Real-time effective-wavelength error correction for HDVSI
View Patent ↗An error correction for effective-wavelength variations is implemented by adjusting the filter parameters of the quadrature demodulation algorithm of a high definition vertical scanning process using a phase step that accounts for phase-step changes associated with variations in the effective wavelength irradiating the sample when the surface is curved. The nominal phase step is replaced in the filter with an actual phase step size that includes a phase parameter generated for each pixel by calibration or modeling of a specific type of surface. This substitution eliminates all errors produced by surface-dependent variations in the effective wavelength of the irradiating light.
1. A method for producing a height map of a sample with correction of errors produced by variations in an effective wavelength illuminating the sample, the method comprising the following steps:
producing a plurality of signals from a fringe pattern; and
calculating the map of the sample by application of a quadrature-demodulation technique on said signals;
wherein said quadrature-demodulation technique utilizes a filter with a phase parameter equal to an empirical value added to a nominal phase step used to produce said plurality of signals; and
wherein said empirical value is a measure of a phase error produced by a variation of said effective wavelength illuminating the sample.
2. The method of claim 1 , wherein said empirical value is obtained by a calibration process whereby a spectral amplitude of a correlogram produced at each pixel by said plurality of signals is calculated using a quadrature-demodulation method.
3. The method of claim 1 , wherein said empirical value is obtained with a parameterized model fitted empirically to said sample.
4. The method of claim 3 , wherein said sample has a surface corresponding substantially to a spherical section, and said parameterized model is represented by
δ xy =αr β ,
where r is a distance between a pixel x,y and a center pixel x 0 ,y 0 on said surface, and α and β are empirical parameters.
5. The method of claim 1 , wherein said correction of errors is carried out in real time during said calculating step.
6. The method of claim 1 , wherein said signals are interference signals.
7. A method for producing a height map of a sample with correction of errors produced by variations in an effective wavelength illuminating the sample, the method comprising the following steps:
producing a plurality of signals from a fringe pattern;
calculating a coarse map of the sample by application of a coherence-peak sensing technique on said signals;
calculating a fine map of the sample by application of a quadrature-demodulation technique on said signals; and
combining the coarse map and the fine map to produce a high-definition map of the sample;
wherein said quadrature-demodulation technique utilizes a filter with a phase parameter equal to an empirical value added to a nominal phase step used to produce said plurality of signals; and
wherein said empirical value is a measure of a phase error produced by a variation of said effective wavelength illuminating the sample.
8. The method of claim 7 , wherein said empirical value is obtained by a calibration process whereby a spectral amplitude of a correlogram produced at each pixel by said plurality of signals is calculated using a quadrature-demodulation method.
9. The method of claim 7 , wherein said empirical value is obtained with a parameterized model fitted empirically to said sample.
10. The method of claim 9 , wherein said sample has a surface corresponding substantially to a spherical section, and said parameterized model is represented by
δ xy =αr β ,
where r is a distance between a pixel x,y and a center pixel x 0 ,y 0 on said surface, and α and β are empirical parameters.
11. The method of claim 7 , wherein said correction of errors is carried out in real time during said calculating step.
12. The method of claim 7 , wherein said signals are interference signals.
13. Interferometric apparatus for producing a height map of a sample surface with correction of errors produced by variations in an effective wavelength illuminating the sample, the apparatus comprising:
means for producing a plurality of interference signals; and
means for calculating the map of the sample by application of a quadrature-demodulation technique on said interference signals;
wherein said quadrature-demodulation technique utilizes a filter with a phase parameter equal to an empirical value added to a nominal phase step used to produce said plurality of interference signals; and
wherein said empirical value is a measure of a phase error produced by a variation of said effective wavelength illuminating the sample.
14. The apparatus of claim 13 , wherein said empirical value is obtained by a calibration process whereby a spectral amplitude of a correlogram produced at each pixel by said plurality of interference signals is calculated using a quadrature-demodulation method.
15. The apparatus of claim 13 , wherein said empirical value is obtained with a parameterized model fitted empirically to said sample.
16. The apparatus of claim 15 , wherein said sample has a surface corresponding substantially to a spherical section, and said parameterized model is represented by
δ xy =αr β ,
where r is a distance between a pixel x,y and a center pixel x 0 ,y 0 on said surface, and α and β are empirical parameters.
17. The apparatus of claim 13 , wherein said correction of errors is carried out in real time during said calculating step.