Lithography system and a machine learning controller for such a lithography system
A lithography system configured to apply a pattern to a substrate, the system including a lithography apparatus configured to expose a layer of the substrate according to the pattern, and a machine learning controller configured to control the lithography system to optimize a property of the pattern, the machine learning controller configured to be trained on the basis of a property measured by a metrology unit configured to measure the property of the exposed pattern in the layer and/or a property associated with exposing the pattern onto the substrate, and to correct lithography system drift by adjusting one or more selected from: the lithography apparatus, a track unit configured to apply the layer on the substrate for lithographic exposure, and/or a control unit configured to control an automatic substrate flow among the track unit, the lithography apparatus, and the metrology unit.
1. A lithography system configured to apply a pattern to a substrate, comprising:
a track unit configured to apply a layer on the substrate for lithographic exposure;
a lithography apparatus configured to expose the layer according to the pattern;
a metrology unit configured to measure a property of the exposed pattern in the layer and/or measure a property associated with exposing the pattern onto the substrate;
a control unit configured to control an automatic substrate flow among the track unit, the lithography apparatus, and the metrology unit; and
a machine learning controller configured to control the lithography system to optimize a property of the pattern, the machine learning controller configured to be trained on the basis of the measured property and to correct lithography system drift by adjusting one or more selected from: the lithography apparatus, the track unit and/or the control unit.
2. A lithography system configured to apply a pattern to a substrate, the system comprising:
a lithography apparatus configured to expose a layer of the substrate according to the pattern; and
a machine learning controller configured to control the lithography system to optimize a property of the pattern, the machine learning controller configured to be trained on the basis of a property measured by a metrology unit configured to measure a property of the exposed pattern in the layer and/or a property associated with exposing the pattern onto the substrate, and to correct lithography system drift by adjusting one or more selected from: the lithography apparatus, a track unit configured to apply the layer on the substrate for lithographic exposure, and/or a control unit configured to control an automatic substrate flow among the track unit, the lithography apparatus, and the metrology unit.
3. The lithography system according to claim 2 , further comprising:
the track unit configured to apply the layer on the substrate for lithographic exposure;
the metrology unit configured to measure the property of the exposed pattern in the layer; and
the control unit configured to control the automatic substrate flow among the track unit, the lithography apparatus, and the metrology unit.
4. The lithography system according to claim 2 , wherein the lithography system drift comprises at least lithography apparatus drift, track unit drift, control unit drift, and/or metrology unit drift.
5. The lithography system according to claim 2 , wherein the machine learning controller is configured to be trained on the basis of at least two lots of substrates.
6. The lithography system according to claim 2 , wherein the machine learning controller comprises a first controller configured to control overlay between pattern layers and/or a second controller configured to control a critical dimension of the pattern.
7. The lithography system according to claim 6 , wherein the machine learning controller comprises both the first and second controller, wherein the first controller comprises a first sub-controller configured to generate a first drift control signal for a pattern overlay, wherein the second controller comprises a third sub-controller configured to generate a second drift control signal for a critical dimension of the pattern, and wherein the machine learning controller is configured to generate an edge-to-edge placement signal which is a combination of the first and second drift control signals.
8. The lithography system according to claim 7 , wherein the edge-to-edge placement signal is a warning signal, and wherein the machine learning controller is configured to control the lithography apparatus, track unit and/or control unit on the basis of the warning signal if it exceeds a threshold level.
9. The lithography system according to claim 8 , wherein the edge-to-edge placement signal is a third drift control signal, wherein the machine learning controller is configured to control the lithography apparatus, track unit and/or control unit with the third drift control signal.
10. The lithography system according to claim 7 , wherein the machine learning controller is configured to deduce at least one of the drift control signals from the measured property and to correct the lithography system for drift by adjusting the lithography apparatus, track unit and/or control unit with the deduced drift control signal.
