Mirror Surface Characterization with a Michelson Interferometer

Optics Rotation Projekt 1, Fall 2003
Tatjana Vavilkin
SUNY at Stony Brook
Adviser: Prof. Tom Weinacht


Introduction

The goal of this Optics Rotation project was to built a Michelson interferometer in order to map the surface features of a mirror.



Experimental Set-Up

This quite simple set up consists of four mirrors, a beam splitter, a telescope, combination of two polarization filters and a laser. It is a He-Ne laser with the wavelength 633nm and 1mW output power. After reflection on the beam splitter, a part of the beam goes through the telescope and is expanded in order to light up bigger area on the mirror. It is reflected back to the screen and formes interference fringes with the part passed through the beam splitter. The interference pattern on the screen are recorded with a CCD camera. The two polarization filters are needed to diminish the intensity of the beam in order to avoid saturation of the CCD camera. A neutral density filter can be used instead.




This are the fringe pattern seen on the screen.


In order to analyse the image there should be more fringes. Therefore the set-up was modified. The camera was placed instead of the screen, so the interference fringes occurred directly on the chip. The mirrors were also tilted a little. We observed Fizeau fringes which can be interpreted as contours of the surface under examination.




Analyses

The interference fringes were analysed with MATLAB and White Light Code by Florian Langhojer. Additional images of the background and both interfering beams apart were taken and subtracted from the interference image in order to improve it. Only one line of the image was taken. The White Light Code analysed the spacing between fringes. As result it gives the phase distribution along the mirror.


Now it is possible to calculate back to the features of the mirror surface. In this case the variation of the illuminated mirror surface is only 20nm.


Conclusion

Next step with this experimental set-up would be to replace a flat mirror with a deformable mirror and measure its surface. It is also necessary to develop software that can analyse the surface in 2D instead in 1D.


Acknowledgements

I thank Prof. Tom Weinacht, Florian Langhojer and David Cardoza for their help and advice.



vavilkin@grad.physics.sunysb.edu