizing is employed to reduce the illumination power density delivered to wafer 103.
• computing system 130 communicates a control signal to illumination power attenuator 102 to control illumination power based on the three dimensional images detected by any of detectors 115, 120, and 125.
• illumination power In general, illumination power
• Attenuator 102 is optional. Thus, in some other
• inspection system 100 does not include
• system 100 may include a deflector (not shown) in the illumination path.
• the deflector may be an acousto-optical
• the deflector may include a mechanical scanning assembly, an electronic scanner, a rotating mirror, a polygon based scanner, a resonant scanner, a piezoelectric scanner, a galvo mirror, or a galvanometer.
• the deflector scans the light beam over the specimen.
• the deflector may scan the light beam over the specimen at an approximately constant scanning speed.
• System 100 includes collection optics 116, 117, and
• detectors 115, 120, and 125 are communicated to computing system 130 for processing the signals and determining the presence of defects and their locations.
• any of collection optics 116-118 may be a lens, a compound lens, or any appropriate lens known in the art.
• any of collection optics 116-118 may be a reflective or partially reflective optical component, such as a mirror.
• FIG. 2 it is to be understood that the collection optics may be arranged at any
• the collection angle may vary depending upon, for example, the angle of incidence and/or topographical characteristics of the specimen.
• Each of detectors 115, 120, and 125 generally function to convert the reflected and scattered light into an electrical signal, and therefore, may include
• a particular detector may be selected for use within one or more embodiments of the invention based on desired
• an efficiency enhancing detector such as a time delay integration (TDI) camera may increase the signal-to-noise ratio and
• CCD charge-coupled device
• phototubes and photomultiplier tubes may be used, depending on the amount of light available for inspection and the type of inspection being performed.
• a photomultiplier tube is used for detecting light scattered from a specimen.
• Each detector may include only one sensing area, or possibly several sensing areas (e.g., a detector array or multi- anode PMT) .
• System 100 can use various imaging modes, such as bright field and dark field modes.
• detector 125 generates a bright field image. As illustrated in FIG. 2, some amount of light scattered from the surface of wafer 103 at a narrow angle is
• Collection optics 118 includes imaging lens 107 that images the reflected light collected by objective lens 109 onto detector array 125.
• An aperture 182, Fourier filter 106, or both are placed at the back focal plane of objective lens 109.
• Various imaging modes such as bright field, dark field, and phase contrast can be implemented by using different illumination apertures 124, collection apertures, Fourier filters 106, or combinations thereof.
• the configuration of the imaging mode can be determined based on DOI signal and three-dimensional images.
• detectors 115 and 120 generate dark field images by imaging scattered light collected at larger field angles.
• inspection system 100 includes selectable collection polarization elements 181.
• computing system 130 communicates command signal 122F to collection polarization elements 181.
• elements 181 adjust the polarization of the collected light provided onto the surface of detector 125.
• inspection system 100 includes a selectable Fourier filter 106.
• Computing system 130 communicates command signals 122D to Fourier filter 106.
• Fourier filter 106 adjusts the Fourier filtering properties of the Fourier filter (e.g., by changing the specific Fourier filter elements located in the collection beam path) .
• the inspection system 100 includes collection aperture 182. As depicted in FIG. 2, computing system 130 communicates command signal 122G to collection aperture 182. In response, collection aperture 182 adjusts the amount of light collected from the surface of the wafer 103 that is transmitted to the corresponding detector. In one embodiment the collection aperture 182 is an assembly that provides varied aperture shapes controlled by command signal 122G communicated from computing system 130.
• System 100 also includes various electronic
• system 100 may include
• the processor may be coupled directly to an ADC by a transmission medium.
• the processor may receive signals from other electronic components coupled to the ADC. In this manner, the processor may be indirectly coupled to the ADC by a transmission medium and any intervening electronic components.
• wafer positioning system 114 moves wafer 103 under beam 111 based on commands 126 received from computing system 130.
• Wafer positioning system 114 includes a wafer chuck 108, motion controller 113, a rotation stage 110, translation stage
• Z-translation stage 121 is configured to move wafer 103 in a direction normal to the surface of wafer 103 (e.g., the z-direction of
• Translation stage 112 and rotation stage 110 are configured to move wafer 103 in a direction parallel to the surface of wafer 103 (e.g., the x and y directions of coordinate system 123) . In some other embodiments, wafer 103 is moved in the in-plane directions
• Wafer 103 is supported on wafer chuck 108.
• wafer 103 is located with its geometric center approximately aligned with the axis of rotation of rotation stage 110.
• rotation stage 110 spins wafer 103 about its geometric center at a specified angular velocity, ⁇ , within an acceptable tolerance.
• translation stage 112 translates the wafer 103 in a direction approximately perpendicular to the axis of rotation of rotation stage 110 at a specified velocity, VT.
• Motion controller 113 coordinates the spinning of wafer 103 by rotation stage 110 and the translation of wafer 103 by translation stage 112 to achieve a desired in-plane