11. The lithography system according to claim 10 , wherein the machine learning controller is configured to deduce the at least one drift control signal also on the basis of lithography apparatus information, the lithography apparatus information including at least one selected from: information about a substrate chuck of the lithography apparatus used for exposure, information about the dynamics of a patterning device support of the lithography apparatus, information about the dynamics of a substrate stage of the lithography apparatus, information about substrate alignment, information about substrate leveling, information about an optical property of a projection system of the lithography apparatus, and/or information about a parameter or property associated with exposing the pattern of the patterning device onto a substrate.
12. The lithography system according to claim 10 , wherein the machine learning controller is configured to deduce the at least one drift control signal also on the basis of substrate process information, the substrate process information including at least one selected from: spin coating information, baking information, etching information and/or the sequence of the substrate in the lot of substrates.
13. The lithography system according to claim 10 , wherein the machine learning controller is configured to deduce the at least one drift control signal also on the basis of plant information regarding the plant housing the lithography system, the plant information including environmental data comprising at least one selected from: temperature in the plant and/or humidity in the plant.
14. The lithography system according to claim 10 , wherein the first controller comprises a second sub-controller configured to deduce a real-time overlay control signal and the second controller comprises a fourth sub-controller configured to deduce a real-time critical dimension control signal, wherein the real-time control signals are based on lithography apparatus information including at least one selected from: information about a substrate chuck of the lithography apparatus used for exposure, information about the dynamics of a patterning device support of the lithography apparatus, information about the dynamics of a substrate stage of the lithography apparatus, information about substrate alignment, information about substrate leveling, information about an optical property of a projection system of the lithography apparatus and/or information about a parameter or property associated with exposing the pattern of the patterning device onto a substrate, wherein the real-time control signals correspond to substrate-to-substrate lithography system variation, and wherein the machine learning controller is configured to correct the lithography apparatus, track unit and/or control unit for the substrate-to-substrate lithography system variation with the real-time control signals, and wherein the machine learning controller is configured to deduce the real-time control signals also on the basis of the measured property.
15. A machine learning controller for use in a lithography system, the machine learning controller configured to control the lithography system to optimize a property of a pattern to be exposed on a substrate, the machine learning controller configured to be trained on the basis of a property measured by a metrology unit configured to measure a property of the exposed pattern in the layer and/or a property associated with exposing the pattern onto the substrate, and to correct lithography system drift by adjusting one or more selected from: a lithography apparatus configured to expose a layer of the substrate according to the pattern, a track unit configured to apply the layer on the substrate for lithographic exposure, and/or a control unit configured to control an automatic substrate flow among the track unit, the lithography apparatus, and the metrology unit.
16. The machine learning controller according to claim 15 , wherein the lithography system drift comprises at least lithography apparatus drift, track unit drift, control unit drift, and/or metrology unit drift.
17. The machine learning controller according to claim 15 , wherein the machine learning controller is configured to be trained on the basis of at least two lots of substrates.
18. The machine learning controller according to claim 15 , wherein the machine learning controller comprises a first controller configured to control overlay between pattern layers and/or a second controller configured to control a critical dimension of the pattern.
19. The machine learning controller according to claim 18 , wherein the machine learning controller comprises both the first and second controller, wherein the first controller comprises a first sub-controller configured to generate a first drift control signal for a pattern overlay, wherein the second controller comprises a third sub-controller configured to generate a second drift control signal for a critical dimension of the pattern, and wherein the machine learning controller is configured to generate an edge-to-edge placement signal which is a combination of the first and second drift control signals.
20. The machine learning controller according to claim 19 , wherein the edge-to-edge placement signal is a warning signal, and wherein the machine learning controller is configured to control the lithography apparatus, track unit and/or control unit on the basis of the warning signal if it exceeds a threshold level.
21. The machine learning controller according to claim 19 , wherein the machine learning controller is configured to deduce at least one of the drift control signals from the measured property and to correct the lithography system for drift by adjusting the lithography apparatus, track unit and/or control unit with the deduced drift control signal.