Context. Cold planets, including all habitable planets, produce only scattered light emission in the visual to near-infrared wavelength range. For this reason it is highly desirable to adapt the technique for the direct imaging of reflected light from extra-solar planets. Aims. For the nearby system ε Eri, we want to set much deeper detection limits for the expected scattered radiation from the radial velocity planet candidate (≈0.7 MJ) and the warm dust using the VLT/SPHERE adaptive optics (AO) instrument with the ZIMPOL imaging polarimeter. Methods. We carried out very deep imaging polarimetry of ε Eri based on 38.5 h of integration time with a broad-band filter (λc = 735 nm) for the search of the polarization signal from a planet or from circumstellar dust using AO, coronagraphy, high precision differential polarimetry, and angular differential imaging. The data were collected during 12 nights within four epochs distributed over 14 months and we searched for a signal in the individual epochs. We also combined the full data set to achieve an even higher contrast limit considering the Keplerian motion using the K-Stacker software. All data were also combined for the search of the scattering signal from extended dust clouds. We improved various data reduction and post-processing procedures and also developed new ones to enhance the sensitivity of SPHERE/ZIMPOL further. The final detection limits were quantified and we investigated the potential of SPHERE/ZIMPOL for deeper observations. Results. The data of ε Eridani provide unprecedented contrast limits but no significant detection of a point source or an extended signal from circumstellar dust. For each observing epoch, we achieved a 5 σ𝒩 point source contrast for the polarized intensity CP = Qϕ/I★ between 2 × 10−8 and 4 × 10−8 at a separation of ρ ≈ 1″, which is as expected for the proposed radial velocity planet at a quadrature phase. The polarimetric contrast limits are close to the photon noise limits for ρ > 0.6″ or about six times to 50 times better than the intensity limits because polarimetric imaging is much more efficient for speckle suppression. Combining the data for the search of a planet moving on a Keplerian orbit with the K-Stacker technique improves the contrast limits further by about a factor of two, when compared to an epoch, to about CP = 0.8 × 10−8 at ρ = 1″. This would allow the detection of a planet with a radius of about 2.5 RJ. Should future astrometry provide strong constraints on the position of the planet, then a 3 σ𝒩 detection at 1″ with CP ≈ 5 × 10−9 would be within reach of our data. The surface brightness contrast limits achieved for the polarized intensity from an extended scattering region is about 15 mag arcsec−2 at 1″ or up to 3 mag arcsec−2 deeper than previous limits. For ε Eri, these limits exclude the presence of a narrow dust ring and they constrain the dust properties. The photon statistics would allow deeper limits but we find a very weak systematic noise pattern probably introduced by polarimetric calibration errors. Conclusions. This ε Eri study shows that the polarimetric contrast limits for reflecting planets with SPHERE/ZIMPOL can be improved to a level below Cp < 10−8 by just collecting more data during many nights using software such as K-Stacker, which can combine all data considering the expected planet orbit. Contrast limits of Cp ≈ 10−9 are within reach for ε Eri if the search can be optimized for a planet with a well-known orbit. This limit is also attainable for other bright nearby stars, such as α Cen or Sirius A. Such data also provide unprecedented sensitivity for the search of extended polarized emission from warm circumstellar dust.
Read full abstract