SPHERE dynamical and spectroscopic characterization of HD142527B
R. Claudi, A.-L. Maire, D. Mesa, A. Cheetham, C. Fontanive, R. Gratton, A. Zurlo, H. Avenhaus, T. Bhowmik, B. Biller, A. Boccaletti, M. Bonavita, M. Bonnefoy, E. Cascone, G. Chauvin, A. Delboulbè, S. Desidera, V. D'Orazi, P. Feautrier, M. Feldt, F. Flammini Dotti, J.H. Girard, E. Giro, M. Janson, J. Hagelberg, M. Keppler, T. Kopytova, S. Lacour, A.-M. Lagrange, M. Langlois, J. Lannier, H. Le Coroller, F. Menard, S. Messina, M. Meyer, M. Millward, J. Olofsson, A. Pavlov, S. Peretti, C. Perrot, C. Pinte, J. Pragt, J. Ramos, S. Rochat, L. Rodet, R. Roelfsema, D. Rouan, G. Salter, T. Schmidt, E. Sissa, P. Thebault, S. Udry, A. Vigan
AAstronomy & Astrophysics manuscript no. HD142527_V8.1_arxiv c (cid:13)
ESO 2018December 20, 2018
SPHERE dynamical and spectroscopic characterization ofHD142527B (cid:63)
R. Claudi , A.–L. Maire , D. Mesa , , A. Cheetham , C. Fontanive , R. Gratton , A. Zurlo , , , H. Avenhaus , , , T.Bhowmik , B. Biller , , A. Boccaletti , M. Bonavita , , M. Bonnefoy , E. Cascone , G. Chauvin , , A.Delboulbé , S. Desidera , V. D’Orazi , P. Feautrier , M. Feldt , F. Flammini Dotti , , J.H. Girard , , E. Giro , M.Janson , , J. Hagelberg , M. Keppler , T. Kopytova , , S. Lacour , A.–M. Lagrange , M. Langlois , , J.Lannier , H. Le Coroller , F. Menard , S. Messina , M. Meyer , , M. Millward , J. Olofsson , , , A. Pavlov , S.Peretti , C. Perrot , C. Pinte , , J. Pragt , J. Ramos , S. Rochat , L. Rodet , R. Roelfsema , D. Rouan , G.Salter , T. Schmidt , E. Sissa , P. Thebault , S. Udry , A. Vigan . (A ffi liations can be found after the references) Received ....; accepted .....
ABSTRACT
Aims.
HD142527 is one of the most frequently studied Herbig Ae / Be stars with a transitional disk that hosts a large cavity that is upto about 100 au in radius. For this reason, it has been included in the guaranteed time observation (GTO) SpHere INfrared survey forExoplanets (SHINE) as part of the Spectro-Polarimetric High-contrast Exoplanet REsearch (SPHERE) at the Very Large Telescope(VLT) in order to search for low-mass companions that might explain the presence of the gap. SHINE is a large survey within about600 young nearby stars are observed with SPHERE with the aim to constrain the occurrence and orbital properties of the giant planetpopulation at large ( > Methods.
We used the IRDIFS observing mode of SPHERE (IRDIS short for infrared dual imaging and spectrograph plus IFS orintegral field spectrograph) without any coronagraph in order to search for and characterize companions as close as 30 mas of thestar. Furthermore, we present the first observations that ever used the sparse aperture mask (SAM) for SPHERE both in IRDIFS andIRDIFS_EXT modes. All the data were reduced using the dedicated SPHERE pipeline and dedicated algorithms that make use of theprincipal component analysis (PCA) and reference di ff erential imaging (RDI) techniques. Results.
We detect the accreting low–mass companion HD142527B at a separation of 73 mas (11.4 au) from the star. No othercompanions with mass greater than 10 M J are visible in the field of view of IFS ( ∼
100 au centered on the star) or in the IRDISfield of view ( ∼
400 au centered on the star). Measurements from IFS, SAM IFS, and IRDIS suggest an M6 spectral type forHD142527B, with an uncertainty of one spectral subtype, compatible with an object of M = . ± .
06 M (cid:12) and R = . ± .
07 R (cid:12) .The determination of the mass remains a challenge using contemporary evolutionary models, as they do not account for the energyinput due to accretion from infalling material. We consider that the spectral type of the secondary may also be earlier than the typewe derived from IFS spectra. From dynamical considerations, we further constrain the mass to 0 . + . − . M (cid:12) , which is consistentwith both our spectroscopic analysis and the values reported in the literature. Following previous methods, the lower and upperdynamical mass values correspond to a spectral type between M2.5 and M5.5 for the companion. By fitting the astrometric points,we find the following orbital parameters: a period of P = −
137 yr; an inclination of i = − ◦ , a value of Ω = − ◦ for the longitude of node, and an 68% confidence interval of ∼ −
57 au for the separation at periapsis. Eccentricity and time atperiapsis passage exhibit two groups of values: ∼ ∼ e , and ∼ ∼ T . While theseorbital parameters might at first suggest that HD142527B is not the companion responsible for the outer disk truncation, a previoushydrodynamical analysis of this system showed that they are compatible with a companion that is able to produce the large cavity andother observed features. Key words.
Star: Formation, Protoplanetary Disks, Instrumentation: high angular resolution, Techniques: imaging spectroscopy,Stars: Individual: HD142527
1. Introduction
Planet formation from disks (Williams & Cieza 2011; Mordasiniet al. 2015) imprints characteristic structures on the disks (e.g.,Muto et al. 2012). One of the most striking structures that can beproduced in this manner are the wide gaps carved by massiveor multiple forming planets in protoplanetary disks (Dodson-Robinson & Salyk 2011; Zhu et al. 2011). Disks with large gaps, (cid:63)
Based on observations collected at the European Organisationfor astronomical research in the southern emisphere under ESO pro-grammes 095.C–0298, 096.C–0241, 097.C–0865 and 189.C–0209. or cavities, are often referred to as “transitional disks”, and theyare observed around Herbig stars or their less massive counter-parts, TTauri stars (Strom et al. 1989). A striking example of atransitional disk is the young Herbig Ae star (F6III, age ∼ ∼
100 au in radius. For this reason, itis one of the most freequently studied objects of this type (seeSection 2). Transitional disks are believed to be in the evolution-ary stage between optically thick gas-rich disks and older diskswhere most of the gas has been dissipated (see, e.g., Espaillatet al. (2014) and references therein). In a handful of cases the
Article number, page 1 of 15 a r X i v : . [ a s t r o - ph . S R ] D ec & A proofs: manuscript no. HD142527_V8.1_arxiv connection between the presence of a gap and the existence of(candidate) low-mass companions has been established. Somecompanion candidates were discovered in the LkCa 15 system(Kraus & Ireland 2012; Sallum et al. 2015), MWC 758 (Reg-giani et al. 2017), HD169142 (Reggiani et al. 2014; Biller et al.2014; Osorio et al. 2014; Fedele et al. 2017; Ligi et al. 2018),and HD100546 (Mulders et al. 2013; Quanz et al. 2013, 2015;Brittain et al. 2013, 2014; Quanz et al. 2015). HD142527 alsohas a low-mass companion that resembles an M star (Lacouret al. 2016). The recent advent of dedicated high-contrast im-agers such as Spectro-Polarimetric High-contrast Exoplanet RE-search (SPHERE, Beuzit et al. 2008), Gemini Planet Imager(GPI, Macintosh et al. 2014), and Subaru Coronagraphic Ex-treme Adaptive Optics (SCExAO, Jovanovic et al. 2016) o ff ersthe possibility to probe regions of transitional disk systems thatlie closer to the star than was possible with previous imaging in-struments and to constrain the presence of any companion thatcould be responsible for opening up an observed gap or cavity.In this paper we present new deep images of the central re-gions of the HD142527 system obtained with SPHERE as part ofthe SPHERE consortium guaranteed time observations (GTO).These data include both non-coronagraphic direct images andsparse aperture masking (SAM) data acquired with the near-infrared channels integral field spectrograph (IFS) and infra-reddual imaging and spectrograph (IRDIS). The outline of the paperis as follows: in Sect. 2 we summarize the main characteristicsof the HD142527 system. In Sect. 3 we describe the SPHEREnear-infrared observations; in Sect. 4 we describe the reductionmethods we applied, and the results are discussed in Sect. 5. Themass estimate of the companion is discussed in Sect. 6, and theorbital properties of HD142527B are derived in Sect. 7. In Sect.8 we outline the conclusions.
2. HD142527
HD142527 is a young 5 ± . ±
20 pc from theSun (Acke & van den Ancker 2004; Mendigutía et al. 2014).Its spectral type of F6 IIIe (Houk 1978; Henize 1976; Waelkenset al. 1996) corresponds to a mass of 2 . ± . (cid:12) , (Fukagawaet al. 2006; Verhoe ff et al. 2011). We adopt the distance recentlyrefined by Gaia Collaboration et al. (2016) to 156 + − pc and theproper motion of µ α = − . ± .
