arXiv:astro-ph/0406108v1 4 Jun 2004 8C08ad70.C-0116(A) and 68.C-008 jena.de n eaieyde sescin2,w r bet eetwid detect to able are we go- 2), and section images (see field deep wide relatively relatively ing using By fi (FOV). small their view of of because searches those to accessible not AU, ris rtidcto fsc osbeiflec ol be could influence di possible apparent such of pl indication of first evolution A and orbits. stability the on prov multip and stellar planet-formation could of on implication they possible because the about interest, The hints some systems. particular stellar of multiple are in planets found were Som far. planets so these discovered of were planets extrasolar 100 than More Introduction 1. eetn eieo opnos ihsprto pto up separation i with an companions, however, us of out, regime leave teresting searches host (sub)stella Those planet optics. close (RV) adaptive radial-velocity ing for the searched of already companions have eccentricit days. low groups 40 about very Several than a shorter is have period to orbiting their when tend planets that systems mul- out stellar Furthermore in pointed multiple planets (2002)). (2004) of Mazeh al. that & et and Eggenberger (Zucker systems one stellar only tiple with systems in ets edo Send ⋆ DI ilb netdb adlater) hand by inserted be will (DOI: srnm Astrophysics & Astronomy o-asselrcmaino h lnths trH 75289 HD star host planet the of companion stellar low- A ae nosrain bando aSlai S programs ESO in Silla La on obtained observations on Based ueotadtoa tla opnosbyn 4 Uadsub and AU 140 beyond H companions to stellar down additional c companion out are (sub)stellar rule companion further the No confirme of 75289. later HD colors months of and 10 spectrum taken infrared image The SofI star. second A 75289. HD of 1 e words. Key AU. 2000 to 3 2 4 Abstract. ??? Accepted ; ??? Received .8mNTtreyasltr edtce omvn compani co-moving a detected we later, ima an three Comparing NTT 2000). m al. 3.58 et (Udry planet radial-velocity ff rn eussto requests print srpyiaice nttt nvri¨tJn,Schill Universit¨at Jena, Institut, Astrophysikalisches uoenSuhr bevtr,Karl-Schwarzschild-Str. Observatory, Southern European Israel 69978, Aviv Tel University, Aviv Tel hu igrLnesenat atnug trwre5 07 5, Sternwarte Tautenburg, Th¨uringer Landessternwarte ff rnebtentems-eidrlto o plan- for relation mass-period the between erence erpr ntedtcino e o-asselrcompanio stellar low-mass new a of detection the on report We tr:o-as lntr systems planetary Stars:low-mass, aksMgae,[email protected] Mugrauer, Markus : .Mugrauer M. aucitn.text no. manuscript 1 .Neuh¨auser R. , ∼ r¨ ßhn23 74 ea Germany Jena, 07745 erg¨aßchen 2-3, 1 licity 1000 .Mazeh T. , anet eld the ide n- se in 7 atnug Germany Tautenburg, 778 ,878Grhn,Germany Garching, 85748 2, y e e = - r eo MS iha mg eotie ihSf tteESO the at SofI with obtained we image an with 2MASS of ge 9mgcudb eetd ihteSf eeto ii ecan we limit detection SofI the With detected. be could mag 19 tla opnoswt assm with companions stellar nitn iha 2t 5mi-eunesa ttedistance the at star main-sequence M5 to M2 an with onsistent evn taeyta visstrto ls otehs st host the ob to an close by saturation achieved avoids is that com- sensitivity strategy planetary serving The wide even detected. Myrs, be hundred can to panions tens few a only of h otensml.W ol eetcmo rprmotion proper common detect could We sample. northern the e Our companions. wide larg of detection relatively the using ( by detectors and IR array companions) close of detection tr like stars fr separation projected with ∼ companions to sensitive are we nHwi nrhr ape n h .8mEONTT ESO m 3.58 the and UKIRT m sample) 3.8 the (northern with stars Hawaii target our on of obtai most have for we image far, of first So a planets. companions giant (sub)stellar harbor to wide known stars unknown all for pr observing search an to started have gram we 2000, of end the at Therefore, essniiealsyI uvy ie2MASS like th surveys by IR far all-sky so sensitive found less be not could which companions (sub)stellar eaaindw otesen ii ( limit seeing the to down separation a hl suhr ape.I otcsstesniiiyo th of sensitivity the su is cases cameras most IR In sample). (southern Chile nlctd21.465 located on 0 Uu osvrl10 U o on Vpae host planet RV young For AU. 1000 several to up AU 100 4 3 2 1 h omnpoe oino D729Bwt t host its with B 75289 HD of motion proper common the d T:NwTcnlg Telescope Technology New NTT: Telescope Sky InfraRed Southern Kingdom the United UKIRT: of Survey Infrared Near Deep DENIS: Survey Sky All Micron 2 2MASS: 2 ff .Alves J. , r led ile n e srmti ofimto in confirmation astrometric new one yielded already ort fH 58,aGVsa hthrosoeknown one harbors that star G0V a 75289, HD of n ι o H 75)or 17051) (HD Hor 3 n .Guenther E. and , ffi ± .2 rsc (621 arcsecs 0.023 in odtc uselrcmain with companions substellar detect to cient → ag O fmr hn10aces for arcsecs) 100 than more of FOV large ≥ 0.050 ǫ r H 24) iha age an with 22049), (HD Eri ± 4 M 0A t2 c east pc) 29 at AU 10 ⊙ ∼ 1 rm40A up AU 400 from ′′ .Ti mle that implies This ). 1 rDENIS or uut7 2018 7, August 2 r( ar . ⋆ 4 ned om → o- in e e e 3 - 2 M. Mugrauer et al.: A low-mass stellar companion of the planet host star HD 75289 of a companion of the star HD89744 (Mugrauer et al. 2004 AN submitted), suggested by Wilson et al. 2001. The compan- ion is separated by about 2500AU from its host star, with an effective temperature (Te f f ) about 2200K and a mass between 0.072 and 0.081 M⊙, depending on the evolutionary model and the assumed age. HD89744B is either a very low mass stellar or a heavy companion to a RV planet host star. In this paper we report astrometric and spectroscopic evidence for a new stellar companion found in our southern survey around the G0V star HD75289, for which Udry et al. (2000) found a planet with msini = 0.42 MJup in a 3.51 orbit.

