Various embodiments set forth techniques for characterizing films on optically clear substrates using ellipsometry. In some embodiments, a spectroscopic ellipsometer is configured to generate a light beam that has a relatively small spot size and is substantially absorbed by an optically clear substrate, thereby reducing or eliminating reflections from an interface between the substrate and air. Optical simulations can be performed to determine values for various parameters associated with the ellipsometer that minimize the reflections from the interface between the substrate and air and maximize reflections from an interface between a film and the substrate. In addition, graded films that include multiple layers can be analyzed using models of multiple layers.

A method for determining properties of a coating on a transparent film, a method for manufacturing a capacitor film and a device configured to determine properties of a coating on a transparent film are disclosed. In an embodiment a method includes moving the transparent film with the coating on a path which passes between a light source and a sensor, illuminating, by the light source, the coating on the transparent film, detecting, by the sensor, an intensity of transmitted light from the light source and calculating, by a processor, the properties of the coating on the transparent film based on the detected intensity of transmitted light.

A system and methods for OCD metrology are provided including receiving training data for training an OCD machine learning (ML) model, including multiple pairs of corresponding sets of scatterometric data and reference parameters. For each of the pairs, one or more corresponding outlier metrics are by calculated and corresponding outlier thresholds are applied whether a given pair is an outlier pair. The OCD MIL model is then trained with the training data less the outlier pairs.

A system and methods for OCD metrology are provided including receiving multiple first sets of scatterometric data, dividing each set into k sub-vectors, and training, in a self-supervised manner, k2 auto-encoder neural networks that map each of the k sub-vectors to each other. Subsequently multiple respective sets of reference parameters and multiple corresponding second sets of scatterometric data are received and a transfer neural network (NN) is trained. Initial layers include a parallel arrangement of the k2 encoder neural networks. Target output of the transfer NN training is set to the multiple sets of reference parameters and feature input is set to the multiple corresponding second sets of scatterometric data, such that the transfer NN is trained to estimate new wafer pattern parameters from subsequently measured sets of scatterometric data.

A substrate inspection system includes an imaging unit provided in a substrate processing apparatus and configured to acquire image data by imaging a surface of a substrate for color information on which a film is formed; a film thickness measurement unit provided in the substrate processing apparatus and configured to measure a film thickness of a substrate for film thickness measurement on which a film is formed under same conditions as on the substrate for color information; and a model creation unit configured to create a film thickness model corresponding to a correlation between information about color change on the surface of the substrate for color information caused by forming the film, which is acquired based on the image data, and the film thickness of the substrate for film thickness measurement, which is measured by the film thickness measurement unit.

Described are systems and techniques directed to optical inspection of moving products (wafers, substrates, films, patterns) that are being transported to or from processing chambers in device manufacturing systems. Implementations include a system that has a first source of light to direct a first light to a first location on a surface of a product. The first light generates, at the first location, a first reflected light. The system further includes a first optical sensor to generate a first data representative of a first reflected light, and a processing device, in communication with the first optical sensor to determine, using the first data, a property of the product.

An apparatus for measuring the thickness of graphene oxide coatings deposited on a support substrate are described. The apparatus includes a light source and a photodetector which can be placed directly into a coating line to provide continuous feedback on the thickness of a fabricated graphene oxide coating, enabling fabrication of controlled thickness coatings and real-time quality monitoring.

A measuring unit of a measuring apparatus includes a light source that emits light in a predetermined wavelength region, a condenser lens that applies the light emitted by the light source, to a plate-shaped workpiece held by a chuck table, a collimating lens that forms return light reflected by the plate-shaped workpiece into parallel light, a transmission filter that transmits interference light of the return light formed into the parallel light, a sensor that has coordinates for receiving the interference light transmitted through the transmission filter and detecting light intensity, and a controller that determines a coordinate position at which the light intensity detected by the sensor is high, as the thickness or height of the plate-shaped workpiece.

A coating system may include a coating chamber; a substrate holder to move a substrate along a motion path; and a sensor device in the coating chamber, wherein the sensor device is configured to move along the motion path, and wherein the sensor device is to perform a spectral measurement on the substrate.

A plasma processing apparatus and method with an improved processing yield, the plasma processing apparatus including detector configured to detect an intensity of a first light of a plurality of wavelengths in a first wavelength range and an intensity of a second light of a plurality of wavelengths in a second wavelength range, the first light being obtained by receiving a light which is emitted into the processing chamber from a light source disposed outside the processing chamber and which is reflected by an upper surface of the wafer, and the second light being a light transmitted from the light source without passing through the processing chamber; and a determination unit configured to determine a remaining film thickness of the film layer by comparing the intensity of the first light corrected using a change rate of the intensity of the second light.