08 mas / yr, µ δ = − . ± . / yr.HD142527 is one of the most frequently studied HerbigAe / Be stars with a transitional disk because its disk is seen al-most face-on and its protoplanetary cavity, extending to a radiusof about 100 au, can be investigated because of its record size(Fukagawa et al. 2006). The gap in this system extends to be-tween 30 and 130 au (Verhoe ff et al. 2011), and the outer diskextends to about 600 au. The system is seen at a low inclination: i = ± ◦ (Perez et al. 2015). The HD142527 disk has longbeen posited as a possible site of planet formation because of theextremely high fraction of crystalline silicates, which have possi-bly formed by a massive companion that induced spiral densitywaves in the disk material (van Boekel et al. 2004; Avenhauset al. 2017). Fukagawa et al. (2006) imaged in scattered light theouter edge of the gap as well as a spiral feature in the outer disk.These features have recently been confirmed in the visible polar-ized light by Avenhaus et al. (2017). The protoplanetary disk ofHD142527 shows high near-infrared excess (Malfait et al. 1999;Fukagawa et al. 2006) that indicates that optically thick materiallies close to the star; this might be a remnant of the original innerdisk. The central star still accretes at a rate of 9 . × − M (cid:12) / yr (Garcia Lopez et al. 2006; Casassus et al. 2012; Avenhaus et al.2014; Biller et al. 2012; Salyk et al. 2013), suggesting that thereis still enough material close to the star for it to be funneled ontothe star. The outer radius of this inner disk is likely within 10 au(Verhoe ff et al. 2011; Fukagawa et al. 2013; Close et al. 2014).Fukagawa et al. (2006) also found an o ff set of 20 au betweenthe location of the star and the center of the disk, which maybe caused by an unseen (at the epoch) eccentric binary compan-ion. The gap appears to be completely depleted of both large andsmall dust grains, as evidenced from scattered light images aswell as (sub-) millimeter observations. Nonetheless, CO gas hasbeen detected within the gap (Casassus et al. 2013).Using SAM with NACO (short for NAOS-CONICA or Nas-myth adaptive optics system near infrared imager and spectro-graph, Tuthill et al. 2010) at the Very Large Telescope (VLT),Biller et al. (2012) detected an asymmetry in the brightness dis-tribution around the central star with a barycenter emission lo-cated at a projected separation of 88 ± . ± . . ± . . ± . . ± . σ errors),relative to the primary star and disk. They interpreted this asym-metry as a low-mass stellar companion ( ∼ . (cid:12) ) orbiting at ≈
12 au from the star, well inside the gap. On the basis of theirobservations with near infrared coronagraphic imager (NICI) atGemini south, Casassus et al. (2013) disputed the presence ofa companion, but Close et al. (2014) confirmed the companion’sexistence through direct-imaging observations in the R and in H α bands. The latter implies mass accretion, and the authors quan-tified it as ∼ . × − M (cid:12) yr − . Recently, Lacour et al. (2016)observed HD142527 from R − to M –band wavelengths with theNACO and Gemini Planet Imager (Greenbaum et al. 2014) in-struments using the SAM technique. They constrained the com-panion mass and radius with evolutionary models as 0 . ± . (cid:12) and 0 . ± .
15 R (cid:12) , respectively, and derived a younger age(1 . + . − . Myr) than for HD142527A. The characteristics of thissystem, and in particular the existence of the wide gap beyondthe low-mass companion, make the HD142527 system a primetarget for the search for circumbinary planets (Bonavita et al.2016).
3. Observations
The SPHERE planet-finder instrument installed at the VLT(Beuzit et al. 2008) is a highly specialized instrument dedi-cated to high-contrast imaging at optical and near-infrared wave-lengths. It is equipped with an extreme adaptive optics systemcalled SAXO (Sphere Adaptive Optics for eXoplanet Observa-tion) (Fusco et al. 2014; Petit et al. 2014), with a 41 ×
41 actua-tor wavefront control, pupil stabilization, and di ff erential tip-tiltcontrol. It also employs stress-polished toric mirrors for beamtransportation (Hugot et al. 2012). The SPHERE instrument isequipped with several coronagraphic devices for stellar di ff rac-tion suppression, including apodized Lyot coronagraphs (Soum-mer 2005) and achromatic four-quadrant phase masks (Boc-caletti et al. 2008). The instrument has three science subsystems:the infrared dual-band imager and spectrograph (IRDIS; Dohlenet al. 2008), an integral field spectrograph (IFS; Claudi et al.2008, 2016), and a rapid-switching imaging polarimeter (ZIM-POL; Thalmann et al. 2008).Our observations were part of the SHINE (SpHere INfraredsurvey for Exoplanets, Chauvin et al. 2017) survey and were per-formed with SPHERE in the IRDIFS_EXT mode (direct imag-ing). In this mode, IRDIS observes in dual-band imaging (DBI;Vigan et al. 2010) with the K12 filter pair (wavelength K1 = Article number, page 2 of 15laudi et SPHERE GTO Cons.: SPHERE dynamical and spectroscopic characterization of HD142527B
Table 1.
HD142527 Observations.
UT Date Instr. Mode Instr. Filter DIT × NDIT a N a exp Field Rot. Seeing Strehl True NorthCorrection(s) ( ◦ ) (”) @1 . µ m ( ◦ )2015-05-13 IRDIFS_EXT IFS YJH / ND1 4.0 ×
56 8 40.10 0.3 0.8 − . ± . / ND1 0.84 ×
88 16IFS YJH / ND2 8.0 × / ND2 8.0 × × − . ± . ×
32 122016-03-26 IRDIFS_EXT IFS YJH / ND1 4.0 ×
56 16 73.84 0.7 0.7 − . ± . / ND1 0.84 ×
88 32IFS YJH / ND2 8.0 × / ND2 2.0 ×
19 122016-06-13 IRDIFS_EXT IFS YJH / ND1 4.0 ×
60 16 64.48 0.8 0.6 − . ± . / ND1 0.84 ×
49 642017-05-16 SAM IFS YJH 4.0 ×
16 24 73.00 0.6 0.7 − . ± . ×
32 4IRDIS K1–K2 0.84 ×
52 202018-04-14 SAM IFS YJH 2.0 ×
18 16 52.00 0.6 0.9 − . ± . ×
36 16 a NDIT refers to the number of integrations per datacube, DIT to the integration time, N exp to the number of datacubes µ m; K2 = µ m), while IFS obtains low-resolution (R = µ m. The target has been observedwith SAM both in IRDIFS and IRDIFS_EXT mode. In contrastto the IRDIFS_EXT mode, in the IFS mode IRDIS observes indual-band imaging with H23 filter pair (wavelength H2 = µ m; H3 = µ m), while IFS performs low-resolution spec-troscopy at R ∼
50 in the wavelength range 0.95 – 1.35 µ m. We observed HD142527 during five nights (2015–03–13, 2016–03–26, 2016–06–13, 2017–05–16, and 2018–04–14) as part ofthe SPHERE GTO program (IRDIFS and SAM) plus one in tech-nical time dedicated to the commissioning of the SAM observ-ing mode (2015–07–03). The instrumental setups used duringthese observations are all described in Table 1. The main goal ofthese observations was to detect and characterize the stellar com-panion around HD142527 and to place strong constraints on thepresence of other possible companions. Because the star – com-panion separation is expected to be smaller than (88 ± /
10 neu-tral density filter (ND1) were slightly saturated. For this reasonwe took additional unsaturated exposures using the 1 /
100 neutraldensity (ND2) filter to properly calibrate the flux from the starand ND1 images. No saturation was present in the IFS images,so all IFS images could be used as science images. The observ-ing conditions were worse for the IRDIS observations, which arenot saturated on the third night, so that we were able to use onlythe ND1 filter to take our science images. In order to attenu-ate the residual speckle noise with the angular di ff erential imag-ing (ADI; Marois et al. 2006a) technique in the post-processingphase, we acquired our observations in pupil-stabilized mode. HD142527 was also observed in the SAM mode of SPHERE(Cheetham et al. 2016) on the night of March 7, 2015, duringa technical night to test this observing mode, and additionally,as part of SHINE, on the nights of May 16, 2017, and April14, 2018. This was the first time in which the SAM mode wasused with SPHERE. The object was observed in IRDIFS (July3, 2015) and in IRDIFS_EXT (May 16, 2017 and April 14,2018) modes with a seven-hole pupil-aperture mask with thesame layout as was used with NACO. The observing conditionsduring the 2015 observation were poor, with thick clouds, low-coherence times, and high values of seeing a ff ecting the perfor-mance of the AO system.To calibrate the systematic e ff ects that are present in SAM(and similar interferometric techniques), the calibrator stars HD142695 and HD 142277 were also observed. The observationsused 0.84 s and 8 s exposure times for IRDIS and IFS, respec-tively (see Table 1).
4. Data reduction and analysis
The data gathered in the six observation nights have been re-duced following the necessary data reduction recipes for the in-dividual instruments in both the classical (IRDIFS) mode andIFS and IRDIS in SAM mode. In this paragraph the data reduc-tion is discussed for each instrument.
For the IFS data we began with the reduction of calibration data(dark, detector flat, spectral position frames, wavelength calibra-tion, and instrument flat) using the data reduction and handling(DRH) software (Pavlov et al. 2008). For a more detailed de-scription of each of these steps, we refer to Mesa et al. (2015).The raw science data frames were then averaged so that the ro-tation between any two frames was on the order of 0.3 degrees.This left us with 128 di ff erent frames for the first night and 248 Article number, page 3 of 15 & A proofs: manuscript no. HD142527_V8.1_arxiv
Fig. 1.
