High-performance 3D waveguide architecture for astronomical pupil-remapping interferometry

TL;DR

This paper introduces a high-performance 3D waveguide architecture for astronomical pupil-remapping interferometry, significantly enhancing direct imaging capabilities for exoplanet detection from ground-based telescopes.

Contribution

It presents a novel photonic device that improves contrast sensitivity and reduces interference, enabling more effective exoplanet imaging with ground-based telescopes.

Findings

Closure phase measurement scatter of ~0.2° achieved.

Waveguide throughput exceeds 70%.

Contrast-ratio sensitivity improved by an order of magnitude.

Abstract

The detection and characterisation of extra-solar planets is a major theme driving modern astronomy, with the vast majority of such measurements being achieved by Doppler radial-velocity and transit observations. Another technique -- direct imaging -- can access a parameter space that complements these methods, and paves the way for future technologies capable of detailed characterization of exoplanetary atmospheres and surfaces. However achieving the required levels of performance with direct imaging, particularly from ground-based telescopes which must contend with the Earth's turbulent atmosphere, requires considerable sophistication in the instrument and detection strategy. Here we demonstrate a new generation of photonic pupil-remapping devices which build upon the interferometric framework developed for the {\it Dragonfly} instrument: a high contrast waveguide-based device which recovers robust complex visibility observables. New generation Dragonfly devices overcome problems caused by interference from unguided light and low throughput, promising unprecedented on-sky performance. Closure phase measurement scatter of only $\sim 0.2^\circ$ has been achieved, with waveguide throughputs of $> 70\%$. This translates to a maximum contrast-ratio sensitivity (between the host star and its orbiting planet) at $1 \lambda/D$ (1$\sigma$ detection) of $5.3 \times 10^{-4}$ (when a conventional adaptive-optics (AO) system is used) or $1.8 \times 10^{-4}$ (for typical `extreme-AO' performance), improving even further when random error is minimised by averaging over multiple exposures. This is an order of magnitude beyond conventional pupil-segmenting interferometry techniques (such as aperture masking), allowing a previously inaccessible part of the star to planet contrast-separation parameter space to be explored.