Machine learning based inverse optical proximity correction and process model calibration

1. A method comprising:

obtaining a first patterning device pattern from simulation of an inverse lithographic process that predicts, from a substrate target layout, an output patterning device pattern, the output patterning device pattern corresponding to a pattern configured to be transferred, by a lithographic apparatus, from a patterning device onto a substrate with the aim to form the substrate target layout;

receiving substrate data corresponding to a substrate exposed using the first patterning device pattern; and

training, by a hardware processor system, an inverse process model configured to predict a second patterning device pattern using the substrate data related to the exposed substrate and the first patterning device pattern,

wherein the training the inverse process model is iterative, an iteration comprising:

determining one or more model parameter values of the inverse process model based on the substrate data and the first patterning device pattern; and

adjusting the one or more model parameter values until a first cost function of the inverse process model is improved.

2. The method of claim 1 , wherein the inverse process model is a machine learning model configured to predict the second patterning device pattern using the substrate target layout as input.

3. The method of claim 1 , wherein the inverse process model is a convolutional neural network.

4. The method of claim 1 , wherein the first cost function is a difference between the first patterning device pattern and a predicted patterning device pattern obtained from the inverse process model.

5. The method of claim 1 , further comprising predicting, using the trained inverse process model, a patterning device pattern based on a given substrate target layout.

6. A computer program product comprising a non-transitory computer readable medium having instructions therein, the instructions, when executed by a computer system, configured to cause the computer system to at least:

obtain a first patterning device pattern from simulation of an inverse lithographic process that predicts, from a substrate target layout, an output patterning device pattern, the output patterning device pattern corresponding to a pattern configured to be transferred, by a lithographic apparatus, from a patterning device onto a substrate with the aim to form the substrate target layout;

receive substrate data corresponding to a substrate exposed using the first patterning device pattern; and

train an inverse process model configured to predict a second patterning device pattern using the substrate data related to the exposed substrate and the first patterning device pattern,

wherein the training of the inverse process model is iterative, an iteration comprising:

determination of one or more model parameter values of the inverse process model based on the substrate data and the first patterning device pattern; and

adjustment of the one or more model parameter values until a first cost function of the inverse process model is improved.

7. The computer program product of claim 6 , wherein the inverse process model is a machine learning model configured to predict the second patterning device pattern using the substrate target layout as input.

8. The computer program product of claim 6 , wherein the instructions are further configured to cause the computer system to predict, using the trained inverse process model, a patterning device pattern based on a given substrate target layout.

9. A computer program product comprising a non-transitory computer readable medium having instructions therein, the instructions, when executed by a computer system, configured to cause the computer system to at least:

obtain an initial patterning device pattern from a trained inverse process model that predicts, from a substrate target layout, the initial patterning device pattern, wherein the initial patterning device pattern corresponds to a pattern to be transferred, by a lithographic apparatus, from a patterning device onto a substrate with the aim to form the substrate target layout; and

determine, by computer simulation using a calibrated process model, a patterning device layout based on the initial patterning device pattern and a substrate target layout,

wherein the determination of the patterning device layout is iterative, an iteration comprising:

determination, by computer simulation using the calibrated process model, a simulated pattern based on the initial patterning device pattern;

evaluation of a cost function, wherein the cost function computes a difference between the simulated pattern and the substrate target layout; and

adjustment of the initial patterning device pattern such that the cost function is reduced.

10. A method comprising:

obtaining an initial patterning device pattern from a trained inverse process model that predicts, from a substrate target layout, the initial patterning device pattern, wherein the initial patterning device pattern corresponds to a pattern to be transferred, by a lithographic apparatus, from a patterning device onto a substrate with the aim to form the substrate target layout; and

determining, by computer simulation using a calibrated process model, a patterning device layout based on the initial patterning device pattern and a substrate target layout,

wherein the determining the patterning device layout is iterative, an iteration comprising:

determining, by computer simulation using the calibrated process model, a simulated pattern based on the initial patterning device pattern;

evaluating a cost function, wherein the cost function computes a difference between the simulated pattern and the substrate target layout; and

adjusting the initial patterning device pattern such that the cost function is reduced.

11. The method of claim 10 , wherein the calibrated process model is calibrated based on substrate data related to the exposed substrate and the patterning device pattern from a simulation of an inverse lithographic process.

12. The method of claim 10 , wherein the initial patterning device pattern is a final patterning device layout that does not require adjustment to the initial patterning device pattern.

13. The computer program product of claim 9 , wherein the calibrated process model is calibrated based on substrate data related to the exposed substrate and the patterning device pattern from a simulation of an inverse lithographic process.

14. A method comprising:

obtaining a patterning device pattern from simulation of an inverse lithographic process that predicts the patterning device pattern based on a substrate target layout, wherein the patterning device pattern corresponds to a pattern configured to be transferred, by a lithographic apparatus, from a patterning device onto a substrate with the aim to form the substrate target layout,

wherein the simulation of the inverse lithographic process is an iterative process, an iteration comprising:

determining, via simulation, a simulated substrate pattern based on an initial patterning device pattern;

evaluating a cost function, wherein the cost function computes a difference between the simulated substrate pattern and the substrate target layout; and

adjusting the initial patterning device pattern such that the cost function is reduced;

receiving, by a hardware processor system, substrate data corresponding to a substrate exposed using the patterning device pattern; and

calibrating, by the hardware processor system, a process model of a patterning process based on the substrate data related to the exposed substrate and the patterning device pattern.

15. The method of claim 14 , wherein the calibrating the process model is an iterative process, an iteration comprising:

determining values of model parameters of the process model based on the substrate data and the patterning device pattern; and

adjusting the values of the model parameters until a cost function of the process model is improved.

16. The method of claim 15 , wherein the cost function is a difference between the substrate data and a predicted pattern obtained from the calibrated process model.

17. The method of claim 16 , wherein the difference is measured in terms of a performance parameter of the patterning process, the performance parameter including at least one selected from: a contour of a feature, critical dimension, and/or a process window.

18. The method of claim 14 , wherein the process model is a mask model, an optical model, a resist model, and/or an etch model.

19. The method of claim 14 , wherein the process model is a physics based model and/or a machine learning model.

20. A computer program product comprising a non-transitory computer readable medium having instructions therein, the instructions, when executed by a computer system, configured to cause the computer system to at least:

obtain a patterning device pattern from simulation of an inverse lithographic process that predicts the patterning device pattern based on a substrate target layout, wherein the patterning device pattern corresponds to a pattern configured to be transferred, by a lithographic apparatus, from a patterning device onto a substrate with the aim to form the substrate target layout,

wherein the simulation of the inverse lithographic process is an iterative process, an iteration comprising:

determination, via simulation, of a simulated substrate pattern based on an initial patterning device pattern;

evaluation of a cost function, wherein the cost function computes a difference between the simulated substrate pattern and the substrate target layout; and

adjustment of the initial patterning device pattern such that the cost function is reduced;

receive substrate data corresponding to a substrate exposed using the patterning device pattern; and

calibrate a process model of a patterning process based on the substrate data related to the exposed substrate and the patterning device pattern.