Non-coronagraphic IRDIFS_EXT and SAM images of the HD142527 system.
Upper panel : Images acquired on June 13, 2016.
Left :PCA post-processing IFS image (averaged over all channels). The red cross marks the position of the central star. In this image the companion ofHD142527 is clearly visible.
Right : Central part of the PCA post-processing IRDIS image of HD142527.
Central panel : Composition of the threeRDI post-processing IFS images (from left to right: May 2015, March 2016, and June 2016). The orbital motion of B is clearly detected.
Bottompanel : Reconstructed images produced from the SAM data using the MiRA algorithm. The images show that we observe significant orbital motionfor the companion HD 142527B between the SAM epochs, and that the point-source model used to calculate its position is a good approximationto the observed structure for the second night, when a larger dataset was obtained. For thethird night, we had 96 frames after binning. For each of theseframes, we then corrected the bad pixels and the e ff ects of thecross-talk between di ff erent lenslets of the IFU (Antichi et al.2009) using dedicated IDL procedures described in more detailin Mesa et al. (2015). On each frame, we then ran the DRHscience recipe, which uses the calibration files and produces awavelength-calibrated datacube composed of 39 monochromaticimages. Each of these images was then corrected for the dif-ferent exposure times and neutral density filters used. We thendetermined the position of the star for each frame using the CN- TRD IDL procedure and recentered each of them to the nom-inal position of the image center. Finally, to retrieve the com-panion image, we performed a principal component analysis(PCA) procedure as described in Mesa et al. (2015), exploit-ing both the angular and spectral information. As we observedthe object in a non-coronagraphic mode and to reduce concernfor the self-cancellation that is present in the PCA method, wealso performed a point spread function (PSF) subtraction (ref-erence di ff erential imaging, RDI) in a region very close to thestar. We devised an automatic search in the SPHERE Data Cen- http: // / docs / cntrd.htmlArticle number, page 4 of 15laudi et SPHERE GTO Cons.: SPHERE dynamical and spectroscopic characterization of HD142527B ter (SDC, Delorme et al. 2017) for reference sequences obtainedusing the same observing mode (non-coronagraphic observationin IRDIFS_EXT mode). We found that the best match was anobservation of HD100546, which provides the highest correla-tion with the observation of HD142527, taken in February 2017,although this star also hosts a bright disk. This sequence in-cludes 533 individual 3D (x, y, lambda) datacubes. We are awarethat HD100546 has a bright disk that is well visible in scatteredlight (see, e.g., Sissa et al. (2018)). However, first, the model isobtained by an automatic procedure. Second, the inner disk ofHD 100546 is not resolved in SPHERE images and the appar-ent separation of the inner edge of the intermediate ring is atabout twice that of the companion of HD142527. Hence, thedisk of HD 100546 has no impact on the present discussion.For each separate wavelength image, we first accurately recen-tered each image on the peak of the di ff raction image. For eachmonochromatic image of the science datacube, our code thensearched for the monochromatic image of the reference datacubethat provides the highest cross-correlation within a circle witha radius of 12 pixels from the nominal center. This is selectedas the best -matching monochromatic image. The flux of thebest-matching monochromatic image is normalized to the valueof the science monochromatic image and then subtracted. Thesubtracted image is normalized to the peak of the original im-age, thus we therefore obtain a contrast image. The subtractedmonochromatic image is derotated and a median of the derotatedmonochromatic images is then made over time. A high-pass fil-tering is then made by subtracting the current median over anarea of 21 ×
21 pixels centered on each pixel. The final imagesare then obtained by collapsing the datacubes along the spectralaxis.
Data reduction for the IRDIS observations was performed fol-lowing the procedures described in Zurlo et al. (2014, 2016). TheIRDIS raw images were pre-reduced by performing backgroundsubtraction, bad-pixel correction, and flat fielding. As these arenon-coronagraphic images, no satellite spots or PSF referenceimages were taken. Satellite spots are fiducial spots symmet-ric with respect to the central star created by using a periodicgrid mask downstream a coronagraph (Sivaramakrishnan & Op-penheimer 2006; Marois et al. 2006b) or by applying a periodicmodulation on an adaptive optics deformable mirror (Langloiset al. 2013). We used one of the images in the sequence as a ref-erence in order to apply the SDC data reduction procedure. Theframes within the resulting datacube were then aligned with oneanother after finding for each of them the location of the star us-ing the CNTRD IDL procedure and recentering each of them tothe nominal position of the image center. For all epochs, afterthe preprocessing of each frame, the speckle pattern subtractionwas performed using both the PCA (Soummer et al. 2012) andthe TLOCI (Marois et al. 2014) algorithms, combined with theADI technique.
The SAM IFS data were converted into cleaned wavelength-extracted cubes using the SPHERE DRH in the same manneras the coronagraphic frames, but without any data binning. TheIRDIS and IFS wavelength cubes were then processed using theIDL-based aperture masking pipeline developed at the Univer-sity of Sydney, with recent modifications allowing the simul- taneous processing of multiwavelength data. A more thoroughdescription of the pipeline can be found in Tuthill et al. (2000),Kraus et al. (2008), and the references therein, but a brief sum-mary follows.The data were background subtracted, flat fielded, andcleaned of bad pixels and cosmic rays. The cleaned cubes werethen windowed with a super-Gaussian function of the form e ( − ar ) before closure phases were extracted from the Fouriertransforms of the images. Calibration of the closure phases wasperformed on each wavelength individually by subtracting aweighted sum of the corresponding measurements taken on thecalibrator stars.The calibrated closure phases were fit with a binary analyt-ical model with the following free parameters: the separation,position angle, and a contrast value for each wavelength chan-nel. The IFS and IRDIS data were fit separately, resulting infour free parameters for IRDIS and 41 parameters for IFS. Thebest-fit parameters were estimated using emcee - a Python im-plementation of the a ffi ne-invariant Markov chain Monte Carlo(MCMC) ensemble sampler (Foreman-Mackey et al. 2013). Toaccount for uncalibrated systematics, a constant was added inquadrature to each closure phase uncertainty to ensure that thebest-fitting model had a reduced χ of 1. These constants wereestimated to be 0 . ◦ and 0 . ◦ for IRDIS and IFS, respectively,in comparison to initial median uncertainties of 0 . ◦ and 0 . ◦ .These high values suggest that the imperfect calibration domi-nates the uncertainties.To estimate the detection limits from the observations, aMonte Carlo simulation was performed. We generated a set of10,000 simulated datasets drawn from a Gaussian distribution,consistent with the measured uncertainties. For each point on agrid of separation and contrast, our detection limits were definedas the point at which at least 99.9% of the datasets were fit bet-ter by a point-source model than the binary model. The 99.9%criteria yields a set of 3.3 σ detection limits, which were approx-imately 7.5 mag for separations between 50-250 mas.
5. Results
The final non-coronagraphic images for IFS, IRDIS, and SAMare shown in Figure 1. Images are from the last night ofIRDIFS_EXT observations; similar images were obtained on theother two nights when the system was observed with the sameinstrumental mode. The companion is clearly visible in the IFSimages with a signal-to-noise ratio (S / N) of ∼
30 (the procedureused to evaluate the S / N is fully described in Zurlo et al. 2014),but it is only marginally resolved in the IRDIS frames becauseof the very small angular separation of the companion and theless favorable pixel scale of IRDIS. Therefore, we did not usethe IRDIS non-coronagraphic data for the astrometric and pho-tometric characterization of HD142527B. The central panel ofthe same figure shows a composition of three post-processingIFS images of the HD142527 system taken at di ff erent epochs(May 2015, March 2016, and June 2016). The companion is alsoclearly visible in the SAM data with an S / N greater than 60 withIRDIS and greater than 40 with IFS. The S / N in the SAM ob-servations was calculated by comparison of the best-fit flux ratiowith the detection limits.We used the MiRA image reconstruc-tion algorithm (Thiébaut 2008) to produce the images in the bot-tom panel of Figure 1. MiRA uses an inverse-problem approachto reconstruct an image from the limited information providedby the closure phases and power spectrum. The images confirmthat the data are consistent with a binary companion and showits orbital motion between the IFS and SAM datasets.
Article number, page 5 of 15 & A proofs: manuscript no. HD142527_V8.1_arxiv
Fig. 2.
