A technique that corrects for large spherical aberrations in fast optics may make possible new applications in optical inspection, lithography, and even communications. Using cheap, holographically corrected lenses and mirrors, researchers at the Laser and Optics Research Center at the U.S. Air Force Academy in Colorado Springs, CO, have been able to image objects with sub micron resolution and micromachine 1-µm diameter holes in foil held under vacuum. The latter was only possible because of the large working distances achieved using their scheme. The work is an extension of research done at the University of Adelaide in Australia, which concentrated on correcting aberrations in telescopes.1 This work is now being adapted for use in optical receivers.
Making microscope objectives with high resolution generally requires a trade-off between cost and working distances. Single, fast lenses have a large amount of spherical aberration. One can compensate for this by making the aperture very small, but that requires the objective to be very close to the object to be imaged. Multi-element systems are often used instead. These have a much better working distance, but they can also be very expensive.
One solution to this problem, developed by Geoff Andersen and his colleagues in the U.S.2,3 is shown in Figure 1. A spatial filter is used as part of the object of a hologram, with a collimated beam used as the reference. The spatial filter acts as a point source of light that picks up aberrations as it travels through the microscope objective (which, in this case, is a Fresnel lens). After recording and replaying the hologram with the object beam -- the spatially filtered light traveling through the objective -- the perfect, collimated reference beam is reconstructed. Thus, the spherical aberrations caused by the optics are removed. When the spatial filter is then replaced by an object, the point at the same location as the pinhole will be correctly imaged, and its neighbors will be imaged with relatively little error (see Figure 2).
The system has some drawbacks. Its working distance is an order of magnitude larger than some microscopes, which means it can be used to image objects that have to be contained in, for example, a vacuum. However, its field of view can be very small -- just a few microns across -- because the image is ideally corrected only at the center point and aberrations get worse towards the edges. Also, because it uses a hologram, the system is quite wavelength dependent: the more aberration corrected for, the smaller the bandwidth of light that can be used. Finally, as with any diffractive system, some of the light is not correctly shaped by the hologram and has to somehow be filtered out of the system.
Figure 2. Examples of images from the holographically-corrected microscope. (a) Blood cells, approximately 5-µm in diameter. (b) Microchip with 0.7-µm track widths.
One application that can tolerate these problems is lithography, where the hologram is reconstructed in phase-conjugate with a patterned "reference" beam. The result is a real image in space that can be used to pattern a substrate in the "object" plane. The advantage of lithography is that it is generally performed with narrow bandwidths anyway, and the field of view, although small, can easily be scanned to pattern a large area. Particularly useful is the fact that the holographic correction technique can be used with reflective imaging optics (mirrors) not just transmissive optics. In fact, much of the earlier telescope work corrected reflective systems. In lithography, the adoption of mirror-based systems means that very short wavelength radiation (ultraviolet and beyond), normally absorbed by transmissive optics, can be used. This means smaller spot and feature sizes.
Another narrow-bandwidth application under investigation is the use of a holographically corrected telescope as part of an optical data transmitter and receiver.4 Andersen has demonstrated that even with a very poor-quality primary mirror with more than 2000 waves of aberration, an inexpensive telescope can operate at the diffraction limit. This would allow the system to handle data transmission rates up to 100 GHz.
1. Geoff Andersen, Jesper Munch, and Peter Veitch, Compact, holographic correction of aberrated telescopes, Applied Optics 36, pp. 1427-1432, 1997.
2. Geoff Andersen and R. J. Knize, Holographically corrected microscope with a large working distance, Applied Optics 37 (10), 1 April 1998.
3. G. Andersen and R. J. Knize, A high resolution, holographically corrected microscope with a Fresnel lens objective at large working distances, Optics Express 2 (13), 22 June 1998.
4. Geoff Andersen and R. J. Knize, Holographically corrected telescope for high bandwidth optical communications, submitted to Applied Optics.