Process window analysis

1. A method comprising:

acquiring, by a processor, first inspection data collected by using a first inspection modality during a first inspection process of a substrate having multiple instances of a pattern of features, wherein each of the instances of the pattern of features is formed by a fabrication process utilizing different respective sets of process parameters associated with the fabrication process of the substrate;

identifying, by the processor, defects at locations in the instances of the pattern based on the first inspection data;

analyzing, by the processor, characteristics of the defects identified in the first inspection data to select a first set of the locations at which the first inspection data is indicative of an influence of the process parameters associated with the fabrication process on the defects;

responsive to selecting the first set of the locations, causing, by the processor, a second imaging modality to acquire second inspection data of the substrate at the first set of the locations during a second inspection process, wherein the second inspection modality corresponds to a finer resolution than the first inspection modality;

analyzing, by the processor, one or more defects appearing in the second inspection data to select, from the first set of the locations, a second set of the locations at which the second inspection data is indicative of a range of the process parameters, wherein the process parameters comprise a focal adjustment and an exposure level applied in the fabrication process to form one or more of the instances of the pattern of features on the substrate; and

providing the range of the process parameters for fabrication of a target device that is to include the pattern.

2. The method according to claim 1 , wherein the first inspection modality is an optical imaging modality, and wherein the second inspection modality is an electron imaging modality.

3. The method according to claim 1 , wherein the substrate corresponds to a semiconductor wafer, and wherein the fabrication process comprises lithographic patterning of the semiconductor wafer, and the instances of the pattern correspond to dies defined by the lithographic patterning.

4. The method according to claim 1 , wherein analyzing the characteristics of the defects comprises:

clustering the defects into a plurality of bins corresponding to the locations of the defects in the instances of the pattern;

sorting the bins so as to prioritize the bins that are indicative of the influence of the process parameters on the defects; and

selecting the first set of the locations from among the prioritized bins.

5. The method according to claim 4 , wherein clustering the defects comprises defining the bins responsively to respective features of the design occurring in the bins.

6. The method according to claim 4 , wherein sorting the bins comprises:

identifying a correlation in one or more of the bins between changes in at least one of the process parameters and the defects in the one or more of the bins; and

prioritizing the one or more of the bins responsively to the identified correlation.

7. The method according to claim 1 , wherein analyzing the characteristics of the defects comprises:

computing, for the defects, corresponding values of a utility function that are indicative of an information content of the first inspection data at the locations of the defects; and

selecting the first set of the locations in response to the computing of the corresponding values of the utility function.

8. A system comprising:

a memory; and

a processor, operatively coupled with the memory, to:

acquire first inspection data collected by using a first inspection modality during a first inspection process of a substrate having multiple instances of a pattern of features, wherein each of the instances of the pattern of features is formed by a fabrication process utilizing different respective sets of process parameters associated with the fabrication process of the substrate;

identify defects at locations in the instances of the pattern based on the first inspection data;

analyze characteristics of the defects identified in the first inspection data to select a first set of the locations at which the first inspection data is indicative of an influence of the process parameters associated with the fabrication process on the defects;

responsive to selecting the first set of the locations, cause a second imaging modality to acquire second inspection data of the substrate at the first set of the locations during a second inspection process, wherein the second inspection modality corresponds to a finer resolution than the first inspection modality;

analyze one or more defects appearing in the second inspection data to select, from the first set of the locations, a second set of the locations at which the second inspection data is indicative of a range of the process parameters, wherein the process parameters comprise a focal adjustment and an exposure level applied in the fabrication process to form one or more of the instances of the pattern of features on the substrate; and

provide the range of the process parameters for fabrication of a target device that is to include the pattern.

9. The system according to claim 8 , wherein the first inspection modality is an optical imaging modality, and wherein the second inspection modality is an electron imaging modality.

10. The system according to claim 8 , wherein the substrate corresponds to a semiconductor wafer, and wherein the fabrication process comprises lithographic patterning of the semiconductor wafer, and the instances of the pattern correspond to dies defined by the lithographic patterning.

11. The system according to claim 8 , wherein the substrate is a focus/exposure matrix (FEM) wafer.

12. The system according to claim 8 , wherein the processor is further to:

cluster the defects into a plurality of bins corresponding to the locations of the defects in the instances of the pattern, to sort the bins so as to prioritize the bins that are indicative of the influence of the process parameters on the defects; and

select the first set of the locations from among the prioritized bins.

13. The system according to claim 12 , wherein the bins are defined responsively to respective features of the design occurring in the bins.

14. The system according to claim 12 , wherein the processor is further to:

identify a correlation in one or more of the bins between changes in at least one of the process parameters and the defects in the one or more of the bins; and

prioritize the one or more of the bins responsively to the identified correlation.

15. The system according to claim 8 , wherein the processor is further to:

compute, for the defects, corresponding values of a utility function that are indicative of an information content of the first inspection data at the locations of the defects; and

select the first set of the locations in response to the corresponding values of the utility function.

16. A non-transitory computer-readable medium storing instructions, which when executed by a processor, cause the processor to:

acquire first inspection data collected by using a first inspection modality during a first inspection process of a substrate having multiple instances of a pattern of features, wherein each of the instances of the pattern of features is formed by a fabrication process utilizing different respective sets of process parameters associated with the fabrication process of the substrate;

identify defects at locations in the instances of the pattern based on the first inspection data;

analyze characteristics of the defects identified in the first inspection data to select a first set of the locations at which the first inspection data is indicative of an influence of the process parameters associated with the fabrication process on the defects;

responsive to selecting the first set of the locations, cause a second imaging modality to acquire second inspection data of the substrate at the first set of the locations during a second inspection process, wherein the second inspection modality corresponds to a finer resolution than the first inspection modality;

analyze one or more defects appearing in the second inspection data to select, from the first set of the locations, a second set of the locations at which the second inspection data is indicative of a range of the process parameters, wherein the process parameters comprise a focal adjustment and an exposure level applied in the fabrication process to form one or more of the instances of the pattern of features on the substrate; and

provide the range of the process parameters for fabrication of a target device that is to include the pattern.