• motion controller 113 coordinates the movement of wafer 103 by translation stage 121 to achieve a desired out-of-plane scanning motion of wafer 103 within inspection system 100.
• Wafer 103 may be positioned relative to the optical subsystems of inspection system 100 in a number of
• wafer 103 is repeatedly scanned in the lateral directions (e.g., x- direction and y-direction) at each different z-position. In some examples, wafer 103 is scanned at two or more different z-positions, corresponding to two or more depths
• wafer 103 is
• Defect review mode is typically employed to perform more detailed investigation of defects (e.g., higher image resolution, higher focal depth resolution, or both) .
• the wafer is moved to a number of different z-positions with respect to the focal plane of the inspection system to image different depths of the wafer stack.
• the position of the focal plane of the inspection system is adjusted optically to a number of different z-positions with respect to the wafer to image different depths of the wafer stack.
• the images collected at each z-position are aggregated to form a three dimensional volume image of a thick
• the optical subsystem 140 including both the illumination and collection subsystems, generates a focused optical image at each of a plurality of focus planes located at a plurality of different depths of a structure under measurement (e.g., a vertically stacked structure) .
• the alignment of the focus plane of the optical subsystem at each different depth is achieved by optical adjustment that moves the focus plane in the z- direction, specimen positioning in the z-direction, or both.
• One or more detectors detect the light collected at each of the plurality of different depths and generate a plurality of output signals indicative of the amount of light collected at each of the plurality of different depths .
• Optical inspection system 100 generates three dimensional images of a thick semiconductor structure from a volume measured in two lateral dimensions (e.g., parallel to the wafer surface) and a depth dimension (e.g., normal to the wafer surface) .
• computing system 130 arranges the outputs from one or more of the measurement channels (e.g., from one or more of detectors 115, 120, and 125) into a volumetric data set that corresponds to the measured volume.
• computing system 130 generates a three-dimensional image of the measured volume by assembling a stack of the series of two-dimensional images acquired at each different focus offset.
• Focus offset is the relative distance between the most reflective surface of the specimen and the focal plane of the
• the parameter to be scanned is not limited to the focus offset.
• a defect image having more than three dimensions is generated by computing system 130.
• both focus offset and illumination direction are scanned for a given (x,y) location.
• computing system 130 generates a four dimensional image of the measured volume by assembling the series of two-dimensional images acquired at each different focus offset and each different illumination angle into a fourth order tensor.
• a series of images for a predefined set of focus offsets is collected while keeping illumination intensity and other system parameters
• a series of images are acquired at various (x,y) locations for a number of different wafer locations within the focal plane of the inspection system.
• image misalignment between different focus offsets must be minimized. In some examples, this is achieved by
• positions at different focus offsets are aligned after data collection using alignment targets.
• defect detection is performed directly from image data generated by inspection system 100.
• one or more feature vectors are extracted from the collected image data and defect detection is performed based on the measured feature vectors.
• a feature vector is an n-dimensional vector of numerical features that represent an object
• a defect detection algorithm includes one or more selectable threshold values that adjust the
• the defect detection algorithm detects fewer defects of interest from a set of three dimensional images.
• highly permissive threshold values are selected, the defect detection
• an optimized measurement recipe tuned to a particular measurement application also includes a selection of detection algorithm threshold values that maximizes the capture rate of real defects, while minimizing the capture rate of nuisance (i.e., false) defects .
• computing system 130 generates and communicates command signals 122A- G such that illumination power, illumination apertures, collection apertures, spectral band, Fourier filters, illumination polarization, collection polarization, or any combination thereof, are selected in accordance with a specified optical mode.
• an inspection system such as inspection system 100 includes other selectable optical system settings such as angle of incidence, azimuth angle, etc. Each distinct combination of optical system settings is referred to as a distinct optical mode of the optical inspection system 100.
• an inspection system such as
• Exemplary performance objectives include, but are not limited to minimizing the response of the nominal structure in the three dimensional image, enhancing the response of the defect signal in the three dimensional image, minimizing the response of wafer noise or nuisance signals in the three dimensional image,
• an optimized measurement recipe for a particular measurement application includes a
• measurement recipe optimization for a three dimensional optical inspection system such as inspection system 100 includes a selection of an optimal optical mode from thousands of possible system configurations and a selection of detection
• DOIs may be present throughout the depth of a three dimensional semiconductor structure.
• the fact that DOIs may be present throughout the depth of a three dimensional semiconductor structure presents significant practical challenges for defect discovery and recipe optimization because the amount of available image data is so vast (i.e., three dimensional images, not just two dimensional images) and defect verification is so time consuming (i.e., wafer de- processing is required to verify defects) .
