The conjugate differential method has been applied to the absolute test of flat, cylindrical, and axicon surfaces. In the
previous work, simulations and correspond experiments have been carried out to verify the feasibility of the method. To
analyze the influences of different factors upon the measurement result, the conjugate differential method is discussed in
detail. Considering the characteristics of the test surface such as surface types and surface profiles, the application ranges
of the conjugate differential method are discussed into three parts. According to the three surface types using the
conjugate differential method, the method can be extended to the absolute test of the spherical surfaces based on
spherical coordinate system. The reconstructed errors caused by different aberrations expressed as Zernike polynomial
terms show that they are more sensitive to high order aberration terms of the surface under test. And for surfaces with
different frequency distributions, the surface with less mid-spatial frequency information is less sensitive to the sampling
frequency. The influence from the other factors in interferometric test are also discussed into three parts. The influences
from the uncertainty of shifts are correlated with the increased aperture diameters, since the integration error caused by
the shift error increases gradually with the expanding of the integration path. The integration error changes by the
influences from the coherent noise and pixel noise related to pixel deviations. The reconstructed deviations get increased
while the peak pixel deviation is increasing. For the balance of the differential deviation and integration error, the
optimization of sampling resolution should take considered for accuracy improvement.
Multilayer defects which reside on the top or inside the multilayer are one of the most critical concerns in the extreme ultraviolet lithography (EUVL) manufacturing process. We proposed the transport of intensity equation and partial least-square regression (TIE & PLSR) method to inspect the defect and reconstruct its geometric parameters: height and full width at half maximum (FWHM). The transport of intensity equation (TIE) is employed to retrieve the phase of the multilayer defect from the two scattering images, which collected at two adjacent propagation distances. Comparing the simulated ideal phase, the phase deformations caused by different top heights and widths of the defects are analyzed. The optical properties maximum, minimum and fitting Zernike coefficients are used to parameterize the phase deformation. Partial least-squares regression (PLSR) is applied to associate the optical properties of the phase deformation with the geometric parameters of the defects, and reconstruct geometric parameters of the measured defect from the established data library. The reconstruction error is less than 0.2% in simulation experiment.
An absolute testing method for cylindrical surfaces is presented in a null test setup with a computer-generated hologram. The absolute test exploits the symmetry of cylinders, which allows us to introduce a certain shift of the test surface both parallel to and rotated about the centerline while the null test condition is still maintained. With two shifts of the cylindrical surface, four measurements belonging to two groups in conjugate positions can be accomplished to obtain the absolute differential map with the interferometer and null optics errors removed. The absolute surface can be obtained by wavefront reconstruction from local differential data. A simulation of the method is presented to estimate the error propagation. Experimental absolute test results of a concave cylindrical surface with 100-mm radius are given. The measured profiles are compared with those obtained from a commercial profiler, showing a difference of less than 15 nm (root-mean-square).
Optical freeform surfaces are complex surfaces with non-rotational symmetry that break through the limitations of conventional optical element, and are widely used in advanced optics application for system configuration simplifying and performance enhancing. Due to the geometrical complexity and optical particularity of optical freeform surfaces, there is, as yet, a lack of precision freeform surfaces testing. Computer generated hologram (CGH) null testing method are discussed in this paper to test the optical freeform surfaces such as off-axis aspheric surfaces. CGH design based on ray tracing and NURBS interpolation are included. Simuation in Zemax is given to verify the result of calculation. The alignment and fiducial sections are added to the CGH to lead the alignment of the freeform surface and CGH with sixdimensional adjustment. The CGH was designed and fabricated to test an off-axis aspheric with Fizeau configuration.