The Phase Induced Amplitude Apodizer and Nuller -- High transmission, high dispersion coronagraphy at 2λ/D

TL;DR

This work presents the Phase Induced Amplitude Apodizer and Nuller (PIAAN), a high-transmission, small-inner-working-angle coronagraphic IFU designed to enable High Dispersion Coronagraphy for Proxima Cen b. By combining partial PIAA apodization with a seven-fiber single-mode array, the approach nulls starlight without sacrificing companion coupling, achieving theoretical contrasts near $7 \cdot 10^{-7}$ and transmission around $72\%$ at $2\lambda/D$ over a $30\%$ bandwidth, with lab demonstrations reaching $\sim3 \cdot 10^{-5}$ contrast across the band. A rigorous tolerance analysis (including tip-tilt, coma, manufacturing, and XAO residuals) and a wavefront-control strategy using eigenmodes enable robust performance, addressing the stringent requirements of sky operation. Prototyping with a CaF$_2$ rod-based PIAA and a 7-fiber IFU validates the concept, revealing strong stability and a clear path toward deployment on the VLT for the RISTRETTO instrument, albeit with performance ultimately limited by XAO capabilities. This work thus establishes PIAAN as a leading technology candidate for high-dispersion exoplanet characterization at small angular separations.

Abstract

Context. Proxima Cen b is the prime target for the search of life around a nearby exoplanet by characterizing its atmosphere in reflected light. Due to the very high star/companion contrast (<1E-6), High Dispersion Coronagraphy is the most promising technique to perform such a characterization. Aims. With a maximum separation of 37 mas, Proxima b can be observed with a VLT in the visible. It requires a coronagraph providing high contrast (< 1E-4 ) very close from the star (< 2 λ/D ) over a broad spectral range (~30%), with a high transmission of the companion (> 50%). We look for an optimal solution that takes benefit of the properties of single-mode fibers. Methods. We introduce the Phase Induced Amplitude Apodizer and Nuller (PIAAN), a coronagraphic integral field unit, designed to feed a diffraction limited spectrograph. It uses a pupil remapping optics with moderate apodization, combined to a single mode fiber integral field unit. It exploits the properties of single mode fibers to null the star light without reducing the companion coupling. The study focuses on a proper tolerance analysis and proposes a wavefront optimization strategy. A prototype is built to demonstrate its performance. Results. We show that the PIAAN can theoretically provide contrasts of 7E-7 and a transmission of 72% at 2 λ/D over a bandwidth of 30%. A prototype is built and characterized and the proposed wavefront control strategy is also demonstrated in the lab. We reach contrast levels of 3E-5 over the full bandwidth, as expected from the tolerance analysis. Conclusions. We demonstrated the potential of the new PIAAN coronagraph, from simulations to prototype. Its performance will eventually be limited by the XAO capabilities. It is the main coronagraph candidate for the RISTRETTO instrument to observe Proxima Cen b on the VLT and its first technology milestone.

Figures (20)

• Figure 1: Expected SNR for a Proxima Cen b observation on a VLT as a function of contrast for different coronagraph transmissions at 2$\lambda/D$. We assume a total optical transmission of 5% (excluding the coronagraph transmission), a Strehl ratio of 70%, and $\sigma_{det}=6e^-$ for a reduced pixel in 1h of exposure time lovis_2016a.
• Figure 2: Working principle of the PIAA Nuller. Top: coupling of the star (on-axis). Bottom: coupling of the companion at 2$\:\lambda/D\:$. Left: PIAA PSF and overlayed IFU at 620nm, with the 'companion' lenslet showcased. Right: Horizontal cut view of the PSF in log scale, with yellow part showing the electric field coupled to the fiber at the companion position.
• Figure 3: Schematics of the PIAAN simulation. The Front-End part is used to inject residual XAO phase screen and to simulate an optical relay with realistic optics and off-pupil aberrations. The MLA can be simulated with data from manufacturers if required.
• Figure 4: Expected performance for a PIAAN optimizing detection of a companion at 2$\:\lambda/D\:$ with a 30% bandwidth, i.e. in the RISTRETTO case between 620 and 840 nm. Behavior outside the optimization band is shown in the light gray areas. Left: star null $\rho_\odot$, showing the influence of the residual secondary obstruction. Right: companion coupling, mostly independent of the residual obstruction.
• Figure 5: Influence of the PIAA apodization on the PIAAN performance. Coupling are averaged for 30% bandwidth. Optimal solution is shown with the dashed line.