부산대학교 도서관

Studying the formation of planet and protoplanetary disks around young stars requires angular resolutions of less than 1 milli-arcsecond which can only be achieved using interferometry. The Planet Formation Imager initiative expressed the need for kilometric baseline mid-infrared interferometry with tens of telescopes [1]. Current state-of-the-art $[8, 13] \mu \mathrm{m}$ interferometer at VLTI is able to combine four telescopes on up to $130\mathrm{m}$ baselines with a direct interferometry approach. Direct interferometry uses bulky mid-infrared free space delay lines to recombine the light from separated telescopes and recover the interferometric observable from the astronomical object. Heterodyne interferometry relies on a different approach which consists in detecting the heterodyne beating between the astronomical signal and a local oscillator at each telescope. The resulting heterodyne signals are radiofrequency signals that can be either digitized or analogically transmitted, amplified and processed before being correlated. Unlike direct schemes, heterodyne interferometry can be easily scaled up to $\mathrm{N}$ telescopes without any major additionnal noise contribution, making it competitive for studying planet formation despite its relatively lower sensitivity. The SNR of the heterodyne detection chain in the shot-noise limited regime scales as the square-root of its bandwidth $\sqrt{\Delta v}$, Thus, infrared detectors and correlators with large bandwidth bandwidth $(\geq 10$ GHz) represent key technologies for the development of sensitive heterodyne interferometry.