The current disclosure describes techniques for managing vertical alignment or overlay in semiconductor manufacturing using machine learning. Alignments of interconnection features in a fan-out WLP process are evaluated and managed through the disclosed techniques. Big data and machine learning are used to train a classification that correlates the overlay error source factors with overlay metrology categories. The overlay error source factors include tool signals. The trained classification includes a base classification and a Meta classification.

Systems, apparatuses, and methods are provided for measuring intensity using off-axis illumination. An example method can include illuminating a region of a surface of a substrate with a first radiation beam at a first incident angle and, in response, measuring a first set of photons diffracted from the region. The example method can further include illuminating the region with a second radiation beam at a second incident angle and, in response, measuring a second set of photons diffracted from the region. The example method can further include generating measurement data for the region based on the measured first set of photons and the measured second set of photons.

In a lithographic process, product units such as semiconductor wafers are subjected to lithographic patterning operations and chemical and physical processing operations. Alignment data or other measurements are made at stages during the performance of the process to obtain object data representing positional deviation or other parameters measured at points spatially distributed across each unit. This object data is used to obtain diagnostic information by performing a multivariate analysis to decompose a set of vectors representing the units in the multidimensional space into one or more component vectors. Diagnostic information about the industrial process is extracted using the component vectors. The performance of the industrial process for subsequent product units can be controlled based on the extracted diagnostic information.

An overlay metrology system may receive images of an overlay target on a sample, where the overlay target includes Moir structures in which first-layer features and second-layer features partially overlap, where overlap regions include regions of overlap between the first-layer and second-layer features, where non-overlap regions include regions of non-overlap between the first-layer and second-layer features. The system may further, determine a coarse overlay measurement based at least in part on the one or more non-overlap regions of the one or more images, determine a fine overlay measurement based at least in part on the one or more overlap regions of the one or more images, and generate an output overlay measurement based on the coarse overlay measurement and the fine overlay measurement.

Disclosed is a method of determining a value for a parameter of interest from a target on a substrate. The method comprises obtaining metrology data comprising single-wavelength parameter of interest values which were obtained using a respective different measurement wavelength; and determining said value for the parameter of interest from a stack sensitivity derived weighted combination of said single-wavelength parameter of interest values. Also disclosed is a method of selecting wavelengths for a measurement based on at least the derivative of the stack sensitivity with respect to wavelength.

A method includes applying a first voltage to a source of a first transistor of a detection unit of a semiconductor detector in a test wafer and applying a second voltage to a gate of the first transistor and a drain of a second transistor of the detection unit. The first transistor is coupled to the second transistor in series, and the first voltage is higher than the second voltage. A pre-exposure reading operation is performed to the detection unit. Light of an exposure apparatus is illuminated to a gate of the second transistor after applying the first and second voltages. A post-exposure reading operation is performed to the detection unit. Data of the pre-exposure reading operation is compared with the post-exposure reading operation. An intensity of the light is adjusted based on the compared data of the pre-exposure reading operation and the post-exposure reading operation.

The invention relates to a method for evaluating at least one marker (200) on a physical object (50) for metrological detection of the object (50), comprising the following steps: providing (101) at least one rendering of the marker (200), wherein the at least one rendering results in each case from a sensor detection, determining (102) a reference marking (210) of the marker (200) on the basis of the provided rendering of the marker (200), wherein the reference marking (210) has at least two partial areas (220) of different display properties, wherein the determining (102) of the reference marking (210) comprises the following steps: transforming (103) into one or more color spaces in the provided rendering of the marker (200) to display the at least two partial areas (220) of different display property with essentially the same display property, recognizing (104) the reference marking (210) on the basis of the partial areas (220) displayed with essentially the same display property, providing (105) a display of the at least two partial areas (220) with the different display property, recognizing (106) at least one reference point of the reference marking (210) based on the provided display having the different display property.

Disclosed is a method of correcting a measured spectrum for the effects of a source spectrum resulting from an illumination source. The method comprises obtaining a measured spectrum in terms of a measurement parameter, the measured spectrum being obtained from captured diffracted radiation from a periodic structure following illumination of said periodic structure using source radiation from said illumination source, the periodic structure being the spectrometer grating and an object being measured; determining an estimate of the source spectrum from the measured spectrum; and correcting the measured spectrum using the estimate of the source spectrum.

A method can include directing radiation toward at least two targets using an optical scanning system so as to generate first and second portions of scattered radiation. A first target can include a plurality of first grating line structures including features having a first bias value. A second target can include a plurality of second grating line structures including features having a second bias value. The method can include detecting the first and second portions of scattered radiation, generating a first measurement signal indicative of a first target position based on the first bias features, and generating a second measurement signal indicative of a second target position based on the second bias features. The method can include analyzing an effect of the first and second bias values on the first and second positions to determine at least one property of the set of targets.

A method for checking a lithography mask for a repair of the lithography mask, the lithography mask having a plurality of edges between partial regions of the lithography mask and the object of the repair lying in an adjustment of a profile of a selected edge in a repair portion of the selected edge, comprises: a) capturing an image representation of a repair region of the lithography mask comprising the repair portion of the selected edge, b) determining the profile of the selected edge in the repair portion on the basis of the captured image representation of the repair region, b1) determining a reference profile on the basis of a profile of an edge corresponding to the selected edge, the corresponding edge being an edge which should not be repaired or a portion of the selected edge which should not be repaired, the corresponding edge being determined on the basis of the captured image representation of the repair region, and c) comparing the determined profile of the selected edge with a reference profile.