Comparison among the extracted spectra for HD142527B ob-tained by both direct imaging and SAM modes of SPHERE in the IFSwavelength range (0 . − . µ m). In the upper part of the plot is in-dicated the value of the average value of the error for each spectrum.The SAM IFS spectrum taken in 2015 is limited to the YJ band of IFS(0 . − . µ m). For a clearer view of the plot all points are connectedby continuous lines. Figure 1 shows that the PCA and RDI post-processing tech-niques allow us to very clearly detect HD142527B in IFS data.However, at this small separation, self-subtraction will severelydiminish the flux in the spectrum of the companion. To avoidself-subtraction due to PCA, we calculated photometry fromeach IFS spectral channel and the corresponding errors fromimages obtained with RDI, exploiting the method described inZurlo et al. (2014). All the extracted spectra (for IFS and SAM)are shown together in Figure 2. In order to facilitate compari-son between the spectra taken at di ff erent epochs, we plot onlypoints falling in the IFS wavelength range (0 . − . µ m). TheH H and K K points from SAM observations are not shown,but are listed in Table 2.Each extracted spectrum of HD142527B can be fit with spec-tra of young field dwarfs from Allers & Liu (2013) to estimateits spectral type. This procedure was executed for all the epochs,and the results are shown in Table 3. The low reduced χ values( <
1) that we obtain in some fits are probably due to the combi-nation of two e ff ects: i ) all the errors could be overestimates, and ii ) some of the 39 measurements of each spectrum could have acertain covariance degree between closer spectral channels. Thefit is useful, however, to form an impression of the spectral typeof the secondary. Figure 3 shows the best-fitting spectra overlaidon the IFS spectrum from May 14, 2015. In order to constrainthe physical characteristics of the companion, we compared theextracted spectra with a set of BT-Settl models (Allard 2014).The models were selected in a grid with the e ff ective tempera-ture and the surface gravity ranging in the following intervals:1000 ≤ T e f f ≤ . ≤ log( g ) ≤ .
5, with an in-cremental step of 100 K for the former and 0.5 for the latter. Inall cases the fit with models with T e f f = − g ) ∼ . Fig. 3. Top : Extracted spectrum for HD142527B (green squares) com-pared with the three best-fit spectra (red, orange, and blue solid lines)from the Allers & Liu (2013) obtained from the data taken on the nightof May 13, 2015.
Bottom : Same as the left panel, but compared withthe three best-fit models (red, orange, and blue solid lines) from theBT-Settl library Allard (2014).
The fit results for a range of spectral types are presented inTable 3. The best-fit spectra match well at all epochs (except forthe spectrum from June 13, 2016), thus we determine a spectraltype between M5 and M6 for HD142527B with an uncertainty of ± e f f correspond-ing to the best-fit spectral type for each individual observation(Col. 3 of Table 3). The extracted spectra seem to be di ff erentfrom each other (see Figure 2). We evaluated the Y–H color ofeach spectrum using the value of the magnitude discussed in thenext section and reported in Table 2. We obtain a variation from0.85 up to 1.40 with an average value of 1 . ± .
20 for the Y–Hcolor. Even if the Y–H color of HD142527B seems to change atdi ff erent epochs, the measurement errors we estimate thereforedo not allow us to be confident about this color variation. Thisis also supported by the spectrum obtained in the night of June13, 2016, which has a bluer color (Y–H = . ± .
28) than theothers, but has a later spectral type (M7V, see Table 3) accordingto the fitting procedure with young field dwarf spectra. These re-sults do not allow us to draw a firm conclusion on the spectraltype of HD142527B. It could be both later or earlier than theM5–M6 spectral type stated here.
We estimated the wide-band photometry in the Y, J, and H spec-tral bands by adopting a distance of 156 pc and using the mag-nitude for the star obtained from 2MASS (Cutri et al. 2003). Wecombined the median contrasts obtained for all the wavelengthchannels below 1.15 µ m for the Y band, between 1.15 and 1.35 µ m for the J band, and between 1.35 and 1.65 µ m for the H band. Article number, page 6 of 15laudi et SPHERE GTO Cons.: SPHERE dynamical and spectroscopic characterization of HD142527B
Table 2.
Absolute magnitudes of HD142527 B obtained from the IFS data for Y, J, and H spectral bands. SAM data also contain the H , H and K and K bands of IRDIS . These magnitudes are not corrected for the several reddening contributions from the inner part of the HD142527 system(see text). Date Y J H H H K K ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± = . ± .
21 mag, J = . ± .
17 mag, and H = . ± .
19 magas mean values for the absolute magnitudes of the secondary.In the determination of the absolute magnitudes ofHD142527B, we considered that the interstellar reddening cor-rection for the primary and the secondary are identical, that is,assuming for the system a visible extinction of A V = . ff et al. 2011; Lacour et al. 2016) and a dust reddening of R V = . E ( B − V ) = . >
90 %) due to the stel-lar surface. The remaining 10%, which is due to the circumstel-lar emission, mainly a ff ects the J band, but the Y band is as-sumed to be una ff ected. However, the red color of HD142527B(J-H = . ± .
25 mag) is consistent with the presence of anoptically thick circum-secondary disk around HD142527B it-self, as has previously been identified by Lacour et al. (2016)and Biller et al. (2012). Furthermore, Close et al. (2014), wereable to detect HD142527B by means of H α (0 . µ m) obser-vations that also allowed them to evaluate the mass accretionrate onto HD142527B (5 . × − M (cid:12) yr − ), which is about 1%of the accretion onto the primary star (Close et al. 2014). Boththese contributions (the disk and the accretion) are significant atlonger wavelengths. From the NIR photometry of Biller et al.(2012), we can assume, as Close et al. did, a ∼ . H = .
09 mag for the Hband. To summarize, the absolute magnitudes of HD142527B,taking into account the circumsecondary material discussed be-fore, become Y = . ± .
21 mag, J coor = . ± .
17 mag andH corr = . ± .
19 mag.It is worth pointing out that the ZIMPOL polarimetric ob-servation made by Avenhaus et al. (2017) did not find any fluxenhancement that would have been due to a dust disk aroundHD142527B, even though they reached a better contrast thanthe observations reported in Rodigas et al. (2014), where such adisk is observed. Therefore Avenhause et al. did not confirm thepresence of the circum-secondary disk. However, considering asimple model of a reflective optical disk with a radius ≤ / hill Table 3.
Reduced χ value for the fits of the HD142527B with the coolfield dwarfs of the Allers & Liu (2013) sample. The e ff ective tempera-tures of the di ff erent spectral types are taken from Pecaut & Mamajek(2013). Date Sp T e f f
Reduced χ a ±
200 0.4432015-07-03 M5V b ±
200 1.0132016-03-26 M5V b ±
200 1.7602016-06-13 M7V c ±
200 0.4192017-05-16 M6V a ±
200 1.035 a + b c σ limit for the observation of HD142527presented by Avenhaus et al. (2017, panel a of their Figure 2).These magnitudes suggest some degree of variability be-tween the di ff erent epochs, with a peak-to-valley excursion of0.52 in the Y band and about 0.4 and 0.5 mag in the J and Hband. It is not clear if this is a real e ff ect or an artifact of our re-ductions. Since we worked di ff erentially with respect to the pri-mary, we should take into consideration the possibility that thisis a variable star. Primary variability was noted by Biller et al.(2012), who reported a significant variation between the 2MASSphotometry and the Malfait et al. (1998) photometry. HD142527is a Herbig Ae object that is listed as probable δ Scuti pulsatorin Marconi & Palla (1998). However, Kurtz & Müller (2001) didnot find any δ Scuti pulsation, but a stellar variability with a pe-riod of about 6.0 d. We observed the star (Messina & Millward,2017 Priv. Comm.) from April 2017 to August 2017 for a totalof 13 nights, confirming the period of 6.0 d with a peak-to-valleyamplitude of 0.13 mag in B and 0.09 mag in V and R. We con-clude that the photometric variability amplitude of the primary isnot su ffi cient to explain the amplitude of the variation we foundin our observation of B. On the other hand, in an extensive NIRphotometrical study of the RCW38 star-forming region, Dörret al. (2013) found that most of the low-mass stars in their sam-ple exhibit irregular light curves with typical timescales of a fewdays and amplitudes between 0.1 and 0.4 mag in K band. Someof them show variations and outbursts with amplitudes above 1mag. If this is the case for HD142527B, as it may be becauseits very young age (1.0 - 5.0 Myr), the variable behavior of NIRphotometry of young stars can most likely explain the photomet-rical variation shown in Table 2. At the York Creek Observatory, Georgetown, AustraliaArticle number, page 7 of 15 & A proofs: manuscript no. HD142527_V8.1_arxiv
Fig. 4. Top left : IFS and IRDIS contrast limits achieved within 250 mas using the SAM observing mode.
Top right : IFS and IRDIS contrast limitsachieved during the non-coronagraphic observations of HD142527.
Bottom left : Minimum detectable mass as a function of separation in the fieldof view of IFS and IRDIS, adopting an age of 1 Myr and the models of BHAC2015.
