Abstract: Disclosed are methods and apparatus for inspecting a photolithographic reticle. A plurality of reference far field images are simulated by inputting a plurality of reference near field images into a physics-based model, and the plurality of reference near field images are generated by a trained deep learning model from a test portion of the design database that was used to fabricate a test area of a test reticle. The test area of a test reticle, which was fabricated from the design database, is inspected for defects via a die-to-database process that includes comparing the plurality of reference far field reticle images simulated by the physic-based model to a plurality of test images acquired by the inspection system from the test area of the test reticle.

Abstract: Metrology methods and targets are provided for reducing or eliminating a difference between a device pattern position and a target pattern position while maintaining target printability, process compatibility and optical contrast—in both imaging and scatterometry metrology. Pattern placement discrepancies may be reduced by using sub-resolved assist features in the mask design which have a same periodicity (fine pitch) as the periodic structure and/or by calibrating the measurement results using PPE (pattern placement error) correction factors derived by applying learning procedures to specific calibration terms, in measurements and/or simulations. Metrology targets are disclosed with multiple periodic structures at the same layer (in addition to regular target structures), e.g., in one or two layers, which are used to calibrate and remove PPE, especially when related to asymmetric effects such as scanner aberrations, off-axis illumination and other error sources.

Abstract: Process control methods, metrology targets and production systems are provided for reducing or eliminating process overlay errors. Metrology targets have pair(s) of periodic structures with different segmentations, e.g., no segmentation in one periodic structure and device-like segmentation in the other periodic structure of the pair. Process control methods derive metrology measurements from the periodic structures at the previous layer directly following the production thereof, and prior to production of the periodic structures at the current layer, and use the derived measurements to adjust lithography stage(s) that is part of production of the current layer. Production system integrate lithography tool(s) and metrology tool(s) into a production feedback loop that enables layer-by-layer process adjustments.

Abstract: Methods and systems for estimating values of process parameters, structural parameters, or both, based on x-ray scatterometry measurements of high aspect ratio semiconductor structures are presented herein. X-ray scatterometry measurements are performed at one or more steps of a fabrication process flow. The measurements are performed quickly and with sufficient accuracy to enable yield improvement of an on-going semiconductor fabrication process flow. Process corrections are determined based on the measured values of parameters of interest and the corrections are communicated to the process tool to change one or more process control parameters of the process tool. In some examples, measurements are performed while the wafer is being processed to control the on-going fabrication process step. In some examples, X-ray scatterometry measurements are performed after a particular process step and process control parameters are updated for processing of future devices.

Abstract: Systems and methods are provided which utilize optical microcavity probes to map wafer topography by near-field interactions therebetween in a manner which complies with high volume metrology requirements. The optical microcavity probes detect features on a wafer by shifts in an interference signal between reference radiation and near-field interactions of radiation in the microcavities and wafer features, such as device features and metrology target features. Various illumination and detection configurations provide quick and sensitive signals which are used to enhance optical metrology measurements with respect to their accuracy and sensitivity. The optical microcavity probes may be scanned at a controlled height and position with respect to the wafer and provide information concerning the spatial relations between device and target features.

Abstract: A method and system for measuring overlay in a semiconductor manufacturing process comprise capturing an image of a feature in an article at a predetermined manufacturing stage, deriving a quantity of an image parameter from the image and converting the quantity into an overlay measurement. The conversion is by reference to an image parameter quantity derived from a reference image of a feature at the same predetermined manufacturing stage with known overlay (“OVL”). There is also disclosed a method of determining a device inspection recipe for use by an inspection tool comprising identifying device patterns as candidate device care areas that may be sensitive to OVL, deriving an OVL response for each identified pattern, correlating the OVL response with measured OVL, and selecting some or all of the device patterns as device care areas based on the correlation.

Abstract: Metrology targets, production processes and optical systems are provided, which enable metrology of device-like targets. Supplementary structure(s) may be introduced in the target to interact optically with the bottom layer and/or with the top layer of the target and target cells configurations enable deriving measurements of device-characteristic features. For example, supplementary structure(s) may be designed to yield Moiré patterns with one or both layers, and metrology parameters may be derived from these patterns. Device production processes were adapted to enable production of corresponding targets, which may be measured by standard or by provided modified optical systems, configured to enable phase measurements of the Moiré patterns.

Abstract: Common events between layers on a semiconductor wafer are filtered. Common events should contain the majority of defects of interest. Only nuisance events that are common between layers on the semiconductor wafer remain, which reduces the nuisance rate. Defects that are common across layers can be filtered based on, for example, defect coordinates, a difference image, or defect attributes.

Abstract: Methods and systems for accelerated training of a machine learning based model for semiconductor applications are provided. One method for training a machine learning based model includes acquiring information for non-nominal instances of specimen(s) on which a process is performed. The machine learning based model is configured for performing simulation(s) for the specimens. The machine learning based model is trained with only information for nominal instances of additional specimen(s). The method also includes re-training the machine learning based model with the information for the non-nominal instances of the specimen(s) thereby performing transfer learning of the information for the non-nominal instances of the specimen(s) to the machine learning based model.

Abstract: Methods and systems disclosed herein can measure thin film stacks, such as film on grating and bandgap on grating in semiconductors. For example, the thin film stack may be a 1D film stack, a 2D film on grating, or a 3D film on grating. One or more effective medium dispersion models are created for the film stack. Each effective medium dispersion model can substitute for one or more layers. A thickness of one or more layers can be determined using the effective medium dispersion based scatterometry model. In an instance, three effective medium dispersion based scatterometry models are developed and used to determine thickness of three layers in a film stack.

