Every object in the universe with mass has the potential to bend light a little; the harder part is being able to measure that change.
This behavior of light, taking apparently curved paths around very massive objects, was one of the earliest tests of General Relativity. General Relativity suggested that all light should travel in locally straight lines, but if space itself was being warped due to the presence of massive objects, then the direction a beam of light might find to be “straight” might not look so straight to an external observer.
An extreme case of the deflection of light exists as gravitational lensing. Gravitational lensing occurs when you have two objects lined up, one behind the other, and the light from the more distant object winds up curling around the closer, massive object, like water from a tap flowing around your fist. As a result, you can get multiple images of the same object; instead of a single jet of water hitting the bottom of the sink, the water splits, and hits the sink in a few places. The way that the light flows around the heavier object tells you about the exact alignment of the two objects, but it also tells you in great detail about the concentration of mass for the object in front. You can imagine that if you ran water over your fist, it would create a different landing pattern than if you ran it over a bowl, and different again if you ran water over a wine glass.
Typically we hear about gravitational lensing from studies of distant galaxies, where we’re able to observe an extremely distant object in great detail because gravitational lensing spread the light out over a wider area of the sky, or because the gravitational lens magnified the light from the object, making it brighter and easier to spot. But these galaxies aren’t the only thing we can observe using these methods; smaller scale objects work just as well.
In fact, this is one of the ways that you can find exoplanets. Say you’re looking at two stars which happen to be lined up; the light from the more distant one will bend around the mass of the star in front. If you have planets orbiting that star, the mass of the planet can bend the light a little further than you might otherwise expect. If the planet is massive enough, and all is very perfectly aligned, this additional distortion can wind up producing an extra image of the background star. Because the effect is relatively small, this method works best if the two stars are very closely aligned, which also means it works best if your planet is also very close in to its parent star.
The tricky part is measuring it! All the nearby stars are measurably in motion, and so the alignments between any two stars are fleeting. On top of that, you need extremely well calibrated data to be able to catch these small changes to a star’s light, as it passes behind another star. One such survey, the Optical Gravitational Lensing Experiment (OGLE) has been running for 25 years, and while its main goal isn’t finding exoplanets, it’s found a few through gravitational lensing measurements anyways. However, an upcoming NASA mission, the Wide Field Infra-Red Survey Telescope (WFIRST) has the detection of exoplanets by microlensing events as one of its main science goals, so when it launches in the 2020s, we should expect to be inundated with new exoplanets all over again - they're expecting to find at least 2,000 new planets.
Sign up for the mailing list for updates & news straight to your inbox!