• the three dimensional volume of a semiconductor wafer subject to defect discovery
• information about the measurement application 136 under consideration is received by computing system 130 from a user input source 135.
• the user input source 135 is an entity such as a user or operator having knowledge of the structures under inspection and expected defects.
• structural information 136 is an entity such as a user or operator having knowledge of the structures under inspection and expected defects.
• inspection system 100 includes expected stack depth of defect of interest, wafer level signature of defect of interest, refractive index of the 3-D stack, etc.
• inspection system 100 includes peripheral devices useful to accept inputs from an operator (e.g., keyboard, mouse, touchscreen, communication ports, etc.) to communicate structural information 136 from the user to inspection system 100.
• operator e.g., keyboard, mouse, touchscreen, communication ports, etc.
• a user also communicates an initial set of optical modes for inspection system 100.
• a user of inspection system 100 typically performs preliminary modeling or employs past experience to arrive at an initial set of optical modes of inspection system 100, which are most likely to result in the best inspection results.
• an initial set of optical modes includes tens of different optical modes, but far fewer than the thousands of available optical modes.
• a user also communicates one or more initial focus levels to inspection system 100.
• the one or more initial focus levels include focus levels where defects of interest should be located.
• inspection system 100 performs an inspection of wafer 103 at each of the initial set of optical modes and at each of the one or more initial focus levels.
• the inspections are run in a scanning mode, where a large area of the wafer (e.g., the entire area of the wafer) is inspected at each of the one or more initial focus levels.
• Threshold values of the defect detection algorithm employed during the initial inspections set at highly permissive values that identify many defects (i.e., both real and nuisance
• computing system 130 selects a few of the most promising defects identified in the initial inspections.
• the most promising defects are defects of interest that most closely match the expected defects provided by the user of inspection system 100.
• Inspection system 100 performs a through focus review of the selected defects of interest by locating the wafer 103 with respect to optical inspection subsystem 140 such that a selected defect of interest is in the field of view of inspection system 100.
• a series of measurements are performed at a number of focus levels all the way through the structure under measurement.
• computing system 130 determines one or more focus planes or focus range that best capture the defect of interest.
• the one or more focus planes or focus range is determined based on a best match between a measured defect signature (e.g., image or feature vector) and an expected defect signature.
• a measured defect signature e.g., image or feature vector
• inspection system 100 After determining the one or more focus planes or focus range, inspection system 100 records image patches (e.g., 32x32 pixel patches) associated with defect
• defect locations identified in each of the initial inspections at the one or more focus planes or focus range rather than throughout the entire depth of the structure.
• one hundred million defect locations, or more are imaged at multiple focus levels, and recorded.
• the recorded data are employed during subsequent defect verification and recipe optimization processes. By limiting the amount of recorded data, subsequent defect verification and recipe optimization processes are dramatically simplified.
• the number of optical modes under consideration is reduced based on a comparison of one or more measured wafer level defect signatures and one or more expected wafer level defect signatures.
• computing system 130 selects a few of the most promising optical modes for further consideration (e.g., five or fewer optical modes) .
• computing system 130 varies defect detection algorithm threshold values for each optical inspection mode.
• Computing system 130 varies the threshold values to best match a measured wafer level defect signature with an expected wafer level defect signature for each optical mode.
• a wafer level defect signature is a wafer defect map that
• a wafer level defect signature includes any indication of wafer defects expressed across the entire wafer area under inspection .
• optical modes that best match the expected defect signature are selected for further consideration, and the other optical modes are discarded. In this manner, the number of modes selected for recording, as described hereinbefore, is reduced. Thus the amount of inspection data under consideration during subsequent defect
• the number of optical modes under consideration is reduced based on measured defect signal to noise ratio. In one example, after performing the initial inspections and wafer level
• computing system 130 further selects a few of the most promising optical modes for further
• computing system 130 analyzes the signal to noise ratio associated with each selected optical mode at the one or more focus planes or focus levels. Computing system 130 selects the optical modes with the highest signal to noise ratio for further consideration, and the other optical modes are discarded. In this manner, the number of modes selected for recording, as described hereinbefore, is reduced. Thus the amount of inspection data under
• optical modes are selected for further consideration based on wafer level signature analysis .
• the number of optical modes under consideration is reduced based on SEM review of defects without de-processing. In one example, after performing the initial inspections and wafer level