Bottom right : Minimum detectable mass as a function ofseparation, adopting an age of 5 Myr. The shaded area between the two dashed vertical lines indicates the disk gap as reported by Verhoe ff et al.(2011). To precisely determine the position of the companion from theIFS images taken at di ff erent observing epochs, we insertednegative-simulated companions (built from a 2D Gaussian func-tion) at di ff erent positions around the position of the true com-panion. The best fit to this position was found from the po-sition of the negative simulated companion, which minimizedthe standard deviation in the region around it. The results foreach epoch are listed in Table 4. Since the observations are non-coronagraphic, the position of the primary star could be definedwith high precision because its signal is strong and unbiased.This shows that the dominant source of uncertainty remains thecentering of the secondary star, which is assumed to be accurateto half of a pixel scale. The main contribution to the error on theposition angle is dominated by the uncertainty on the true north(TN) angle, calculated by observing an astrometric calibrationfield (Maire et al. 2016). The best limiting contrast curves as a function of separation ob-tained for both IFS and IRDIS (see Mesa et al. (2015)) for acomplete description of the method we used to evaluate the lim-iting contrast curves) are shown in the upper panels of Figure4. Here we also show the best limiting contrast curve obtainedwith the SAM observing mode for IRDIS and IFS for separa- tions smaller than 250 mas (39 au). The lower panel shows theminimum detectable mass as a function of separation in the fieldof view (FoV) of IFS and IRDIS. In the figure, the inner andouter limits of the disk gap are also highlighted with verticaldashed lines. We considered the mass limits for ages of 5 Myr,as stated for HD142527 by Fukagawa et al. (2006), and of 1 Myr,which resulted from comparing the photometry of HD142527Bwith models from Bara ff e et al. (2015, henceforth BHAC 2015)(see Section 6.1). The non-coronagraphic setup that was used forthe observations means that the achieved contrast at sep ≥ . −
10 M J (depending on the age of the system).Boehler et al. (2017), in their analysis of ALMA observationsof HD142527, found a compact source in the continuum map aswell as CO emission at about 50 au from the central star. They in-terpreted this as material orbiting a low-mass companion. Exceptfor HD142527B, we detect no companion with a mass higherthan 10 M J orbiting HD142527 in the large gap between the in-ner and the circumbinary disks. Consequently, the third object inthe system proposed by Boehler et al. (2017), if it exists, shouldbe less massive than 10 M J . At larger orbital radii up to 500 Article number, page 8 of 15laudi et SPHERE GTO Cons.: SPHERE dynamical and spectroscopic characterization of HD142527B
Table 4.
Astrometric position derived for HD142527B for each individual epoch.
Date ∆ RA ∆ Dec Separation Position Angle(mas) (mas) (mas) (deg)2015-05-13 65 ± − ± ± ± ± − . ± . ± ± ± − ± ± ± ± − ± ± ± . ± . . ± . ± ± ± ± ± . ± . − J (depending on the age of the system) are detected.
6. HD142527B mass estimate
The spectroscopic results discussed in Section 5.1 can be usedto estimate the mass and radius of HD142527B. The analysis ofPecaut & Mamajek (2013) on cool pre-main-sequence stars alsoprovides a tentative determination of the mass and radius of suchobjects based on spectral type. Our determination of the spectraltype yields a value of M = . + . − . M (cid:12) and R = . + . − . R (cid:12) . The mass, radius, and age of this star can be obtained by com-parisons between the measured absolute magnitudes and modelabsolute magnitudes from theoretical evolutionary tracks. In thiscase, we considered the standard evolutionary models of BHAC2015. In Figure 5 we display the isochrones for models withmasses ranging between 0.05 M (cid:12) to 0.80 M (cid:12) and for ages be-tween 0.5 Myr to 8 Myr, on which we overlay the observedpoints (blue dots) at di ff erent epochs and their average. A suit-able set of parameters for this comparison is M = . ± . (cid:12) , R = . ± . (cid:12) , and an age of about 1 Myr. Although thismass determination agrees with our previous estimate based onspectroscopic data and with the result reported in Lacour et al.(2016), it is most likely not conclusive. The absolute magnitudesof HD142527B are systematically too bright compared to themodels. This is mainly due to the accretion of matter onto thisyoung star (Close et al. 2014) that is not taken into account inthe evolutionary tracks of structures, which instead only accountfor gravitational contraction. The accretion produces a hotter ob-ject with a larger radius. We tried to account for the contributionof the disk (see Section 5.2), but other e ff ects can contribute tothe luminosity of the secondary, such as the inner disk. It is ex-pected that if such e ff ects were accounted for in the isochrones,the points would shift toward larger magnitudes. If we were totake the higher temperature and brighter absolute magnitude thatis due to accretion into account in the model, the isochrone sys-tem would move mainly parallel to the arrow in the Figure 5,rendering this analysis inconclusive. An alternative and promising way to constrain the mass ofHD142527B is to evaluate the proper motion variation of theprimary star that is due to the presence of the secondary. To thispurpose, we searched for significant di ff erences in proper mo-tion of the star as measured at di ff erent epochs (see Table 5).HD142527 is present in several catalogs, and the comparisonof GAIA and Tycho2 (Høg et al. 2000) proper motions gives ∆ µ α = (2 . ± .
00) mas / yr and ∆ µ δ = (1 . ± .
00) mas / yr. The Fig. 5.
Evolutionary tracks and isocontours of masses as function of thetemperature and absolute magnitude J as evaluated in BAHC 2015 Theblue squares are the observed absolute magnitudes of HD142527B cor-rected for the contribution due to the circum-secondary disk, and theblack square is their mean with error bars. The arrow qualitatively in-dicates how the isochrone system would change with respect to the ob-served points if the accretion process is taken into account in the models.
GAIA and SPM (Girard et al. 2011, Southern Proper MotionCatalogue) ∆ µ α = (2 . ± .
40) mas / yr and ∆ µ δ = (9 . ± . / yr and GAIA and UCAC5 (Zacharias et al. 2017) ∆ µ α = (1 . ± .
9) mas / yr and ∆ µ δ = (0 . ± .
90) mas / yr also showsignificant di ff erences in proper motion.We were therefore able to use the code for orbitalparametrization of astrometrically identified new systems (CO-PAINS, Fontanive et al. in prep.) to evaluate the characteristics ofthe possible companions that are compatible with the observed ∆ µ . The code uses Eq. 1 from Makarov & Kaplan (2005), de-rived by Makarov & Kaplan (2005), to estimate the change in astellar proper motion that is induced by a companion for a rangeof possible masses and separations, ∆ µ ≤ π Π R M √ aM Tot . (1)where, M is the mass of the secondary, M Tot is the total massof the binary, a is the semi-major axis in AU, Π is the parallaxof the system in mas, and R takes into account the orbital phaseso that R = (cid:16) + e cos E − e cos E (cid:17) / , where e is the orbital eccentricityand E is the eccentric anomaly. A fine grid of mass and sepa-ration values is explored, and the expected ∆ µ is evaluated andcompared with the observed one. In order to properly take intoaccount the projection e ff ects, the code considers for each pointon the mass-separation grid 10 possible orbital configurations, Article number, page 9 of 15 & A proofs: manuscript no. HD142527_V8.1_arxiv
Table 5.
Values for di ff erent catalogs and epochs of the HD142527 proper motion. Catalog µ α δµ α µ δ δµ δ p δ p Epochmas / yr mas / yr mas / yr mas / yr mas mas yrTGAS a − .
76 0.08 − .
46 0.05 6.40 0.26 2015.00Hipparcos New − .
19 0.93 − .
46 0.79 4.29 0.98 1991.25Tycho2 − . − . − .
16 3.42 − .
78 3.26 2006.12UCAC5 b − . − . a Tycho – GAIA Astrometric Solution b Fifth US Naval Observatory Astrograph Catalog
Fig. 6.
Mass distribution, obtained using the COPAINS, of the com-panions that are compatible with the observed ∆ µ for HD142527, at aphysical separation that is compatible with the SPHERE detection. Thedashed line shows the position of the most likely value, and the shadedarea highlights the region within a 1 σ confidence level. with eccentricities drawn from a uniform distribution (a Gaus-sian distribution can also be used, see Bonavita et al. (2016) forfurther details and other applications of the code).If an estimate of the orbital parameters is available, as in thecase of HD142527B (see Section 7 for the details of the orbitalcharacterization), the code allows us to retrieve the mass distri-bution for companions compatible with the observed trend andthe orbital characteristics. Figure 6 shows the results of the ap-plication of this method to HD142527B. The retrieved mass dis-tribution peaks at 0 . + . − . M (cid:12) and is therefore compatible withthe value obtained in the analysis based on the spectral classifica-tion and the calibration by Pecaut & Mamajek (2013) discussedat the beginning of this section.
7. Orbital properties of HD142527B
We combined the astrometric measurements of the compan-ion reported in the literature (Close et al. 2014; Rodigas et al.2014; Lacour et al. 2016) with the new SPHERE / IFS andSPHERE / SAM measurements (Table 4) to perform a new orbitalstudy. The measurements are shown in Fig. 7. We show the GPIpolarimetric di ff erential imaging measurement in Rodigas et al.(2014) for comparison, but we did not use it for our analysis be-cause of its large uncertainties and because a more accurate GPISAM measurement close in time is available from Lacour et al.(2016). The SPHERE data represent an increase by almost a fac-tor 3 in the observational baseline with respect to the previousstudy of Lacour et al. (2016). In about four years, the position Fig. 7.
Astrometric measurements of HD142527B from the literature(NaCo, MagAO, and GPI data) and this work (SPHERE data). The GPImeasurement (green data point) is shown for comparison, but is not usedin the analysis (see text). angle of the companion decreased by ∼ ◦ and its separationdecreased by ∼
33 mas. The data in Fig. 7 indicate inflections inboth separation and position angle in the orbital motion of thecompanion. We show the evolution of the separation and posi-tion angle of the companion as a function of time in Fig. 8. Theseparation and position angle measured in 2018 deviate from alinear trend based on the data points listed in Lacour et al. (2016)at a significance of ∼ σ , respectively. In addition, thecompanion separation decreases. We therefore conclude that thecompanion is accelerating on its orbit and approaches its periap-sis. We used a least-squares Monte Carlo (LSMC) algorithm tofit the astrometric measurements and derive distributions of the Article number, page 10 of 15laudi et SPHERE GTO Cons.: SPHERE dynamical and spectroscopic characterization of HD142527B
Fig. 8.
