Abstract: A machine learning model that uses composite metrology data determines at least one parameter of a device under test using measured metrology data from the device. The composite metrology data is generated by merging measured metrology data from a reference device with synthetic metrology data calculated from a model of the reference device. The composite metrology data may be generated further based on a synthetic metrology data calculated from a model for a modified reference device. The modified reference device may be generated using variations of at least one parameter of the model to expand the parameter space of the training range.

Abstract: A horizontal Dewar flask is used with an optical metrology device, which may advantageously reduce the vertical height of the device. A thermal transfer member provides thermal transfer between a liquefied gas cooled sensor and liquefied gas in a chamber of the Dewar flask. To compensate for the loss of thermal transfer from the sensor as the liquefied gas evaporates and changes to a gaseous state, the thermal transfer member biases heat transfer to the liquefied gas that is at the bottom of the chamber. The thermal transfer member may have a larger surface area at a bottom portion of the thermal transfer member than the upper portion. For example, the thermal transfer member may include one or more projections that extend into the liquefied gas with greater density at the bottom of the chamber than at the top of the chamber.

Abstract: An optical metrology device performs multi-wavelength polarized confocal Raman spectroscopy. The optical metrology device uses a first light source to produce a first light beam with a first wavelength and a second light source to produce a second light beam with a second wavelength. A dichroic beam splitter partially reflects the first light beam and transmits the second light beam to combine the light beams along a same optical axis that is incident on a sample. The dichroic beam splitter directs the Raman response emitted from the sample in response to the first light beam and the second light beam together towards at least one spectrometer and directs the first light beam away from the at least one spectrometer. A chopper may be used to isolate the Raman response to the first and second light beams that is received and spectrally measured by the at least one spectrometer.

Abstract: An alignment or overlay target that has an optically opaque layer disposed between the top and bottom target structure is measured using opto-acoustic metrology. A classifier library is generated for classifying whether an opto-acoustic metrology signal is on or off the bottom structure. A target may be measured by acquiring opto-acoustic measurement data for the bottom structure of the target and determining a location of the bottom structure using opto-acoustic metrology data acquired from the different locations over the bottom structure and the classifier library. Locations for acquisition of the data may be based on classification results of each measurement and a search pattern. The top structure of the target may be optically imaged. The relative position of the top structure with respect to the bottom structure is determined using the opto-acoustically determined location of the bottom structure and the image of the top structure.

Abstract: Physical modeling and machine learning modeling are combined to analyze signals from multiple data sources, including metrology data acquired from different tool sets or at different process steps, and data related to processing equipment, such as sensor data, process parameters, Advanced Process Control (APC) parameters, context data, etc. At least one physical model is generated and used to analyze metrology signals from metrology tools to extract measurement results for key and non-key parameters of a structure on a sample. At least one machine learning model is built and trained to predict parameters of interest based on the extracted measurement results as well as additional data, including raw measured signals, reference data and/or design of experiment (DOE) data, and data from different tool sets or the same tool as used for the physical modeling.

Abstract: Complex structures, such as gate-all-around (GAA) field effect transistor or high-aspect ratio (HAR) Channel hole etch, etc., in semiconductor devices are measured using a combination of physical modeling and machine learning modeling. Metrology signals collected at different manufacturing process steps, e.g., pre-process step and post-process step of the structure of interest (SOI) may be used. The measurement signals acquired at the pre-process steps are used to determine a first parameter of the SOT, e.g., using physical modeling and machine learning, which may be fed forward and used to generate a physical model of the SOI at the post-process step. A second parameter of the SOI at the post-process step is determined using physical modeling and machine learning and may be fed back and used to generate the physical model of the SOI at the post-process step with post process signals and used to determine other parameters.

Abstract: A metrology target is designed for measuring a feature at the bottom of a trench in a device under test, such as a tungsten recess vertical profile in a wordline in a three-dimensional (3D) NAND. The metrology target follows the design rules for the device under test and includes a tier stack with a plurality of tier stack pairs including, each including a conductor layer, such as tungsten, and an insulator layer, such as silicon dioxide and a trench that extends through the tier stack pairs. The metrology target includes a via that extends through the tier stack pairs and is positioned a lateral distance to the trench to promote access of light to a bottom of the trench, via plasmonic resonance, for measurement of a characteristic of the trench, such as the tungsten recess at the bottom of the wordline slit.

Abstract: A chuck assembly includes an upper surface configured to support a wafer-level package assembly and a clamping mechanism securing the wafer-level package assembly to the upper surface.

Abstract: One example of an inspection system includes a laser, a magnification changer, and a first camera. The laser projects a line onto a wafer to be inspected. The magnification changer includes a plurality of selectable lenses of different magnification. The first camera images the line projected onto the wafer and outputs three-dimensional line data indicating the height of features of the wafer. Each lens of the magnification changer provides the same nominal focal plane position of the first camera with respect to the wafer.

Abstract: As feature sizes of semiconductor chips shrink there is a need for tighter overlay between layers in a lithography process. This means more advanced and larger overlay corrections may be necessary to ensure that die are properly manufactured into chips, especially in reconstituted substrates where the die can shift in the process of creating the substrate. Systems and methods for correcting these overlay errors in a lithographic process are provided. Additional rotation (theta) and projected image size (mag) corrections can be made to correct overlay errors present in reconstituted substrates by adjusting the stage and the reticle. Furthermore, these adjustments can be made allowing site-by-site or zone-by-zone corrections instead of a one-time adjustment of the reticle chuck as has been done in the past. These corrections can alleviate some of the issues associated with fan-out wafer-level packaging (FOWLP) and fan-out panel-level packaging (FOPLP).

