| Imaging ellipsometry is an ellipsometry technique with a high lateral resolution. It can achieve ellipsometry on each pixel of optical imaging of the sample surface simultaneously, and obtain the physical parameters of materials,such as thickness, refractive index, extinction coefficient, surface roughness, component ratio in synthetic materials and so on, and their space distribution. It is a fast, non-invasive and high accuracy measurement technique, which make imaging ellipsometry play an important role in many fields, especially in thin film detection.Unlike traditional transmissive optical system design, in this paper, we develop a broadband spectroscopic imaging ellipsometer, which is free of chromatic aberration, by using an all-reflective focusing optical structure with special polarization control. Specially, we design to use two pairs of off-axis parabolic mirror and plane mirror in specific combinations in the optical path, which can effectively eliminate the change of polarization state caused by all-reflective focusing structure.In order to simplify the complex processes of common calibration method, we proposed a method of using multiple standard samples to calibrate the imaging ellipsometer. To put it simply, this calibration method can be described as:at first, light intensity images with different angles of the compensator were measured from multiple standard samples with known film thickness and optical properties. Then, the calibration parameters of the whole imaging ellipsometer system can be estimated using region average, Fourier analysis and a least square method, which contain the azimuth of polarizer/analyzer, the initial azimuth of compensator, the phase retardation of wave plate and so on.After the system calibration, we can determine the test sample’s spatial distributions of ellipsometric angles Ψ and Δ,and the film thickness by applying the obtained parameters.To measure the SiO2/Si thin films with the thicknesses of 20~3000 A at multiple wavelengths between 200 and 1000nm, we verify the accuracy of our home-made imaging ellipsometer, the result shows that the SiO2 film thickness can be determined within the maximum relative measurement error of 6% by our imaging ellispometer.Finally, we present a technique improving image resolution called super resolution, which can combine different low resolution images to reconstruct a high resolution image. By experiment, the result shows super resolution technology is practical and efficient in the imaging ellipsometer. And we made some future expectations about the application of this technique in imaging ellipsometer. |