Temporal evolution of the separation ( top ) and position angle( bottom ) of HD142527B. For both panels, a linear fit to the data pointsobtained before 2015 is shown (purple solid line) to show the progres-sive deviation from linearity of the SPHERE measurements. The GPImeasurements (green data points) are not considered for the fits. orbital parameters. The approach has previously been describedin Esposito et al. (2013) and Maire et al. (2015). We assumeda distance for the system of 156 pc (Gaia Collaboration et al.2016) and a total system mass of 2.1 M (cid:12) (Lacour et al. 2016).We drew 2 000 000 random realizations of the astrometric mea-surements assuming Gaussian distributions around the nominalvalues. Then, we fit the six Campbell elements simultaneouslyusing a debugged version of the downhill simplex AMOEBA al-gorithm (Eastman et al. 2013). Initial guesses for the orbitalelements were drawn assuming uniform distributions. We con-sidered no priors on the orbital elements, except for the period( P = χ < P = i = ◦ , longitude of node Ω ∼ The customized built-in routine provided by IDL truncates the step-ping scales to floating point precision, regardless of the type of inputdata. ◦ , and argument of periapsis ω = ◦ . For the eccentric-ity and time at periapsis passage, each distribution exhibits twogroups of possible values: ∼ ∼ e , and ∼ ∼ T .The shape of the distributions of the orbital parameters arebroadly consistent with those derived in Lacour et al. (2016) us-ing a Markov chain Monte Carlo algorithm for the parametersin common, except that all the solutions are consistent with acompanion approaching its periapsis passage. Some parametersare better constrained by our updated analysis (inclination, lon-gitude of node, and argument of periapsis), whereas others havelarger ranges (period). The latest data points probe very close tothe periapsis passage, hence we cannot firmly conclude whetherthe companion has passed it. Further monitoring is required toaddress this point. Lacour et al. (2016) found two families oforbital solutions, where the companion approaches periapsis inone case and has recently past periapsis in the other case. Theywere not able to distinguish between these two solutions becauseavailable data o ff er only small orbital coverage and the evolutionof the separation and position angle with time does not deviatefrom linearity. Figure 10 shows the predicted separations and po-sition angles for all the orbital solutions derived in our analysis.The position angle (PA) decrease will continue in the comingyears with a high variation rate, while the separation will reachminimum before increasing again starting from ∼ ff erent colors three representative orbits among the orbital so-lutions. The orbits similar to the orbit shown in blue composethe largest group of all three groups ( ∼
77% of all fitted orbits).They are characterized by a longitude of node larger than 100 ◦ and periods longer than ∼
40 yr. The orbits represented by theorbit marked in red are the second largest group ( ∼
22% of allfitted orbits) and have longitudes of node larger than 100 ◦ , butperiods shorter than 40 yr. Finally, there is a small group of or-bits shown by the orbit colored in green ( ∼ ∼
40 yr, but longitudes of node smaller than 100 ◦ .These three types of orbits also have di ff erent times at perias-tron passages, the first group have T before ∼ ∼ ∼ ∼ M (cid:12) companion with a 50 au apoap-sis and e ∼ . − . ∼
100 au (Holman & Wiegert1999), which is the innermost possible location for the outer disk(e.g., Fukagawa et al. 2006; Casassus et al. 2012; Rameau et al.2012; Rodigas et al. 2014; Avenhaus et al. 2017). Nevertheless,the inclination and longitude of node of HD142527B disagreewith those of the outer circumstellar disk ( i = ◦ , Ω= ◦ , Ver-hoe ff et al. 2011; Perez et al. 2015), as has been reported byLacour et al. (2016). This result would at first sight rule out thatHD142527B is responsible for the outer disk truncation. How-ever, recent hydrodynamical simulations have shown that for aneccentric companion with an almost polar inclination to the outerdisk that approaches its periapsis passage, the interactions of Article number, page 11 of 15 & A proofs: manuscript no. HD142527_V8.1_arxiv
Fig. 9.
LSMC distributions of the six Campbell orbital elements for all the fitted solutions with χ < ff -diagonal diagrams show the correlations betweenpairs of orbital elements. The linear color scale in the correlation plots accounts for the relative local density of the orbital solutions. In the diagonalhistograms, the red solid line represents the 50 percentile values, the red dashed lines show the intervals at 68%, and the green solid line indicatesthe best χ fitted solution. companion and disk can reproduce several of the main observeddisk features, such as its large cavity, and with the correct posi-tion angles, its spiral features and shadows (Price et al. 2018).This orbital configuration is broadly consistent with the resultsfrom our orbital analysis. On the other hand, the orbital planeof the companion is close to the plane of the inner circumstellardisk, as previously suggested by Lacour et al. (2016). The innerdisk has a position angle of 110 ± ◦ from CO(6-5) kinematicsmeasured with ALMA (Casassus et al. 2015). Its mean radiusis estimated to be about 10 au from near-infrared interferometric observations (Anthonioz et al., in prep.). From MIR imaging andSED modeling, Verhoe ff et al. (2011) derived a maximum radialextension for the inner disk up to 30 au. However, Avenhauset al. (2014) imaged the inner circumstellar environment downto ∼ (cid:48)(cid:48) (15 au) in polarized scattered light with NaCo, but didnot detect traces of an inner disk. More recently, Avenhaus et al.(2017) imaged the disk in visible polarized light with SPHEREdown to 25 mas ( ∼ Article number, page 12 of 15laudi et SPHERE GTO Cons.: SPHERE dynamical and spectroscopic characterization of HD142527B
Fig. 10.
Predicted separations ( left ), position angles ( middle ), and sky-plane positions ( right ) for the subset of 100 randomly selected orbitalsolutions shown in Fig. 9. In the right panel we also show with di ff erent colors three representative orbits among the fitted solutions (see text) andan inset providing a zoom around the region that is covered by the data points. Fig. 11.
Same as the diagonal plots in Fig. 9, but for the separation at apoapsis (left) and periapsis (right). with predictions from a modeling of the shadows projected ontothe outer disk (Marino et al. 2015). Using our orbital analysis ofthe companion, we derived the distribution of its separation atperiapsis (Fig. 11, right panel) and find a 68% confidence inter-val of ∼ −
12 au. This results indicates that the shape of theinner disk is strongly a ff ected by HD142527B.
8. Discussion and conclusion
We here presented a detailed characterization of the stellarcompanion of HD142527. We refined the previously estimatedspectral and orbital characteristics of HD142527B using IFSlow-resolution spectra in both IRDIFS_EXT observing modeand also using the SAM technique without a coronagraph. Be-cause the observations were taken in non-coronagraphic mode,the IRDIS images taken in IRDIFS_EXT mode were used notfor photometry, but only to detect the mass limit. Except forHD142527B, we detect no objects with masses greater than 10M J inside the gap of the disk, based on our achieved IFS con-trasts. This constrains the mass of the third object hypothesizedby Boehler et al. (2017) at 50 au from the central star to lowerthan 10 M J . The images obtained with IRDIS exclude the pres-ence of planets with masses greater than 7 M J up to 500 au.However, the non-coronagraphic technique we used in our ob-servations means that the achieved contrast at sep ≥ . e f f between 3030 and 2650 K.The spectra are variable in flux because of several possible fac-tors, such as 1) the variation in stellar temperature, or 2) the con-tribution from the accretion disk around the secondary, and / or 3)the variation in circumstellar material absorption (around the pri-mary or secondary, or around both) that is due to interaction withthe environment and to the young age of the system. In compar-ing the absolute magnitude of the B component to the models,we find (like Lacour et al. (2016)) that the age of the secondaryis very young ( ∼ Article number, page 13 of 15 & A proofs: manuscript no. HD142527_V8.1_arxiv star, we dynamically constrained the mass of the secondary to beM HD B = . + . − . M (cid:12) . This value is in agreement with ourspectroscopic estimate (see Section 6) and with the estimate ofLacour et al. (2016). Following Pecaut & Mamajek (2013) thelower and upper dynamical mass values correspond to a spectraltype between M2.5 and M5.5. To our knowledge, this is the firstdynamical mass determination for HD142527B that was madeby exploiting the di ff erence between nearly instantaneous andlong-term motion. Significant improvements are expected afterthe full GAIA dataset is available.By combining the new SPHERE / IFS and SPHERE / SAM as-trometric measurements with those reported in the literature, weconstrained the orbital properties of HD142527B and obtained aperiod of P = −
137 yr, an inclination of i = − ◦ degrees,a value of Ω = − ◦ degrees for the longitude of node, andan 68% confidence interval of ∼ −
57 au for the separation atperiapsis. Eccentricity and time at periapsis passage exhibit twogroups of values: ∼ ∼ e , and ∼ ∼ T . The orbit of the secondary is moreinclined than the outer circumstellar disk and seems to rule outthat HD142527B is responsible for the truncation of the outerdisk. On the other hand, these results are also consistent with thescenario described by Price et al. (2018), in which a compan-ion close to its periapsis in an eccentric and highly inclined orbitwith respect to the outer disk could be responsible for the largecavity and other observed features and shadows. Our distribu-tion of the orbital parameters are in good agreement with thoseof Lacour et al. (2016). Our solutions are consistent with a com-panion that is approaching its periapsis passage, but the errorson our derived eccentricity and period are too large (but furthermonitoring is required) to conclusively determine whether thecompanion has passed periapsis. The next two or three years ofobservations will be crucial to clarify this point. Acknowledgements.