• computing system 130 further selects a few of the most promising optical modes for SEM review without de-processing of the wafer.
• wafer 103 is transferred to a SEM review tool, and SEM review measurements are performed to verify defects of interest at the surface or even slightly below the surface.
• high energy SEM is utilized to review defects buried within a structure under consideration (e.g., depths up to one micrometer) .
• SEM is not suitable for defect verification of defects at significant depths (e.g., beyond one micrometer) .
• Computing system 130 receives an indication of defects verified by the SEM review tool and selects the optical modes with the highest capture rates of verified defects, and the fewest capture rate of nuisance defects, for further consideration. Other optical modes are discarded. In this manner, the number of modes selected for recording, as described hereinbefore, is reduced.
• optical modes are selected for further consideration based on wafer level signature analysis .
• verified defect images/features are mapped to corresponding defects identified by
• the verified defects and the recorded three dimensional images are employed to train a nuisance filter and optimize the measurement recipe.
• FIG. 3 is a simplified schematic view of one
• the system 150 provides a measurement recipe optimization for inspection of three dimensional semiconductor structures.
• computing system 160 includes inspection system 100 as described with reference to FIG. 2, a defect verification tool 151, and a computing system 160.
• the task performed by computing system 160 are as described herein are
• computing system 130 implemented by computing system 130, or another computing system .
• defect verification tool 151 is an electron beam based analysis tool. In some other embodiments, defect verification tool 151 is an x-ray based analysis tool. In these embodiments, a material removal tool may not be necessary to make the buried defect visible to the x-ray based analysis tool. Thus, an associated material removal tool is optional.
• defect verification is achieved by de-processing wafer 103 and inspecting the exposed defects with inspection system 100.
• a different defect verification tool 151 may not be required.
• a defect verification tool such as a SEM review tool may be integrated with inspection system 100 as a single wafer processing tool, or, alternatively,
• Computing system 130 coordinates the inspection processes, and performs analyses, data handling, and communication tasks.
• computing system 160 coordinates the material removal and review processes, performs analyses, and performs data handling and
• Defect verification can be accomplished in many different ways.
• voltage contrast inspection is performed to verify defects.
• a wafer is decorated in accordance with a small sample plan and voltage contrast measurements are performed on the decorated wafer by a voltage contrast inspection tool.
• wafer fabrication is completed and a bit-map test is performed on the finished wafer to verify defects.
• a wafer is de-processed to remove layers of the multiple layer structure under consideration.
• De-processing may be accomplished by chemical processes, mechanical processes, or both.
• a focused ion beam (FIB) tool is employed to remove material from the surface of a wafer.
• the wafer is de-processed until the buried defects are located at or near the surface of the wafer and can be effectively imaged by defect verification tool 151, e.g., a SEM review tool, inspection system 100, etc.
• defect verification tool 151 e.g., a SEM review tool, inspection system 100, etc.
• verification measurements are stored in a memory (e.g., memory 162 on board computing system 160) .
• a memory e.g., memory 162 on board computing system 160.
• the defect information is stored in the form of a KLA results file (KLARF) .
• KLARF file is a flat ASCII file produced by the defect verification tool 150.
• the same KLARF file format is used to save defect
• computing system 160 samples the identified defects to generate a diversity set of DOIs 153
• computing system 160 bins the defects
• the diversity set of DOIs 153 are saved in a memory (e.g., memory 162 on board computing system 160) .
• Defect verification measurements are performed on the diversity set of DOIs.
• the defect locations and associated defect images from the defect verification measurements are stored in a memory (e.g., memory 162 on board computing system 160) .
• defect information associated with the diversity set of DOIs is also stored in a KLARF file format.
• Defect verification data from the diversity set of DOIs, any other set of verified defects, or a combination thereof, are mapped to the saved through focus defect image patches and corresponding feature vectors.
• the defect verification data and the corresponding through focus defect image patches and corresponding feature vectors are employed to train a nuisance elimination filter.
• computing system 160 trains a through focus image based machine learning network to filter out nuisance defects.
• the machine learning network is trained based on defect images.
• a suitable machine learning network is implemented as a neural network, a support vector machines model, a decision tree model, etc.
• computing system 160 trains a through focus feature based automated classifier to filter out nuisance defects.
• the automated classifier is feature based, rather than image based.
• a suitable through focus feature based automated classifier is implemented as a trained random forest algorithm, etc.
• computing system 160 implements a rule based tree classifier to filter out nuisance
• the rule based tree classifier is feature based, rather than image based.
• a suitable rule based tree classifier is implemented based on manually generated rules .
