Method to check model accuracy during wafer patterning simulation
View Patent ↗A method, and computer program product and system for performing the method, is provided for designing a mask used in the manufacture of semiconductor integrated circuits, in which a model of the lithographic process is used during the mask design process. More particularly, the on-wafer process model is a function of optical image parameters that are calibrated using measurements from a test pattern. An uncertainty metric for the predicted response simulated by the on-wafer process model is computed for a given evaluation point of interest as a function of a distance metric between the collective optical image parameters simulated at the given evaluation point and the collective optical image parameters at the calibration data points. The uncertainty metric preferably is also a function of the sensitivity of the on-wafer process model response to changes in the optical image parameters.
1. A method for optimizing a lithographic process for manufacturing semiconductor integrated circuits, the lithographic process including using an empirically calibrated process model, the method comprising:
providing a calibration test structure;
computing an optical image for said calibration test structure using an optical model;
forming a printed image of said calibration test structure on a wafer using the lithographic process;
determining measurement data from said printed calibration test structure and optical image parameters from said optical image at selected calibration points of said calibration test structure;
providing a process model for determining a predicted on-wafer response value, said process model comprising a function of optical image parameters;
calibrating said process model so that said predicted on-wafer response value predicts said measurement data within a predetermined metric;
providing an uncertainty metric associated with said predicted on-wafer response value for a selected set of values of said optical image parameters, said uncertainty metric comprising a function of a distance metric between said selected set of values of said optical image parameters and said optical image parameters at said selected calibration points;
applying said process model to a non-calibration mask pattern to obtain a predicted non-calibration on-wafer response; and
determining whether said uncertainty metric associated with said predicted non-calibration on-wafer response violates a predetermined criterion.
2. The method according to claim 1 , wherein if said predetermined criterion is violated, then modifying said non-calibration mask pattern in accordance with said predetermined criterion.
3. The method according to claim 1 , wherein at least one of said optical image parameters are selected from the group consisting of maximum intensity, minimum intensity, image slope, image curvature, and integrated intensity.
4. The method of claim 1 , wherein said distance metric comprises a Euclidean distance between said optical image parameters obtain from said optical model at an evaluation point of interest and said optical image parameters from said optical image at said selected calibration points.
5. The method of claim 1 , wherein said uncertainty metric further comprises a function of a sensitivity metric that represents a rate of change of said predicted on-wafer response value as a function of a change in said optical image parameters along said distance metric.
6. The method of claim 5 , wherein said sensitivity metric comprises taking a partial derivative of said process model as a function of one of said optical image parameters.
7. The method of claim 1 , wherein said lithographic process is selected from the group consisting of a resist process and an etch process.
8. The method of claim 1 , wherein said uncertainty metric comprises determining a root-mean-square of said distance metric between said selected set of optical image parameters evaluated at an evaluation point of interest and said optical image parameters at said selected calibration points.
9. The method of claim 1 , further comprising determining a set of uncertainty values of said uncertainty metric over a range of selected values of said optical image parameters, and storing said set of uncertainty values in a look-up table.
10. The method of claim 1 , wherein said uncertainty metric further comprises a function of uncertainty in said measurement data.
11. A computer program product comprising computer readable storage medium having stored therein computer readable instructions which when executed by a computer cause the computer to perform method steps for optimizing a lithographic process for manufacturing semiconductor integrated circuits, the lithographic process including using an empirically calibrated process model, the method comprising:
providing a calibration test structure;
computing an optical image for said calibration test structure using an optical model;
forming a printed image of said calibration test structure on a wafer using the lithographic process;
determining measurement data from said printed calibration test structure and optical image parameters from said optical image at selected calibration points of said calibration test structure;
providing a process model for determining a predicted on-wafer response value, said process model comprising a function of optical image parameters;
calibrating said process model so that said predicted on-wafer response value predicts said measurement data within a predetermined metric;
providing an uncertainty metric associated with said predicted on-wafer response value for a selected set of values of said optical image parameters, said uncertainty metric comprising a function of a distance metric between said selected set of values of said optical image parameters and said optical image parameters at said selected calibration points;
applying said process model to a non-calibration mask pattern to obtain a predicted non-calibration on-wafer response; and
determining whether said uncertainty metric associated with said predicted non-calibration on-wafer response violates a predetermined criterion.
12. The computer program product according to claim 11 , wherein if said predetermined criterion is violated, then said method further comprising modifying said non-calibration mask pattern in accordance with said predetermined criterion.
13. The computer program product according to claim 11 , wherein at least one of said optical image parameters are selected from the group consisting of maximum intensity, minimum intensity, image slope, image curvature, and integrated intensity.
14. The computer program product of claim 11 , wherein said distance metric comprises a Euclidean distance between said optical image parameters obtain from said optical model at an evaluation point of interest and said optical image parameters from said optical image at said selected calibration points.
15. The computer program product of claim 11 , wherein said uncertainty metric further comprises a function of a sensitivity metric that represents a rate of change of said predicted on-wafer response value as a function of a change in said optical image parameters along said distance metric.
16. The computer program product of claim 15 , wherein said sensitivity metric comprises taking a partial derivative of said process model as a function of one of said optical image parameters.
17. The computer program product of claim 11 , wherein said lithographic process is selected from the group consisting of a resist process and an etch process.
18. The computer program product of claim 11 , wherein said uncertainty metric comprises determining a root-mean-square of said distance metric between said selected set of optical image parameters evaluated at an evaluation point of interest and said optical image parameters at said selected calibration points.
19. The computer program product of claim 11 , further comprising determining a set of uncertainty values of said uncertainty metric over a range of selected values of said optical image parameters, and storing said set of uncertainty values in a look-up table.
20. The computer program product of claim 11 , wherein said uncertainty metric further comprises a function of uncertainty in said measurement data.