摘要:
Embodiments are generally directed to neural network training for library-based critical dimension metrology. An embodiment of a method includes optimizing a threshold for a principal component analysis of a spectrum data set to provide a principal component value, estimating a training target for one or more neural networks, training the one or more neural networks based both on the training target and on the principal component value provided from optimizing the threshold for the principal component analysis, and providing a spectral library based on the one or more trained neural networks.
摘要:
Approaches for accurate neural network training for library-based critical dimension (CD) metrology are described. Approaches for fast neural network training for library-based CD metrology are also described. In an example, a method includes optimizing a threshold for a principal component analysis (PCA) of a spectrum data set to provide a principal component (PC) value, estimating a training target for one or more neural networks, training the one or more neural networks based both on the training target and on the PC value provided from optimizing the threshold for the PCA, and providing a spectral library based on the one or more trained neural networks.
摘要:
Approaches for accurate neural network training for library-based critical dimension (CD) metrology are described. Approaches for fast neural network training for library-based CD metrology are also described.
摘要:
A structure formed on a semiconductor wafer is examined by obtaining a first diffraction signal measured using a metrology device. A second diffraction signal is generated using a machine learning system, where the machine learning system receives as an input one or more parameters that characterize a profile of the structure to generate the second diffraction signal. The first and second diffraction signals are compared. When the first and second diffraction signals match within a matching criterion, a feature of the structure is determined based on the one or more parameters or the profile used by the machine learning system to generate the second diffraction signal.
摘要:
To manage data flow in generating different signal formats for use in optical metrology, a project data object is created. A first option data object is created. The first option data object has a set of signal parameters. Different settings of the set of signal parameters correspond to different signal formats for diffraction signals. A version number is associated with the first option data object. The first option data object is linked with the project data object. At least a second option data object is created. The second option data object has a set of signal parameters. Different settings of the set of signal parameters correspond to different signal formats for diffraction signals. The set of signal parameters of the first option data object and the set of signal parameters of the second option data object are set differently. Another version number is associated with the second option data object. The second option data object is linked with the project data object. The project data object, the first option data object, and the second option data object are stored. The version numbers associated with the first option data object and the second option data object are stored. The link between the first option data object and the project data object is stored. The link between the second option data object and the project data object is stored.
摘要:
In allocating processing units of a computer system to generate simulated diffraction signals used in optical metrology, a request for a job to generate simulated diffraction signals using multiple processing units is obtained. A number of processing units requested for the job to generate simulated diffraction signals is then determined. A number of available processing units is determined. When the number of processing units requested is greater than the number of available processing units, a number of processing units is assigned to generate the simulated diffraction signals that is less than the number of processing units requested.
摘要:
A system for examining a patterned structure formed on a semiconductor wafer using an optical metrology model includes a first fabrication cluster, a metrology cluster, an optical metrology model optimizer, and a real time profile estimator. The first fabrication cluster configured to process a wafer, the wafer having a first patterned and a first unpatterned structure. The first patterned structure has underlying film thicknesses, critical dimension, and profile. The metrology cluster including one or more optical metrology devices coupled to the first fabrication cluster. The metrology cluster is configured to measure diffraction signals off the first patterned and the first unpatterned structure. The metrology model optimizer is configured to optimize an optical metrology model of the first patterned structure using one or more measured diffraction signals off the first patterned structure and with floating profile parameters, material refraction parameters, and metrology device parameters.
摘要:
An optical metrology model for a structure to be formed on a wafer is developed by characterizing a top-view profile and a cross-sectional view profile of the structure using profile parameters. The profile parameters of the top-view profile and the cross-sectional view profile are integrated together into the optical metrology model. The profile parameters of the optical metrology model are saved.
摘要:
A structure formed on a semiconductor wafer is examined by obtaining a first diffraction signal measured from the structure using an optical metrology device. A first profile is obtained from a first machine learning system using the first diffraction signal obtained as an input to the first machine learning system. The first machine learning system is configured to generate a profile as an output for a diffraction signal received as an input. A second profile is obtained from a second machine learning system using the first profile obtained from the first machine learning system as an input to the second machine learning system. The second machine learning system is configured to generate a diffraction signal as an output for a profile received as an input. The first and second profiles include one or more parameters that characterize one or more features of the structure.
摘要:
An optical metrology model for a repetitive structure is optimized by selecting one or more profile parameters using one or more selection criteria. One or more termination criteria are set, the one or more termination criteria comprising measures of stability of the optical metrology model. The profile shape features of the repetitive structure are characterized using the one or more selected profile parameters. The optical metrology model is optimized using a set of values for the one or more selected profile parameters. One or more profile parameters of the profile of the repetitive structure are determined using the optimized optical metrology model and one or more measured diffraction signals. Values of the one or more termination criteria are calculated using the one or more determined profile parameters. When the calculated values of the one or more termination criteria do not match the one or more set termination criteria, the selection of the one or more profile parameters and/or the characterization of the profile shape features of the repetitive structure are revised.