Abstract: A non-destructive opto-acoustic metrology device detects the presence and location of non-uniformities in a film stack that includes a large number, e.g., 50 or more, transparent layers. A transducer layer at the bottom of the film stack produces an acoustic wave in response to an excitation beam. A probe beam is reflected from the layer interfaces of the film stack and the acoustic wave to produce an interference signal that encodes data in a time domain from destructive and constructive interference as the acoustic wave propagates upward in the film stack. The data may be analyzed across the time domain to determine the presence and location of one or more non-uniformities in the film stack. An acoustic metrology target may be produced with a transducer layer configured to generate an acoustic wave with a desired acoustic profile based on characteristics of the film stack.

Abstract: One example of an inspection system includes a laser, a magnification changer, and a first camera. The laser projects a line onto a wafer to be inspected. The magnification changer includes a plurality of selectable lenses of different magnification. The first camera images the line projected onto the wafer and outputs three-dimensional line data indicating the height of features of the wafer. Each lens of the magnification changer provides the same nominal focal plane position of the first camera with respect to the wafer.

Abstract: A method for correcting misalignments is provided. An alignment for each device of a group of devices mounted on a substrate is determined. An alignment error for the group of devices mounted on the substrate is determined based on the respective alignment for each device. One or more correction factors are calculated based on the alignment error. The alignment error is corrected based on the one or more correction factors.

Abstract: An ellipsometer includes a focusing system that uses an image of the measurement spot to determine a best focal position for the ellipsometer. The focus signal is produced by splitting off the ellipsometer measurement spot before the signal is analyzed by a polarizer thereby avoiding imagining the spot with a modulated intensity. The focus signal is imaged on a sensor array and based on the position of the spot on the sensor array, the focal position of the ellipsometer may be determined. A single image may be used to determine the focal position of the ellipsometer permitting a real time focus position measurement.

Abstract: An improved stage for the processing of large, thin substrates, such as glass and semiconductor panels. Processing includes lithography, inspection, metrology, grinding, and the like. The stage includes a chuck that moves over a base relative to a device for processing a substrate. The chuck conforms to a geometry of the base while moving relative to the base.

Abstract: A pixelated color mask is combined with a pixelated polarization mask in dynamic interferometry. The color mask includes a wavelength-selective bandpass filter placed in front of each camera pixel such that each set of contiguous four camera pixels is covered by two green bandpass filters, a red bandpass filter, and a blue bandpass filter. The pixelated phase mask is coupled to the color filters such that one polarization filter covers one set of color filters. At least three polarization filters are used to calculate phase. In addition, the color signals can be used, for example, to encode the motion of the interferometer, to provide very high speed autofocus or tip/tilt feedback, to create a color image of the object being measured, to automatically focus the system at different positions for different measurements conducted with different color sources, and to perform heterodyne interferometry with a single, vibration-immune measurement.

Abstract: An ellipsometer uses a broadband light source and a Fresnel cone to produce a simultaneous broadband polarization state generator with no moving parts. The detector of the ellipsometer includes a diffractive element to spatially separate the wavelengths of the light from the sample. The wavelengths may be spatially separated sufficiently that there is no overlap of bands of wavelengths when imaged by a two-dimensional sensor or may be temporally separated. Additionally, the detector separates and simultaneously analyzes the polarizations states of the light from the sample so there is no overlap of polarization states when imaged by a two-dimensional sensor and no moving parts are used. The resulting image with separated wavelengths and polarization states may be used to determine at least a partial Mueller matrix for the sample.

Abstract: An optical metrology device uses a multi-wavelength beam of light that has azimuthally varying polarization states and/or phase states, referred to as a vortex beam. The metrology device focuses the vortex beam on a sample under test over a large range of angles of incidence. The metrology device may detect an image of the vortex beam reflected from the sample and measure the polarization state of the return light as function of the angle of incidence and the azimuth angle, which may be further measured at a plurality of different wavelengths. The vortex beam includes azimuthally varying polarization states, thereby enabling measurement of all desired polarization states without requiring the use of moving optical components. The polarization state information detected over multiple angles of incidence and wavelengths provides data with which an accurate determination of one or more characteristics of a sample may be determined.

Abstract: The light from an optical metrology device is focused into a measurement spot on a sample using a focusing system. The focusing system uses an image of the light reflected from the measurement spot to determine a best focal position at a desired position of the sample. The focusing system selects a characteristic of reflected light, such as polarization state or wavelengths, to use for focusing. The characteristic of the reflected light that is selected for use in determining focal position is affected different by different portions of the sample. For example, light reflected from a top surface of a sample may have a different characteristic than light reflected by an underlying layer. The selected characteristic of the reflected light is used by the focusing system to focus the measurement spot at the top surface or an underlying layer of the sample.

Abstract: A method for monitoring and controlling a substrate singulation process is described. Device edges are imaged and identified for analysis. Discrepancies in device edges are noted and used to modify a singulation process and to monitor the operation of singulation processes for anomalous behavior.