Abstract: A dynamic misregistration measurement amelioration method including taking at least one misregistration measurement at multiple sites on a first semiconductor device wafer, which is selected from a batch of semiconductor device wafers intended to be identical, analyzing each of the misregistration measurements, using data from the analysis of each of the misregistration measurements to determine ameliorated misregistration measurement parameters at each one of the multiple sites, thereafter ameliorating misregistration metrology tool setup for ameliorated misregistration measurement at the each one of the multiple sites, thereby generating an ameliorated misregistration metrology tool setup and thereafter measuring misregistration at multiple sites on a second semiconductor device wafer, which is selected from the batch of semiconductor device wafers intended to be identical, using the ameliorated misregistration metrology tool setup.

Abstract: Techniques and systems for critical dimension metrology are disclosed. Critical parameters can be constrained with at least one floating parameter and one or more weight coefficients. A neural network is trained to use a model that includes a Jacobian matrix. During training, at least one of the weight coefficients is adjusted, a regression is performed on reference spectra, and a root-mean-square error between the critical parameters and the reference spectra is determined. The training may be repeated until the root-mean-square error is less than a convergence threshold.

Abstract: A three-dimensional (3D) microscope includes various insertable components that facilitate multiple imaging and measurement capabilities. These capabilities include Nomarski imaging, polarized light imaging, quantitative differential interference contrast (q-DIC) imaging, motorized polarized light imaging, phase-shifting interferometry (PSI), and vertical-scanning interferometry (VSI).

Abstract: Metrology methods and targets are provided for reducing or eliminating a difference between a device pattern position and a target pattern position while maintaining target printability, process compatibility and optical contrast—in both imaging and scatterometry metrology. Pattern placement discrepancies may be reduced by using sub-resolved assist features in the mask design which have a same periodicity (fine pitch) as the periodic structure and/or by calibrating the measurement results using PPE (pattern placement error) correction factors derived by applying learning procedures to specific calibration terms, in measurements and/or simulations. Metrology targets are disclosed with multiple periodic structures at the same layer (in addition to regular target structures), e.g., in one or two layers, which are used to calibrate and remove PPE, especially when related to asymmetric effects such as scanner aberrations, off-axis illumination and other error sources.

Abstract: Methods and systems for improving a measurement recipe describing a sequence of measurements employed to characterize semiconductor structures are described herein. A measurement recipe is repeatedly updated before a queue of measurements defined by the previous measurement recipe is fully executed. In some examples, an improved measurement recipe identifies a minimum set of measurement options that increases wafer throughput while meeting measurement uncertainty requirements. In some examples, measurement recipe optimization is controlled to trade off measurement robustness and measurement time. This enables flexibility in the case of outliers and process excursions. In some examples, measurement recipe optimization is controlled to minimize any combination of measurement uncertainty, measurement time, move time, and target dose. In some examples, a measurement recipe is updated while measurement data is being collected.

Abstract: Methods and systems for characterizing dimensions and material properties of semiconductor devices by transmission small angle x-ray scatterometry (TSAXS) systems having relatively small tool footprint are described herein. The methods and systems described herein enable Q space resolution adequate for metrology of semiconductor structures with reduced optical path length. In general, the x-ray beam is focused closer to the wafer surface for relatively small targets and closer to the detector for relatively large targets. In some embodiments, a high resolution detector with small point spread function (PSF) is employed to mitigate detector PSF limits on achievable Q resolution. In some embodiments, the detector locates an incident photon with sub-pixel accuracy by determining the centroid of a cloud of electrons stimulated by the photon conversion event. In some embodiments, the detector resolves one or more x-ray photon energies in addition to location of incidence.

Abstract: A method of defect discovery can include providing a nuisance bin in a nuisance filter, partitioning the defect population into a defect population partition, segmenting the defect population partition into a defect population segment, selecting from the defect population segment a selected set of defects, computing one or more statistics of the signal attributes of the defects in the defect population segment, replicating the selected set of defects to yield generated defects, shifting the generated defects outside of the defect population segment, creating a training set, and training a binary classifier. This method can be operated on a system. The method can enable a semiconductor manufacturer to determine more accurately the presence of defects that would otherwise have gone unnoticed.

Abstract: A high-dimensional variable selection unit determines a list of critical parameters from sensor data and parametric tool measurements from a semiconductor manufacturing tool, such as a semiconductor inspection tool or other types of semiconductor manufacturing tools. The high-dimensional variable selection model can be, for example, elastic net, forward-stagewise regression, or least angle regression. The list of critical parameters may be used to design a next generation semiconductor manufacturing tool, to bring the semiconductor manufacturing tool back to a normal status, to match a semiconductor manufacturing tool's results with that of another semiconductor manufacturing tool, or to develop a specification for the semiconductor manufacturing tool.

Abstract: A vacuum hold-down apparatus retains a wafer in a desired position and orientation. A vacuum chuck assembly of the vacuum hold-down apparatus has a vacuum chuck surface with a vacuum communication aperture. A venturi vacuum generator is fixed with respect to the vacuum chuck assembly and communicates with the vacuum chuck surface via the vacuum communication aperture. A positive pressure fluid line communicates with the venturi vacuum generator.

Abstract: Machine learning techniques are used to predict values of fixed parameters when given reference values of critical parameters. For example, a neural network can be trained based on one or more critical parameters and a low-dimensional real-valued vector associated with a spectrum, such as a spectroscopic ellipsometry spectrum or a specular reflectance spectrum. Another neural network can map the low-dimensional real-valued vector. When using two neural networks, one neural network can be trained to map the spectra to the low-dimensional real-valued vector. Another neural network can be trained to predict the fixed parameter based on the critical parameters and the low-dimensional real-valued vector from the other neural network.