A method for overlay metrology may include generating broadband illumination beams and directing the broadband illumination beams to an overlay target on a sample, where the overlay target may include cells having periodic features formed as overlapping grating structures. The method may include generating diffracted light using the periodic features of the overlay target, where the periodic features may act as a diffraction grating to generate diffracted light by separating the broadband illumination beam into a plurality of wavelengths. The method may include generating pupil images of cells of the overlay target, where a distribution of light in the pupil plane may include first-order diffraction lobes, where spectra of the first-order diffraction lobes may be spatially dispersed in the pupil plane. The method may include generating an overlay measurement based on portions of the pupil images corresponding to selected wavelengths of the spectra of the first-order diffraction lobes.

Disclosed herein is a control module for a semiconductor process system, utilizing reinforcement learning (RL) algorithms to autonomously generate and adjust process recipes. It features a comprehensive system digital twin, including subsystem, chamber plasma, and process digital twins, and employs neural network models for efficiency. Using a policy neural network and Monte Carlo Tree Search (MCTS), real-time adjustments are based on calibrated state data from various sensors, enhancing precision and adaptability in manufacturing processes.

Systems and methods provide for localizing a beam to a plurality of regions of interest for creating a synthetically stitched image. The process includes creating at least a first region of interest of a target and at least a second region of interest of the target, each of the first region of the target and the second region of interest defined by one or more previously established rules for polygonal dimensions. Next, performing a first scan of the target to capture at least one first image of the first region of interest and performing a second scan of the target to capture at least one second image of the second region of interest occurs. The captured images are then stitched together to create a stitched final image of the target.

Method and machine utilizes the real-time recipe to perform weak point inspection on a series of wafers during the fabrication of integrated circuits. Each real-time recipe essentially corresponds to a practical fabrication history of a wafer to be examined and/or the examination results of at least one examined wafer of same lot. Therefore, different wafers can be examined by using different recipes where each recipe corresponds to a specific condition of a wafer to be examined, even these wafers are received by a machine for examining at the same time.

A swath imaging system may include an illumination source configured to illuminate a sample with an illumination beam, a stage to scan the sample with a scan pattern including swaths extending along a scan direction when implementing the inspection recipe, one or more TDI sensors configured to capture swath images of the sample, and a controller. The plurality of swaths may be distributed along a step direction orthogonal to the scan direction, and at least some of the plurality of swath images overlap along the step direction. The controller may implement the inspection recipe by receiving the plurality of swath images, combining the plurality of swath images into a uniformized image where overlapping portions of the plurality of swath images are combined within the uniformized image, and generating one or more measurements of the sample based on the uniformized image.

A method including: obtaining a detected representation of radiation redirected by each of a plurality of structures from a substrate additionally having a device pattern thereon, wherein each structure has an intentional different physical configuration of the respective structure than the respective nominal physical configuration of the respective structure, wherein each structure has geometric symmetry at the respective nominal physical configuration, wherein the intentional different physical configuration of the structure causes an asymmetric optical characteristic distribution and wherein a patterning process parameter measures change in the physical configuration; and determining a value, based on the detected representations and based on the intentional different physical configurations, to setup, monitor or correct a measurement recipe for determining the patterning process parameter.

Described are systems and methods for predicting an overlay of a patterned wafer. Methods can include receiving a machine data set comprising parameters at a plurality of positions of a wafer, generating an integrated data set by aggregating the parameters and mapping the aggregated parameters to an equispatial grid, generating, using a trained machine learning (ML) model, a predicted overlay for the patterned wafer based at least in part on processing the integrated data set.