SPHERE is an instrument designed and built by a con-sortium consisting of IPAG (Grenoble, France), MPIA (Heidelberg, Germany),LAM (Marseille, France), LESIA (Paris, France), Laboratoire Lagrange (Nice,France), INAF–Osservatorio di Padova (Italy), Observatoire de Genève (Switzer-land), ETH Zurich (Switzerland), NOVA (Netherlands), ONERA (France) andASTRON (Netherlands) in collaboration with ESO. SPHERE was funded byESO, with additional contributions from CNRS (France), MPIA (Germany),INAF (Italy), FINES (Switzerland) and NOVA (Netherlands). SPHERE also re-ceived funding from the European Commission Sixth and Seventh FrameworkProgrammes as part of the Optical Infrared Coordination Network for Astronomy(OPTICON) under grant number Rll3–Ct–2004–001566 for FP6 (2004–2008),grant number 226604 for FP7 (2009–2012) and grant number 312430 for FP7(2013–2016). This work has made use of the SPHERE Data Centre, jointly oper-ated by Osug / Ipag (Grenoble), Pytheas / Lam / Cesam (Marseille), OCA / Lagrange(Nice) and Observatoire de Paris / Lesia (Paris) and supported by a grant fromLabex OSUG@2020 (Investissements d’avenir ANR10 LABX56). This workhas been in particular carried out in the frame of the National Centre for Compe-tence in Research ’PlanetS’ supported by the Swiss National Science Foundation(SNSF). D.M. acknowledges support from the ESO-Government of Chile JointComittee program “Direct imaging and characterization of exoplanets”. A.Z. ac-knowledges support from the CONICYT + PAI / Convocatoria nacional subven-ción a la instalación en la academia, convocatoria 2017 + Folio PAI77170087.J. O. acknowledges support from the ICM (Iniciativa Científica Milenio) via theNucleo Milenio de Formación planetaria grant, from the Universidad de Val-paraíso and from Fondecyt (grant 1180395).
References
Acke, B. & van den Ancker, M. E. 2004, A&A, 426, 151Allard, F. 2014, in IAU Symposium, Vol. 299, Exploring the Formation and Evo-lution of Planetary Systems, ed. M. Booth, B. C. Matthews, & J. R. Graham,271–272Allers, K. N. & Liu, M. C. 2013, ApJ, 772, 79Antichi, J., Dohlen, K., Gratton, R. G., & et al. 2009, ApJ, 695, 1042Avenhaus, H., Quanz, S. P., Schmid, H. M., et al. 2017, AJ, 154, 33Avenhaus, H., Quanz, S. P., Schmid, H. M., et al. 2014, ApJ, 781, 87 Bara ff e, I., Homeier, D., Allard, F., & et al. 2015, A&A, 577, A42Beuzit, J.-L., Feldt, M., Dohlen, K., et al. 2008, in Proc. SPIE, Vol. 7014,Ground-based and Airborne Instrumentation for Astronomy II, 701418Biller, B., Lacour, S., Juhász, A., & et al. 2012, ApJ, 753, L38Biller, B. A., Males, J., Rodigas, T., & et al. 2014, ApJ, 792, L22Boccaletti, A., Abe, L., Baudrand, J., & et al. 2008, in Proc. SPIE, Vol. 7015,Adaptive Optics Systems, 70151BBoehler, Y., Weaver, E., Isella, A., et al. 2017, ApJ, 840, 60Bonavita, M., Desidera, S., Thalmann, C., et al. 2016, A&A, 593, A38Brittain, S. D., Carr, J. S., Najita, J. R., Quanz, S. P., & Meyer, M. R. 2014, ApJ,791, 136Brittain, S. D., Najita, J. R., Carr, J. S., et al. 2013, ApJ, 767, 159Cardelli, J. A., Clayton, G. C., & Mathis, J. S. 1989, ApJ, 345, 245Casassus, S., Marino, S., Pérez, S., et al. 2015, ApJ, 811, 92Casassus, S., Perez M., S., Jordán, A., et al. 2012, ApJ, 754, L31Casassus, S., van der Plas, G., M, S. P., & et al. 2013, Nature, 493, 191Chauvin, G., Desidera, S., Lagrange, A.-M., et al. 2017, in SF2A-2017: Pro-ceedings of the Annual meeting of the French Society of Astronomy and As-trophysics, ed. C. Reylé, P. Di Matteo, F. Herpin, E. Lagadec, A. Lançon,Z. Meliani, & F. Royer, 331–335Cheetham, A. C., Girard, J., Lacour, S., et al. 2016, in Proc. SPIE, Vol. 9907,Optical and Infrared Interferometry and Imaging V, 99072TClaudi, R., Antichi, J., Baru ff olo, A., et al. 2016, in Proc. SPIE, Vol. 9908,Ground-based and Airborne Instrumentation for Astronomy VI, 99083HClaudi, R. U., Turatto, M., Gratton, R. G., & et al. 2008, in Proc. SPIE, Vol.7014, Ground-based and Airborne Instrumentation for Astronomy II, 70143EClose, L. M., Follette, K. B., Males, J. R., & et al. 2014, ApJ, 781, L30Cutri, R. M., Skrutskie, M. F., van Dyk, S., et al. 2003, VizieR Online DataCatalog, 2246Delorme, P., Meunier, N., Albert, D., et al. 2017, in SF2A-2017: Proceedings ofthe Annual meeting of the French Society of Astronomy and Astrophysics,ed. C. Reylé, P. Di Matteo, F. Herpin, E. Lagadec, A. Lançon, Z. Meliani, &F. Royer, 347–361Dodson-Robinson, S. E. & Salyk, C. 2011, ApJ, 738, 131Dohlen, K., Langlois, M., Saisse, M., & et al. 2008, in Proc. SPIE, Vol. 7014,Ground-based and Airborne Instrumentation for Astronomy II, 70143LDörr, M., Chini, R., Haas, M., Lemke, R., & Nürnberger, D. 2013, A&A, 553,A48Eastman, J., Gaudi, B. S., & Agol, E. 2013, PASP, 125, 83Espaillat, C., Muzerolle, J., Najita, J., & et al. 2014, Protostars and Planets VI,497Esposito, S., Mesa, D., Skemer, A., et al. 2013, A&A, 549, A52Fedele, D., Carney, M., Hogerheijde, M. R., et al. 2017, A&A, 600, A72Foreman-Mackey, D., Hogg, D. W., Lang, D., & Goodman, J. 2013, PASP, 125,306Fukagawa, M., Tamura, M., Itoh, Y., et al. 2006, ApJ, 636, L153Fukagawa, M., Tsukagoshi, T., Momose, M., & et al. 2013, PASJ, 65, L14Fusco, T., Sauvage, J.-F., Petit, C., et al. 2014, in Proc. SPIE, Vol. 9148, AdaptiveOptics Systems IV, 91481UGaia Collaboration, Brown, A. G. A., Vallenari, A., et al. 2016, A&A, 595, A2Garcia Lopez, R., Natta, A., Testi, L., & Habart, E. 2006, A&A, 459, 837Girard, T. M., van Altena, W. F., Zacharias, N., et al. 2011, AJ, 142, 15Greenbaum, A. Z., Cheetham, A., Sivaramakrishnan, A., & et al. 2014, inProc. SPIE, Vol. 9147, Ground-based and Airborne Instrumentation for As-tronomy V, 91477BHenize, K. G. 1976, ApJS, 30, 491Høg, E., Fabricius, C., Makarov, V. V., et al. 2000, A&A, 355, L27Holman, M. J. & Wiegert, P. A. 1999, AJ, 117, 621Houk, N. 1978, Michigan catalogue of two-dimensional spectral types for theHD starsHugot, E., Ferrari, M., El Hadi, K., & et al. 2012, A&A, 538, A139Jovanovic, N., Guyon, O., Lozi, J., et al. 2016, in Proc. SPIE, Vol. 9909, AdaptiveOptics Systems V, 99090WKraus, A. L. & Ireland, M. J. 2012, ApJ, 745, 5Kraus, A. L., Ireland, M. J., Martinache, F., & Lloyd, J. P. 2008, ApJ, 679, 762Kurtz, D. W. & Müller, M. 2001, MNRAS, 325, 1341Lacour, S., Biller, B., Cheetham, A., et al. 2016, A&A, 590, A90Langlois, M., Vigan, A., Moutou, C., et al. 2013, in Proceedings of the ThirdAO4ELT Conference, ed. S. Esposito & L. Fini, 63Ligi, R., Vigan, A., Gratton, R., et al. 2018, MNRAS, 473, 1774Macintosh, B. A., Anthony, A., Atwood, J., et al. 2014, in Proc. SPIE, Vol. 9148,Adaptive Optics Systems IV, 91480JMaire, A.-L., Langlois, M., Dohlen, K., et al. 2016, in Proc. SPIE, Vol. 9908,Ground-based and Airborne Instrumentation for Astronomy VI, 990834Maire, A.-L., Skemer, A. J., Hinz, P. M., et al. 2015, A&A, 576, A133Makarov, V. V. & Kaplan, G. H. 2005, AJ, 129, 2420Malfait, K., Bogaert, E., & Waelkens, C. 1998, A&A, 331, 211Malfait, K., Waelkens, C., Bouwman, J., & et al. 1999, A&A, 345, 181Marconi, M. & Palla, F. 1998, ApJ, 507, L141Marino, S., Perez, S., & Casassus, S. 2015, ApJ, 798, L44 Article number, page 14 of 15laudi et SPHERE GTO Cons.: SPHERE dynamical and spectroscopic characterization of HD142527B
