Machine vision as input to a CMP process control algorithm
During chemical mechanical polishing of a substrate, a signal value that depends on a thickness of a layer in a measurement spot on a substrate undergoing polishing is determined by a first in-situ monitoring system. An image of at least the measurement spot of the substrate is generated by a second in-situ imaging system. Machine vision processing, e.g., a convolutional neural network, is used to determine a characterizing value for the measurement spot based on the image. Then a measurement value is calculated based on both the characterizing value and the signal value.
1. A polishing system, comprising:
a support to hold a polishing pad;
a carrier head to hold a substrate in contact with the polishing pad;
a motor to generate relative motion between the support and the carrier head;
a first in-situ monitoring system to generate a signal that depends on a thickness of a layer in a measurement spot on the substrate;
a second in-situ imaging system to generate an image comprising a plurality of pixels, the image covering at least the measurement spot of the substrate and being taken at substantially the same time as the first in-situ monitoring system generates the signal for the measurement spot on the substrate; and
a controller configured to
receive the image from the second in-situ imaging system,
determine, based on the image and using machine vision processing of the image, a characterizing value representing a classification of a portion of the substrate corresponding to the measurement spot, the classification corresponding to a type of structure on the substrate, wherein the type of structure is selected from a group of structure types including at least one of an array, a scribe line, a periphery, and a contact pad,
receive the signal from the first in-situ monitoring system,
generate a thickness measurement value based on both the characterizing value and the signal, wherein the controller has the characterizing value as an input to a conversion algorithm such that the classification influences the conversion of the signal to the thickness measurement value, and
at least one of halt polishing of the substrate or adjust a polishing parameter based on the thickness measurement value.
2. The system of claim 1 , wherein the machine vision processing comprises an artificial neural network.
3. The system of claim 2 , wherein the machine vision processing comprises a convolutional neural network.
4. The system of claim 2 , wherein the controller is configured to train the artificial neural network by backpropagation using training data including images and known characterizing values for the images.
5. The system of claim 2 , wherein the first in-situ monitoring system comprises a spectrographic monitoring system to generate a measured spectrum for the measurement spot.
6. The system of claim 2 , wherein the first in-situ monitoring system comprises an eddy current monitoring system to generate a signal value for the measurement spot.
7. The system of claim 6 , wherein the artificial neural network is configured to determine a geometry value for a feature that affects current flow in the measurement spot.
8. The system of claim 7 , wherein the geometry value includes at least one of a distance, size or orientation.
9. The system of claim 1 , wherein the controller is configured to determine a plurality of characterizing values for a plurality of different characteristics of the substrate at the measurement spot based on the image using the machine vision processing.
10. The polishing system of claim 1 , wherein the controller is configured to store a plurality of libraries of reference spectra, each respective library of reference spectra, having an associated respective classification, and to select a library from the plurality of libraries based on the classification represented by the characterizing value.
11. A computer program product for controlling processing of a substrate, the compute program product tangibly embodied in a non-transitory computer readable media and comprising instructions for causing a processor to:
receive, from a first in-situ monitoring system, a signal that depends on a thickness of a layer in a measurement spot on a substrate undergoing polishing;
receive image data comprising a plurality of pixels, the image data covering at least the measurement spot of the substrate from a second in-situ imaging system;
determine a characterizing value representing a classification of a portion of the substrate corresponding to the measurement spot based on the image data and using machine vision processing of the image data, the classification corresponding to a type of structure on the substrate, wherein the type of structure is selected from a group of structure types including at least one of an array, a scribe line, a periphery, and a contact pad;
generate a thickness measurement value based on both the characterizing value and the signal by inputting the characterizing value to a conversion algorithm such that the classification influences the conversion of the signal to the thickness measurement value, and
at least one of halt polishing of the substrate or adjust a polishing parameter based on the thickness measurement value.
12. The computer program product of claim 11 , comprising instructions to determine a portion of the image data corresponding to the measurement spot.
13. The computer program product of claim 11 , comprising instructions to synchronize image data from the second in-situ imaging system with signal values from the first in-situ monitoring system.
14. The computer program product of claim 11 , wherein the machine vision processing comprises feeding the image data to an artificial neural network.
15. The computer program product of claim 14 , wherein the artificial neural network comprises a convolutional neural network.
16. The computer program product of claim 11 , wherein the signal comprises a measured spectrum, and the instructions to generate the measurement value comprise instructions to select one of a plurality of libraries of reference spectra based on the classification.
17. The computer program product of claim 11 , wherein the instructions to perform machine vision processing comprise instructions to determine a geometry value of a feature in a portion of the substrate corresponding to the measurement spot.
18. The computer program product of claim 17 , wherein the signal comprises a signal value from an eddy current monitoring system, and the geometry value comprises an orientation of the feature.
19. The computer program product of claim 11 , wherein the instructions to generate the thickness measurement value comprise instructions to store a plurality of libraries of reference spectra, each respective library of reference spectra having an associated respective classification, and to select a library from the plurality of libraries based on the classification represented by the characterizing value.