• the trained nuisance elimination filter 142 is communicated to inspection system 100 and applied to the saved defect images associated with each optics mode under consideration. In this manner, defect detection is emulated using the through focus defect events recorded during defect discovery. Detection threshold values associated with each optical mode are adjusted to achieve a desirable nuisance rate. In one example, the detection threshold values associated with each optical mode are adjusted to achieve a nuisance rate of approximately 30%, and the optical mode that achieves the best defect
• inspection system 100 is selected for implementation as the production measurement recipe for the measurement application under consideration.
• inspection system 100 is depicted in FIG. 3, inspection system 100
• nuisance filter 142 implements nuisance filter 142 and the selected production measurement recipe to identify and classify defects based on an analysis of three dimensional images of thick
• processor 131 is configured to detect and classify defects from a three-dimensional image.
• the processor may include any appropriate processor known in the art.
• the processor may be configured to use any appropriate defect detection and classification algorithm or method known in the art.
• the processor may use a die-to-database comparison, a three- dimensional filter, a clustering algorithm such as a principal component analysis or spectral clustering, a thresholding algorithm, a deep learning algorithm, or any other suitable algorithm to detect and classify defects on the specimen.
• the three dimensional location of a defect of interest is determined based on an analysis of the three dimensional image of a thick semiconductor structure.
• the actual position of a defect within a wafer is measured (e.g., ⁇ x,y,z ⁇ coordinates of the defect) .
• the actual defect position can be used to locate the defect later for further analysis (e.g.,
• the x-position, y-position, and focus offset associated with the peak defect signal within the 3D image is used to evaluate the actual defect position within the wafer structure (e.g., 3D NAND wafer stack) .
• the diffraction orders are suppressed and the actual defect location in the z-direction (e.g., depth) is linearly related to the focus offset associated with the peak signal.
• the defect location in the z-direction is linearly related to the focus offset associated with the peak signal.
• computing system 130 determines the focus offset associated with the peak and determines the defect depth by multiplying the focus offset by a scaling factor.
• the actual defect position is determined by comparing the three dimensional image and one or more simulated three-dimensional images of a defect.
• computing system 130 performs a rigorous coupled wave analysis (RCWA) to simulate the measured defect response. This analysis may be performed
• a measurement library is generated that matches measured data with defect depths
• the trusted reference measurement system is a defect review performed after focus ion beam etching of a specimen under consideration.
• defect locations associated with subsequent measurements are estimated based on library matching.
• the three dimensional image is filtered before defect analysis to improve SNR.
• computing system analyzes the assembled three- dimensional image with a 3D digital filter, or other suitable numerical technique to detect unique three- dimensional structures arising from defects. This
• computing systems 130 and 160 are configured to detect and classify defects based on through focus images or feature vectors.
• Computing systems 130 and 160 may include any appropriate processor (s) known in the art.
• computing systems 130 and 160 may be configured to use any appropriate defect detection
• computing systems 130 and 160 may use a die-to-database comparison or a thresholding algorithm to detect defects on the specimen.
• inspection system 100 may include peripheral devices useful to accept inputs from an operator (e.g., keyboard, mouse, touchscreen, etc.) and display outputs to the operator (e.g., display monitor). Input commands from an operator may be used by computing system 130 to adjust threshold values used to control illumination power. The resulting power levels may be graphically presented to an operator on a display monitor.
• peripheral devices useful to accept inputs from an operator (e.g., keyboard, mouse, touchscreen, etc.) and display outputs to the operator (e.g., display monitor).
• Input commands from an operator may be used by computing system 130 to adjust threshold values used to control illumination power.
• the resulting power levels may be graphically presented to an operator on a display monitor.
• Inspection system 100 includes a processor 131 and an amount of computer readable memory 132.
• Processor 131 and memory 132 may communicate over bus 133.
• Memory 132 includes an amount of memory 134 that stores an amount of program code that, when executed by processor 131, causes processor 131 to execute the defect detection,
• System 150 includes a processor 161 and an amount of computer readable memory 162.
• Processor 161 and memory 162 may communicate over bus 163.
• Memory 162 includes an amount of memory 164 that stores an amount of program code that, when executed by processor 161, causes processor 161 to execute the defect detection, classification, and depth estimation functionality described herein.
• FIG. 6 illustrates a flowchart of an exemplary method 200 useful for detecting defects based on 3-D images of thick structures.
• inspection system 100 described with reference to FIG. 2 is configured to implement method 200.
• the implementation of method 200 is not limited by the specific embodiments described herein.
• providing an amount of illumination light is provided to a semiconductor wafer at a plurality of defect locations in accordance with each of a plurality of optical modes at each of a plurality of focus planes within each of a plurality of vertically stacked structures disposed on the semiconductor wafer.
• an amount of light is imaged from each of the vertically stacked structures in response to the amount of illumination light in accordance with each of the plurality of optical modes at each of the plurality of focus planes at each of the plurality of defect locations.
