Discussion and news about the modern effort to understand the nature of life on Earth, finding planets around other stars, and the search for life elsewhere in the universe

Monday, October 18, 2010

Through the halo

The technology for detecting planets continues to advance. A new paper by Sascha and colleagues contains the first direct imaging data of a planet obtained with an emerging optical technology called an apodized phase plate. Interestingly this builds on a set of ideas that have also been kicking around for a while in microscopy - the techniques for peering into the infinitesimal can also help us peer into the infinite.

The classic difficulty in looking directly at exoplanets is that they always appear extraordinarily close to their parent stars on the sky. Stars are millions to billions of times brighter than planets (depending on the precise wavelength of observation), and the wave nature of light always results in the blurring, or diffraction, of light as it passes through any kind of optical instrument. So the starlight ends up swamping out the feeble emission or reflection from any orbiting planets. Getting rid of the starlight is an entire field of research unto itself. Methods such as coronagraphy, and the rather ominous sounding nulling interferometry all seek to remove or suppress the light of the star, leaving behind the glimmer of any planets. It's been likened to trying to spot fireflies in the glare of a searchlight when you're thousands of miles away - that's not a bad analogy.

The apodized phase plate is a nifty bit of trickery. An earlier paper in 2007 by Kenworthy et al. describes the application to astronomy. In a nutshell the plate (combined with a coronagraph - a disk directly blocking some of the light of the star in the image) is a transparent chunk of material - in this case zinc selenide which is highly transparent in infrared wavelengths - that is modified to introduce a complex spatial pattern of phase changes to incoming light. Ok, so that's not so much a nutshell as a mouthful of walnut. Imagine a watery surface with light reflecting from it. You've undoubtedly seen this happen - wonderfully varied patterns of light and dark occur in the reflected light, the stuff of a summers day, shimmering on the hulls of boats, or off the dirty dishes in a kitchen sink. What you're observing are the phase shifts in the light - tiny time delays because a watery surface is not perfectly flat. If the phase shifts are out of sync light cancels out, if they're in sync it adds together - and we see dark and light regions in the reflected images.

The phase plate exploits the same physics, but in a careful, mathematically controlled way. Its 'rippled' surface adds time delays in pre-determined locations across the beam of light from a distant stellar system. These are cleverly designed so that on one side of the final image the light of the star gets blanked out (nulled), but the light of any planets remains. It's a tough challenge to get this to work. You also have to remove as much of the other sources of blurriness - like that induced by Earth's own atmosphere - as possible. Sascha et al. seem to have accomplished this, re-detecting a giant planet around the young star beta Pictoris, on an orbit of a mere 7 astronomical units (that's between Jupiter and Saturn if it were our solar system).

It's a terrific step. There are certainly many side effects of such trickery - a narrow waveband, imagery of only one side of the star, calibration challenges - but together with many other emerging techniques it does seem that we're well on our way to doing even richer exoplanetary science.

2 comments:

Eniac said...

Excellent post.

One nit to pick: Those beautiful shimmerings of summer days, those are caustics, not phase shift interference. Sunlight is not coherent enough to cause visible interference patterns, normally. One exception is oil films, which can also be very beautiful and exhibit actual interference.

Unknown said...

Yup, mea culpa, trying too hard for a real-world example that also satisfied my poetic conceit ! I guess another example would be the grainy/speckly pattern you see if you point a laser-pointer at a surface - that's interference due to scattering a coherent source (at least that's my guess).