Space Optical Interferometry for High-Resolution Coherent Imaging of Astronomical Objects

Abstract

The hunt for Earth-like exoplanets is one of the great scientific endeavors of the 21st century. To date, the characterization of known exoplanets with instruments like JWST has been spatially unresolved—we can study atmospheric constituents via spectroscopy, but we cannot see continents, synoptic weather, or the geographic distribution of potential spectral biosignatures. The diffraction limit dictates that an optical aperture with sufficient angular resolution to resolve planetary scale features from the vantage point of our solar system would be prohibitively immense—hundreds to thousands of kilometers in diameter at visible wavelengths. Optical interferometry offers perhaps the only path to the required angular resolution by interfering light from multiple telescopes, or sub-apertures, providing Fourier components at a resolution corresponding to the sub-aperture separation. In recent decades, ground-based interferometry has made major strides in sensitivity by calibrating for atmosphere-induced piston errors to enable long coherent integration times. In addition to high-resolution astrometry, these sensitivity gains have allowed for milliarcsecond-scale imaging of bright astronomical objects. Still, optical interferometry from the Earth’s surface faces many fundamental performance limitations, and a space-based system would allow for the study of much dimmer targets at higher diffraction-limited resolutions, taking us a step closer to one day mapping exo-Earths. With recent advances in satellite miniaturization and lower cost-to-orbit, multiple groups have proposed new designs for a first demonstration mission, but no astronomical interferometer has yet flown in space. This work investigates the expected performance of first- and second-generation space optical interferometer concepts for astronomical imaging, focusing on maturing the technology that will be needed to push sensitivity and resolution beyond what is currently possible from the ground. The first mission envisioned is a formation flying pathfinder comprising three CubeSats, aiming to demonstrate the first measurements of interference fringes from starlight collected by separated space telescopes. This feat will require sub-wavelength matching of the optical path lengths traveled by the starlight to maintain the mutual coherence, which is achieved using rapid measurements of the interference fringe itself as a source of path length feedback. The performance of this fringe tracking is modeled via a time-domain control simulation accounting for the micro-vibration disturbance environment that would be expected on a CubeSat platform using reaction wheels for attitude control, which could induce up to 5 µm in optical path length noise and several arcseconds in optical alignment errors if left uncorrected. Also considered are the optical losses and noise sources associated with photon arrival statistics and detection as well as the beam pointing jitter. Simulation results indicate that such a mission would be able to stabilize interference fringes on at least the fifty brightest stars to better than 45 nm even under pessimistic disturbance assumptions, which would be sufficient to demonstrate the feasibility of space interferometry for a subsequent larger mission. The second mission concept analyzed aims to push the limits of faint-object interferometry from space by implementing a dual-feed beam combiner for fringe stabilization and phase referencing using bright off-axis guide stars. A three-telescope interferometer is assessed in its ability to map the surfaces of recently discovered dwarf planets in the outer solar system from a sun-synchronous Earth orbit. The population of stars usable as an interferometric phase reference is found to be within the off-axis field of view permitted by a 1-meter differential optical delay line, aiding image reconstruction via use of the Fourier phase referenced to a fixed point in the sky. Bayesian statistical imaging algorithms are employed to demonstrate recovery of an image from simulated noisy measurements of an 18th magnitude rotating object 37 millarcseconds in diameter, but results indicate that to do this with the integration times permitted by the chosen orbital formation would require approximately Hubble-sized telescope apertures. Potential alternative operational strategies to enable longer integration times with more modest apertures are discussed. Informed by the simulation results, recommendations are presented for near-term technology development in support of space interferometry.