学术文献库

Context. Direct imaging of exoplanets takes advantage of state-of-the-art adaptive optics (AO) systems, coronagraphy, and post-processing techniques. Coronagraphs attenuate starlight to mitigate the unfavorable flux ratio between an exoplanet and its host star. AO systems provide diffraction-limited images of point sources and minimize optical aberrations that would cause starlight to leak through coronagraphs. Post-processing techniques then estimate and remove residual stellar speckles such as noncommon path aberrations (NCPAs) and diffraction from telescope obscurations. Aims. We aim to demonstrate an efficient method to minimize the speckle intensity due to NCPAs during an observing night on VLT/SPHERE. Methods. We implement an iterative dark-hole (DH) algorithm to remove stellar speckles on-sky before a science observation. It uses a pair-wise probing estimator and a controller based on electric field conjugation. This work presents the first such on-sky minimization of speckles with a DH technique on SPHERE. Results. We show the standard deviation of the normalized intensity in the raw images is reduced by a factor of up to 5 in the corrected region with respect to the current calibration strategy under median conditions for VLT. This level of contrast performance obtained with only 1 min of exposure time reaches median performances on SPHERE that use post-processing methods requiring 1h-long sequences of observations. We also present an alternative calibration method that takes advantage of the starlight coherence and improves the post-processed contrast levels rms by a factor of about 3. Conclusions. This on-sky demonstration represents a decisive milestone for the future design, development, and observing strategy of the next generation of ground-based exoplanet imagers for 10m to 40m telescope.