Marois, C., Correia, C., Galicher, R., et al. 2014, in Proc. SPIE, Vol. 9148, Adap-tive Optics Systems IV, 91480UMarois, C., Lafrenière, D., Doyon, R., & et al. 2006a, ApJ, 641, 556Marois, C., Lafrenière, D., Macintosh, B., & Doyon, R. 2006b, ApJ, 647, 612Mendigutía, I., Fairlamb, J., Montesinos, B., & et al. 2014, ApJ, 790, 21Mesa, D., Gratton, R., Zurlo, A., et al. 2015, A&A, 576, A121Mordasini, C., Mollière, P., Dittkrist, K.-M., Jin, S., & Alibert, Y. 2015, Interna-tional Journal of Astrobiology, 14, 201Mulders, G. D., Min, M., Dominik, C., Debes, J. H., & Schneider, G. 2013,A&A, 549, A112Muto, T., Grady, C. A., Hashimoto, J., & et al. 2012, ApJ, 748, L22Osorio, M., Anglada, G., Carrasco-González, C., et al. 2014, ApJ, 791, L36Pavlov, A., Möller-Nilsson, O., Feldt, M., & et al. 2008, in Proc. SPIE, Vol. 7019,Advanced Software and Control for Astronomy II, 701939Pecaut, M. J. & Mamajek, E. E. 2013, ApJS, 208, 9Perez, S., Casassus, S., Ménard, F., et al. 2015, ApJ, 798, 85Petit, C., Sauvage, J.-F., Fusco, T., & et al. 2014, in Proc. SPIE, Vol. 9148, Adap-tive Optics Systems IV, 91480OPrice, D. J., Cuello, N., Pinte, C., et al. 2018, MNRAS, 477, 1270Quanz, S. P., Amara, A., Meyer, M. R., et al. 2015, ApJ, 807, 64Quanz, S. P., Amara, A., Meyer, M. R., et al. 2013, ApJ, 766, L1Rameau, J., Chauvin, G., Lagrange, A.-M., et al. 2012, A&A, 546, A24Reggiani, M., Christiaens, V., Absil, O., et al. 2017, ArXiv e-prints[ arXiv:1710.11393 ]Reggiani, M., Quanz, S. P., Meyer, M. R., et al. 2014, ApJ, 792, L23Rodigas, T. J., Follette, K. B., Weinberger, A., Close, L., & Hines, D. C. 2014,ApJ, 791, L37Sallum, S., Follette, K. B., Eisner, J. A., et al. 2015, Nature, 527, 342Salyk, C., Herczeg, G. J., Brown, J. M., & et al. 2013, ApJ, 769, 21Sissa, E., Gratton, R., Garufi, A., et al. 2018, ArXiv e-prints[ arXiv:1809.01001 ]Sivaramakrishnan, A. & Oppenheimer, B. R. 2006, ApJ, 647, 620Soummer, R. 2005, ApJ, 618, L161Soummer, R., Pueyo, L., & Larkin, J. 2012, ApJ, 755, L28Strom, K. M., Strom, S. E., Edwards, S., Cabrit, S., & Skrutskie, M. F. 1989, AJ,97, 1451Thalmann, C., Schmid, H. M., Boccaletti, A., & et al. 2008, in Proc. SPIE, Vol.7014, Ground-based and Airborne Instrumentation for Astronomy II, 70143FThiébaut, E. 2008, in Proc. SPIE, Vol. 7013, Optical and Infrared Interferometry,70131ITuthill, P., Lacour, S., Amico, P., & et al. 2010, in Proc. SPIE, Vol. 7735, Ground-based and Airborne Instrumentation for Astronomy III, 77351OTuthill, P. G., Monnier, J. D., Danchi, W. C., Wishnow, E. H., & Hani ff , C. A.2000, PASP, 112, 555van Boekel, R., Min, M., Leinert, C., & et al. 2004, Nature, 432, 479Verhoe ff , A. P., Min, M., Pantin, E., et al. 2011, A&A, 528, A91Vigan, A., Moutou, C., Langlois, M., & et al. 2010, MNRAS, 407, 71Waelkens, C., Waters, L. B. F. M., de Graauw, M. S., & et al. 1996, A&A, 315,L245Weingartner, J. C. & Draine, B. T. 2001, ApJ, 548, 296Williams, J. P. & Cieza, L. A. 2011, ARA&A, 49, 67Zacharias, N., Finch, C., & Frouard, J. 2017, AJ, 153, 166Zhu, Z., Nelson, R. P., Hartmann, L., & et al. 2011, ApJ, 729, 47Zurlo, A., Vigan, A., Galicher, R., et al. 2016, A&A, 587, A57Zurlo, A., Vigan, A., Mesa, D., et al. 2014, A&A, 572, A85 INAF – Osservatorio Astronomico di Padova e-mail: [email protected] Max Planck Institute for Astronomy, Königstuhl 17, D–69117 Hei-delberg, Germany INCT, Universidad De Atacama, calle Copayapu 485, Copiapó, At-acama, Chile Geneva Observatory, University of Geneva, Chemin des Maillettes51, 1290 Versoix, Switzerland SUPA, Institute for Astronomy, The University of Edinburgh, RoyalObservatory, Blackford Hill, Edinburgh, EH9 3HJ, UK Núcleo de Astronomia, Facultad de Ingeniería, Universidad DiegoPortales, Av. Ejercito, Santiago, Chile Escuela de Ingeniería Industrial, Facultad de Ingeniería y Ciencias,Universidad Diego Portales, Av. Ejercito 441, Santiago, Chile Aix Marseille Université, CNRS, LAM (Laboratoired’Astrophysique de Marseille) UMR 7326, 13388 Marseille,France ETH Zurich, Institute for Astronomy, Wolfgang-Pauli-Str. 27, CH-8093, Zurich, Switzerland Departamento de Astronomía, Universidad de Chile, Casilla 36-D,Santiago, Chile Millennium Nucleus ”Protoplanetary Disk”, Departamento de As-tronomía, Universidad de Chile, Casilla 36-D, Santiago, Chile LESIA, Observatoire de Paris, PSL Research University, CNRS,Sorbonne Universités, UPMC Univ. Paris 06, Univ. Paris Diderot,Sorbonne Paris Cité, 5 place Jules Janssen, 92195 Meudon, France Univ. Grenoble Alpes, CNRS, IPAG, F-38000, Grenoble, France INAF - Osservatorio Astronomico di Capodimonte, SalitaMoiariello 16, 80131 Napoli, Italy Unidad Mixta Internacional Franco–Chilena de Astronomia,CNRS / INSU UMI 3386 and Departemento de Astronomía, Univer-sidad de Chile, Casilla 36–D, Santiago, Chile Space Telescope Science Institute, 3700 San Martin Drive, Balti-more, MD 21218, USA European Southern Observatory (ESO), Alonso de Córdova 3107,Vitacura, Casilla 19001, Santiago, Chile Department of Astronomy, Stockholm University, AlbaNova Uni-versity Center, 106 91 Stockholm, Sweden Steward Observatory, The University of Arizona, Tucson, AZ 85721 CRAL, UMR 5574, CNRS, Universit de Lyon, Ecole Normale Su-perior de Lyon, 46 Alle d’Italie, F–69364 Lyon Cedex 07, France INAF-Catania Astrophysical Observatory, via S. Sofia, 78, 95123,Catania, Italy York Creek Observatory, Georgetown, 7253, Tasmania, Australia Department of Astronomy, University of Michigan, 1085 S. Univer-sity, Ann Arbor, MI 48109 Instituto de Física y Astronomía, Facultad de Ciencias, Universi-dad de Valparaíso, Av. Gran Bretaña 1111, Playa Ancha, Valparaíso,Chile Núcleo Milenio Formación Planetaria - NPF, Universidad de Val-paraíso, Av. Gran Bretaña 1111, Valparaíso, Chile Monash Centre for Astrophysics (MoCA) and School of Physivs andAstronomy Monash University, Clayton Vic 3800, Australia Xi’an Jiaotong–Liverpool University, department of MathematicalSciences, 111 Ren?ai road, Suzhou Dushu Lake Higher Educationtown,Jiangsu Province, China, cp 21512328