摘要:
Split machine learning systems can be used to generate an output for an input. When the input is received, a determination is made as to whether the input is within a first, second, or third range of values. If the input is within the first range, the output is generated using a first machine learning system. If the input is within the second range, the output is generated using a second machine learning system. If the input is within the third range, the output is generated using the first and second machine learning systems.
摘要:
A structure formed on a wafer can be examined by directing an incident pulse at the structure, the incident pulse being a sub-picosecond optical pulse. A diffraction pulse resulting from the incident pulse diffracting from the structure is measured. A characteristic of the profile of the structure is then determined based on the measured diffraction pulse.
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:
The profile of an integrated circuit structure is determined by obtaining a measured metrology signal and a first simulated metrology signal, which has an associated profile model of the structure defined by a set of profile parameters. When the two signals match within a first termination criterion, at least one profile parameter is selected from the set of profile parameters. A value for the selected profile parameter is determined. A second simulated metrology signal having an associated profile model of the structure defined by a set of profile parameters with at least one profile parameter equal or close to the determined value for the selected profile parameter is obtained. When the measured and the second simulated metrology signals match within a second termination criterion, values for one or more remaining profile parameters are determined from the set of profile parameters associated with the second simulated metrology signal.
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:
Metrology data from a semiconductor treatment system is transformed using multivariate analysis. In particular, a set of metrology data measured or simulated for one or more substrates treated using the treatment system is obtained. One or more essential variables for the obtained set of metrology data is determined using multivariate analysis. A first metrology data measured or simulated for one or more substrates treated using the treatment system is obtained. The first obtained metrology data is not one of the metrology data in the set of metrology data earlier obtained. The first metrology data is transformed into a second metrology data using the one or more of the determined essential variables.
摘要:
Metrology data from a semiconductor treatment system is transformed using multivariate analysis. In particular, a set of metrology data measured or simulated for one or more substrates treated using the treatment system is obtained. One or more essential variables for the obtained set of metrology data is determined using multivariate analysis. A first metrology data measured or simulated for one or more substrates treated using the treatment system is obtained. The first obtained metrology data is not one of the metrology data in the set of metrology data earlier obtained. The first metrology data is transformed into a second metrology data using the one or more of the determined essential variables.
摘要:
A process step in fabricating a structure on a wafer in a wafer application having one or more process steps and one or more process parameters is controlled by determining a correlation between a set of profile models and one or more key profile shape variables. Each profile model is defined using a set of profile parameters to characterize the shape of the structure. Different sets of profile parameters define the profile models in the set. The one or more key profile shape variables include one or more profile parameters or one or more process parameters. One profile model is selected from the set of profile models based on the correlation and a value of at least one key profile shape variable of the process of the wafer application to be used in fabricating the structure. The structure is fabricated in a first fabrication process cluster using the process step and the value of the at least one key profile shape variable. A measured diffraction signal is obtained off the structure. One or more profile parameters of the structure are determined based on the measured diffraction signal and the selected profile model. The one or more determined profile parameters are transmitted to the first fabrication process cluster or a second fabrication process cluster.
摘要:
The accuracy of a library of simulated-diffraction signals for use in optical metrology of a structure formed on a wafer is evaluated by utilizing an identity relationship inherent to simulated diffraction signals. Each simulated diffraction signal contains at least one set of four reflectivity parameters for a wavelength and/or angle of incidence. One of the four reflectivity parameters is selected. A value for the selected reflectivity parameter is determined using the identity relationship and values of the remaining three reflectivity parameters. The determined value for the selected reflectivity parameter is compared to the value in the obtained set of four reflectivity parameters to evaluate and improve the accuracy of the library. The identity relationship can also be used to reduce the data storage in a library.
摘要:
One or more profile parameters of a structure fabricated on a wafer in a wafer application are determined by developing a correlation between a set of profile models and one or more key profile shape variables. The wafer application has one or more process steps and one or more process parameters. Each profile model is defined using a set of profile parameters to characterize the shape of the structure. Different sets of profile parameters define the profile models in the set. The one or more key profile shape variables include one or more profile parameters or one or more process parameters. A value of at least one key profile shape variable of the process step of the wafer application to be used in fabricating the structure is determined. One profile model is selected from the set of profile models based on the determined correlation and the value of the at least one determined key profile shape variable. The structure is fabricated using the process step and the value of the at least one determined key profile shape variable determined. A measured diffraction signal off the fabricated structure is obtained. One or more profile parameters of the fabricated structure are determined based on the measured diffraction signal and the selected profile model.