Detection of dishing and tilting using X-ray fluorescence
A method for testing a material applied to a surface of a sample includes directing an excitation beam, having a known beam-width and intensity cross-section, onto a region of the sample. An intensity of X-ray fluorescence emitted from the region responsively to the excitation beam is measured. A distribution of the material within the region is estimated, responsively to the measured intensity of the X-ray fluorescence and to the intensity cross-section of the excitation beam, with a spatial resolution that is finer than the beam-width.
1. A method for testing a material applied to a surface of a sample, comprising:
directing an excitation beam, having a known beam-width and intensity cross-section, onto a region of the sample;
measuring an intensity of X-ray fluorescence emitted from the region responsively to the excitation beam;
estimating, responsively to the measured intensity of the X-ray fluorescence and to the intensity cross-section of the excitation beam, a distribution of the material within the region with a spatial resolution that is finer than the beam-width; and
outputting an identification of a defect in the region of the sample based on the estimated distribution of the material.
2. The method according to claim 1 , wherein the sample comprises a semiconductor wafer, wherein the region comprises a metal-filled feature on the wafer, and wherein estimating the distribution comprises identifying the defect in the feature.
3. The method according to claim 2 , wherein the defect comprises at least one of a dishing effect and a tilting effect.
4. The method according to claim 1 , wherein directing the excitation beam comprises scanning the beam over a feature on the surface, and wherein measuring the intensity comprises producing a measured XRF profile of the scanned feature.
5. The method according to claim 4 , wherein measuring the intensity comprises subtracting a background noise from the measured XRF profile.
6. The method according to claim 4 , wherein estimating the distribution comprises calculating a convolution between a simulated profile of the feature and a beam model representing the intensity cross section of the excitation beam, so as to produce a simulated XRF profile.
7. The method according to claim 6 , wherein the beam model comprises at least one Gaussian function.
8. The method according to claim 4 , wherein estimating the distribution comprises fitting a simulated XRF profile to the measured XRF profile.
9. The method according to claim 8 , wherein fitting the simulated profile comprises applying an iterative optimization process to the simulated XRF profile of the feature.
10. The method according to claim 9 , wherein applying the iterative process comprises calculating a Figure-of-Merit (FOM) function, so as to quantify a difference between the measured XRF profile and the simulated XRF profile.
11. The method according to claim 9 , wherein applying the iterative process comprises applying a Genetic Algorithm (GA).
12. The method according to claim 1 , wherein the spatial resolution of the estimated distribution is finer than one-half the beam-width.
13. Apparatus for testing a material applied to a surface of a sample, comprising:
a radiation source, which is coupled to direct an excitation beam, having a known beam-width and intensity cross-section, onto a region of the sample;
an array of detectors, which are coupled to measure an intensity of X-ray fluorescence emitted from the region responsively to the excitation beam; and
a processor, which is configured to estimate, responsively to the measured intensity of the X-ray fluorescence and to the intensity cross-section of the excitation beam, a distribution of the material within the region with a spatial resolution that is finer than the beam-width.
14. The apparatus according to claim 13 , wherein the sample comprises a semiconductor wafer, wherein the region comprises a metal-filled feature on the wafer, and wherein the processor is configured to identify a defect in the feature.
15. The apparatus according to claim 14 , wherein the defect comprises at least one of a dishing effect and a tilting effect.
16. The apparatus according to claim 13 , wherein the radiation source is arranged to scan the beam over a feature on the surface, and wherein the array of detectors is arranged to produce a measured XRF profile of the scanned feature.
17. The apparatus according to claim 16 , wherein the processor is configured to subtract a background noise from the measured XRF profile.
18. The apparatus according to claim 16 , wherein the processor is configured to calculate a convolution between a simulated profile of the feature and a beam model representing the intensity cross section of the excitation beam, so as to produce a simulated XRF profile of the feature.
19. The apparatus according to claim 18 , wherein the beam model comprises at least one Gaussian function.
20. The apparatus according to claim 16 , wherein the processor is configured to fit a simulated XRF profile to the measured XRF profile, so as to estimate the distribution of the material.
21. The apparatus according to claim 20 , wherein the processor is configured to apply an iterative optimization process to the simulated XRF profile of the feature, so as to fit the simulated XRF profile to the measured XRF profile.
22. The apparatus according to claim 21 , wherein the processor is configured to calculate a Figure-of-Merit (FOM) function, so as to quantify a difference between the measured XRF profile and the simulated XRF profile.
23. The apparatus according to claim 21 , wherein the iterative optimization process comprises a Genetic Algorithm (GA).
24. The apparatus according to claim 13 , wherein the spatial resolution of the estimated distribution is finer than one-half the beam-width.
25. A computer software product for testing a material applied to a surface of a sample, the product comprising a computer-readable medium, in which program instructions are stored, which instructions, when read by the computer, cause the computer to receive an intensity cross-section of an excitation beam, which is used to excite a region of the sample, to receive a measurement of an intensity of X-ray fluorescence emitted from the region responsively to the excitation beam, to estimate, responsively to the measurement of the intensity of the X-ray fluorescence and to the intensity cross-section of the excitation beam, a distribution of the material within the region with a spatial resolution that is finer than the beam-width, and to output an identification of a defect in the region of the sample based on the estimated distribution of the material.