The VLT/ERIS grating vector Apodizing Phase Plate coronagraph

Abstract

We describe the design, laboratory manufacture, and on-sky testing of the grating vector apodizing phase plate (gvAPP) coronagraph for the Enhanced Resolution Imager and Spectrograph (ERIS) on the Very Large Telescope. We used both laboratory measurements and on-sky observations to characterise the gvAPP in several different filters, from the K to the L band. In testing, the gvAPP reaches its design specification in the transmission of the optic with 90% in the K bands and 60% in the L band. While the gvAPP reaches its designed raw contrast performance of $1 \times 10^{-5}$, it does not reach the post-processed contrast of $5 \times 10^{-5}$ in on-sky observations. Electronic detector noise, due to the Airy core of the coronagraphic point spread function inducing cross-talk between the readout amplifiers, produces a repeated pattern within the coronagraphic regions of the gvAPP. Despite these limitations, we recommend the gvAPP as a tool for characterising substellar companions with known separations and position angles, which allow them to be placed in the coronagraphic dark holes for observations. The ERIS gvAPP's leakage term can also be used as a photometric reference for time series observations; however, we caution that the contrast performance may limit such studies to only the brightest targets. ERIS gvAPP data quality may be improved further with better modelling of detector electronic noise. This work is a pathfinder for Extremely Large Telescope instruments including METIS, which will include gvAPP coronagraphs with improved designs based on these results.

Figures (6)

• Figure 1: eris gvapp phase pattern design and simulated psf. Left: Phase within the defined gvapp pupil. Right: Resultant three psf of the gvapp. We show the two coronagraphic psf and the central leakage term. Three astrometric reference spots can be seen. The scale is logarithmic in normalised intensity.
• Figure 2: Optical transmission of the gvapp measured in the laboratory (black curve) and the on-sky background transmission in the Br-$\alpha$ filter (orange point). Additional measurements with the eris internal calibration unit are shown by the blue points. The blue regions trace all filters available for observations with the gvapp in eris. Note that for on-sky throughput, this transmission curve should be multiplied by 0.23 to account for the EE and division of the flux between coronagraphic psf (see Sect. \ref{['sect:eris_gvapp_throughput']}).
• Figure 3: Geometry of combining the two coronagraphic psf to form the coverage around the target star. The hatched area is masked out during subsequent image combination and processing. The semi-circular dashed lines are a visual aid only; the area at wider separations beyond these circles is not masked in the combined image.
• Figure 4: Raw contrast of the gvapp in the Br-$\alpha$-continuum filter. Left: Raw flux contrast evaluated in $\lambda/D$-sized apertures placed in the dark hole over an aperture placed on the psf centre in the left lobe of a deep integration on HR 8799 at $3.96\, \mathrm{\mu m}$. The orange line traces the median contrast evaluated in a $\pm 10^\circ$ region perpendicular to the main axis of the gvapp psf. The orange markers indicate the iwa and owa provided in Table \ref{['tab:design_param']}. The blue curve displays the raw contrast measured on the gvapp element under laboratory conditions at $2 \, \mathrm{\mu m}$ by 2021JATIS...7d5001B -- the values from 10 to 12.5 $\lambda/D$ are below the sensitivity of the test bench. The rise seen at 13.5 $\lambda/D$ in the laboratory data is due to limitations from internal reflections in the test bench not present in ERIS. Right: Focal plane image of the left lobe of the gvapp on which the raw contrast was evaluated. The shaded orange wedge is the $\pm 10^\circ$ region in which the orange contrast curve was measured.
• Figure 5: Post-processed contrast of the eris gvapp in the Br-$\gamma$, Br-$\alpha$, Br-$\alpha$-continuum, and $K$-peak filters. The contrast curves for the first three filters were calculated after psf subtraction with optimal pca/adi via fake planet injections. The contrast curve for the $K$-peak filter was calculated with the TRAP algorithm Samland2021TRAP:Separations. The angular separation is shown in units of $\lambda/D$ in order to compare the results for the different narrow band filters. The central wavelength and effective width of each filter are indicated in the label of each curve. The shaded area represents the uncertainty of the contrast curve.