• one or more defect locations are selected from the plurality of defect locations.
• a plurality of images are generated at different focus planes through the vertically stacked structure at each of the selected defect locations in accordance with the plurality of optical modes.
• a subset of the plurality of focus planes is selected for storage.
• a stack preferably include detection of defects throughout a stack, including the stack surface and throughout the various depths of a stack. For example, certain embodiments allow defects to be found at depths of up to about three
• defects can be any defect. In another embodiment, defects can be any defect.
• a vertical ONON or OPOP stack under inspection is limited only by the depth of penetration of the illumination light. Transmission through an oxide-nitride-oxide-nitrite (ONON) or oxide- polysilicon-oxide-polysilicon (OPOP) stack is limited less by absorption at longer wavelengths. Thus, longer illumination wavelengths may be employed to effectively inspect very deep structures.
• ONON oxide-nitride-oxide-nitrite
• OPOP oxide- polysilicon-oxide-polysilicon
• the three dimensional imaging techniques described herein can be applied to complex, vertically stacked structures, including, but not limited to 3D negative-AND
• NAND vertical NAND
• TCAT terabit cell array transistors
• VSAT vertical-stacked array transistors
• the vertical direction is generally a direction that is perpendicular to the substrate surface.
• inspection embodiments may be applied at any point in the fabrication flow that results in multiple layers being formed on a substrate, and such layers may include any number and type of materials.
• specimen is used herein to refer to a wafer, a reticle, or any other sample that may be inspected for defects, features, or other information (e.g., an amount of haze or film properties) known in the art.
• wafer generally refers to substrates formed of a semiconductor or non- semiconductor material. Examples include, but are not limited to, monocrystalline silicon, gallium arsenide, and indium phosphide. Such substrates may be commonly found and/or processed in semiconductor fabrication facilities. In some cases, a wafer may include only the substrate
• a wafer may include one or more layers of different materials formed upon a
• One or more layers formed on a wafer may be "patterned” or “unpatterned . "
• a wafer may include a plurality of dies having repeatable pattern features .
• a "reticle” may be a reticle at any stage of a reticle fabrication process, or a completed reticle that may or may not be released for use in a semiconductor fabrication facility.
• a reticle, or a "mask,” is generally defined as a substantially transparent substrate having substantially opaque regions formed thereon and configured in a pattern.
• the substrate may include, for example, a glass material such as quartz.
• a reticle may be disposed above a resist-covered wafer during an exposure step of a lithography process such that the pattern on the reticle may be transferred to the resist.
• Computer-readable media includes both computer storage media and. communication media including any medium that facilitates transfer of a computer program from one place to another.
• a storage media may be any available media that can be accessed by a general purpose or special purpose computer.
• such computer-readable media can comprise RAM, ROM, EEPROM, CD-ROM or other optical disk storage, magnetic disk storage or other magnetic storage devices, or any other medium that can be used to carry or store desired program code means in the form of instructions or data structures and that can be accessed by a general-purpose or special-purpose computer, or a general-purpose or special- purpose processor.
• any connection is properly termed a computer- readable medium. For example, if the software is transmitted from a website, server, or other remote source using a coaxial cable, fiber optic cable, twisted pair, digital subscriber line (DSL) s or wireless
• Disk and disc includes compact disc (CD) , laser disc, optical disc, digital versatile disc (DVD- , floppy disk and. biu-ra.y disc where disks usually reproduce data magnetically, while discs reproduce data optically with lasers. Combinations of the above should also be included within the scope of computer-readable media.
• a detector may include a fiber array.
• inspection system 100 may include more than one light source (not shown) .
• the light sources may be
• the light sources may be configured to generate light having
• the light sources may be configured according to any of the embodiments described herein. In addition one of the light sources may be configured according to any of the embodiments described herein, and another light source may be any other light source known in the art.
• an inspection system may illuminate the wafer over more than one illumination area simultaneously.
• the multiple illumination areas may spatially overlap.
• the multiple illumination areas may be spatially distinct.
• an inspection system may illuminate the wafer over more than one illumination area at different times.
• the different illumination areas may temporally overlap (i.e., simultaneously illuminated over some period of time) .
• the different illumination areas may be
• illumination areas may be arbitrary, and each illumination area may be of equal or different size, orientation, and angle of incidence.
• inspection system 100 may be a scanning spot system with one or more illumination areas that scan independently from any motion of wafer 103.
• an illumination area is made to scan in a repeated pattern along a scan line. The scan line may or may not align with the scan motion of wafer 103.
• wafer positioning system 114 generates motion of wafer 103 by coordinated rotational and translational movements
• wafer positioning system 114 may generate motion of wafer 103 by coordinating two translational movements.
• wafer positioning system 114 may generate motion along two orthogonal, linear axes (e.g., X-Y

