Interferometers and spectrographs


One of the primary difficulties in determining the radial velocities of stars so that we can detect planets around them is measuring exactly what wavelength (or color) a particular spectral feature is at.  
When we put starlight from a telescope through a spectrograph (basically a fancy prism) the light is dispersed into all of the colors of the rainbow with very high resolution.  This means that we can distinguish between incredibly subtle shades of color, which is necessary because as stars wobble from the presence of planets their color varies by incredibly small amounts (less than one part in 1 hundred million!)  
The problem is that when we go back to the telescope to see if these wavelengths (colors) have shifted, we have to make sure that the spectrograph is still calibrated properly — we don’t want to mistake a shift in the instrument for a shift in the star’s light.  The two ways to do this are by keeping your spectrograph ultra-stable (as the European HARPS instrument does) or by having a calibration filter or cell in place that remove very specific colors from your spectrum, allowing you to calibrate any instrumental shifts (like the iodine cell technique I use).
I have now worked with two different teams that have tried to use an interferometer to solve the problem.  The idea is to create a physical device that removes very specific wavelengths of light through interference fringes.  The first project, called TEDI, did this in a very clever way that created spectral beat patterns of starlight against the interferometer light.  This allowed us to turn a low-resolution spectrograph into a high precision velocimeter (I know that’s a highly technical description — but it’s a very technical and fancy project, so I don’t know how else to describe it!)
The other team is led by our own Suvrath Mahadevan at the Center for Exoplanets and Habitable Worlds here at Penn State.  Dr. Mahadevan’s team, and especially his student Sam Halverson (who, totally coincidentally, also worked on TEDI as a Berkeley undergrad) have used a Fabry-P�rot interferometer to create interference fringes with the prototype for the Habitable Zone Planet Finder.  These fringes can be used to calibrate a spectrograph similar to the way that HARPS operates.  This is technique very similar to one that uses the new, Nobel Prize winning technology called frequency laser combs, except that these Fabry-P�rot devices are an order of magnitude cheaper and easier to build.
The first results are shown above.  The image is a close-up of several sections of the spectrum.  The color is green, but in reality these wavelengths are in the infrared portion of the spectrum.  Each dot is a bright interference peak from the interferometer, and each column contains dots of light of slowly but steadily increasing wavelength.  Think of this as a portion of a page that reads top to bottom, like a Chinese text: you “read” the dots from top to bottom (I’m not showing the whole column) and then when you get to the bottom start over at the top of the next column and continue.  There are thousands of these dots over hundreds of columns spanning almost a factor of 2 in wavelength.
It’s a beautiful and exciting result, and I hope that it turns into more reliable exoplanet detection everywhere.
Dr. Mahadevan describes the work he has done here.