Slashdot Mirror
Looking Directly at Extrasolar Planets
D2Deek writes "Science Daily is reporting on a new device called an Optical Vortex Coronagraph that's been invented to directly image planets orbiting other stars by using a special lens that "spins out" the light from the star leaving only the reflected light from the planet." I just can't imagine trying to clean a lens shaped like a giant corkscrew.
A new optical device might allow astronomers to view extrasolar planets directly without the annoying glare of the parent star. It would do this by "nulling" out the light of the parent star by exploiting its wave nature, leaving the reflected light from the nearby planet to be observed in space-based detectors.
About ten years ago, the presence of planets around stars other than our sun was first deduced by the very tiny wobble in the star's spectrum of light imposed by the mutual tug between the star and its satellite. Since then, more than 100 extrasolar planets have been detected in this way. Also, in a few cases the slight diminution in the star's radiation caused by the transit of the planet across in front of the star has been observed. Many astronomers would, however, like to view the planet directly, a difficult thing to do.
Seeing the planet next to its bright star has been compared to trying to discern, from a hundred meters away, the light of a match held up next to the glare of an automobile's headlight. The approach taken by Grover Swartzlander and his colleagues at the University of Arizona is to eliminate the star's light by sending it through a special helical-shaped mask, a sort of lens whose geometry resembles that of a spiral staircase turned on its side.
The process works in the following way: light passing through the thicker and central part of the mask is slowed down. Because of the graduated shape of the glass, an "optical vortex" is created: the light coming along the axis of the mask is, in effect, spun out of the image. It is nulled, as if an opaque mask had been placed across the image of the star, but leaving the light from the nearby planet unaffected.
The idea of an optical vortex has been around for many years, but it has never been applied to astronomy before. In lab trials of the optical vortex mask, light from mock stars has been reduced by factors of 100 to 1000, while light from a nearby "planet" was unaffected (see figure).
Attaching their device to a telescope on Mt. Lemon outside Tucson, Arizona, the researchers took pictures of Saturn and its nearby rings to demonstrate the ease of integrating the mask into telescopic imaging system. This is, according to Swartzlander (520-626-3723, grovers@optics.arizona.edu), a more practical technique than merely attempting to cover the star's image, as is done in coronagraphs, devices for observing our sun's corona by masking out the disk of the sun. It could fully come into its own on a project like the Terrestrial Planet Finder, or TPF, a proposed orbiting telescope to be developed over the coming decade and designed to image exoplanets.
Foo et al., Optics Letters, 15 December 2005 Summary of articles related to optical vortex on Swartzlander's Web page
Someone more patient than I can put in the links to the figures. See http://aip.org/pnu/2005/755.html for everything.
Apparently all the work so far is only computer simulations. There are some serious problems to overcome before this could be a practical system. The author states:
"These calculations assume no aberrations or other scattering sources, and they assume the vortex mask can be made achromatic."
http://www.u.arizona.edu/~grovers/ovc.html
In other words, the lens material is made from unobtanium and the rest of the system has to be perfect. The author certainly knows this will never happen.
Mike
The 472,068,000 doesn't include the 300,000km/s. The real number would be 142,009,200,000,000. (also a year is 365.25 days, but the 15 light-year measurement is much less accurate than that I suspect) The page you point to says the planet's average orbit is 0.2 times the earth-sun distance, about 31.5 million km. That gives an angular distance between the star and planet of about 0.0000127 degrees, or 0.045 arcseconds. The hubble can resolve about 0.07 arcseconds, if we can separate the glare from the star bleeding over, then we are close.
This is why you're taught in courses on physics how to estimate things like the weight of the pyramids or the capacitance of clouds (well we were). So you aren't just a slave to your calculator and can actually detect when a number is BS.