Landscapes

• Physics & Mathematics (AREA)
• Engineering & Computer Science (AREA)
• General Physics & Mathematics (AREA)
• Health & Medical Sciences (AREA)
• Biochemistry (AREA)
• Analytical Chemistry (AREA)
• General Health & Medical Sciences (AREA)
• Chemical & Material Sciences (AREA)
• Immunology (AREA)
• Pathology (AREA)
• Life Sciences & Earth Sciences (AREA)
• Computer Vision & Pattern Recognition (AREA)
• General Engineering & Computer Science (AREA)
• Quality & Reliability (AREA)
• Theoretical Computer Science (AREA)
• Computer Hardware Design (AREA)
• Microelectronics & Electronic Packaging (AREA)
• Electromagnetism (AREA)
• Toxicology (AREA)
• Testing Or Measuring Of Semiconductors Or The Like (AREA)
• Investigating Materials By The Use Of Optical Means Adapted For Particular Applications (AREA)
• Analysing Materials By The Use Of Radiation (AREA)

Priority Applications (3)

Applications Claiming Priority (6)

Publications (2)

Family

ID=62190747

Family Applications (1)

Country Status (6)

Cited By (3)

Families Citing this family (53)

Family Cites Families (48)

• 2017
  • 2017-11-29 US US15/826,019 patent/US11047806B2/en active Active
  • 2017-11-30 WO PCT/US2017/064040 patent/WO2018102596A2/en not_active Ceased
  • 2017-11-30 TW TW106141985A patent/TWI774708B/zh active
  • 2017-11-30 KR KR1020197017165A patent/KR102438824B1/ko active Active
  • 2017-11-30 CN CN201780071387.1A patent/CN109964116B/zh active Active
  • 2017-11-30 JP JP2019528902A patent/JP7080884B2/ja active Active

Also Published As

Similar Documents

Legal Events