2. Imaging, Data Reduction and Calibration

Our own observations of HD75289 were obtained in the H band (1.6 µm) with the 3.58m ESO NTT. This telescope is equipped with active optics which dramatically reduced dome and telescope seeing, yielding images with the seeing limit of the atmosphere. The IR detector is SofI 5, a 1024x1024 HgTeCd detector with 18 µm pixel and a pixel scale of ap- proximately 0.144arcsecs in the so-called small field mode (147arcsecs FOV). To reduce saturation by the bright primary we chose individual integration time to be as short as possible (1.2 s). To reach high sensitivity (i.e. a high limiting magnitude for the detection of faint companions),the total integration time was around 10min, composed of many short integrated images. The auto-jitter technique of the NTT telescope was applied to delete the IR sky background from each raw frame. The data- reduction is donewith the ESO pipeline ECLIPSE 6. All images were flat fielded with a special dome flat image, provided by the NTT science team. At 1arcsec seeing the detection limit (S/N = 3) is 19mag in H for 10min of total integration. Fig. 1. H band images of HD75289 (central bright star) from For calibration we identify 2MASS objects also detected on 2MASS (02/99) at the top and our first NTT/SofI image our NTT images. We use the coordinates of those objects from (01/02). The total integration time is 10min. The co-moving the 2MASS point source catalog to determine the NTT pixel companion is located 21.5arcsecs east to HD75289 and is also scale. We do so on each NTT image and obtain the mean pixel visible in the 2MASS image (marked object). The stars R1 and scale for each NTT run. The averaged pixel scale of all runs R2 are used in Fig.4 as comparison stars. is 143.66±0.15mas (only 0.1% relative uncertainty). With the pixel scale for each run, we can determine the positional dif- ference (separation) between any two stars, for the 1st, and 2nd 3. epoch. The separations between non-moving background stars In our search of wide (sub)stellarcompanionswe have to exam- do not change with time. Using those non-moving background ine hundreds of faint objects close to the RV planet host stars. stars, we can then determine the of stars moving Most of those objects will emerge as ordinary background thru the field. The precision of this method depends on the pre- stars, randomly located close to, but far behind the target stars. cision of gaussian centering per star and on the number of the On the other hand, bound companions share the proper motion ∼ stars used. One can achieve 1/100 of a pixel with gaussian of the host stars. This is so because the orbital motions of wide centering and special astrometric care (e.g. Pravdo&Shaklan companions with separations ≥ 100AU are small compared to ∼ (1996)). In our study, we have achieved a precision of 1/10 of their much faster common proper motions. An astrometric sur- ∼ a pixel ( 20mas), well enough for measuring the proper mo- vey will find these co-moving companions with only 2 images tions of our relatively nearby target stars and their co-moving taken with some epoch difference, depending on the astromet- companions. ric accuracy and the proper motion of the primary stars. Hence astrometry is a very effective tool for companion searches. 5 SofI: Son of Isaac In a first step of our study of HD75289 we compared our first 6 ECLIPSE: ESO C Library for an Image Processing Software epoch NTT image with the 2MASS one. The proper motion of Environment the brightenoughobjects can be derivedby comparingtheir po- M. Mugrauer et al.: A low-mass stellar companion of the planet host star HD 75289 3 sition in the 2MASS and the NTT images. The 2MASS images 31.7 are accurate enough for the detection of co-moving compan- 31.6 31.5 ions, as the proper motion of HD75289, is large enough. 31.4 31.3 The proper motion of most stars, as derived from the NTT(12/02) 2MASS/NTT astrometry and the given epoch difference of 31.2 31.1 2.9yr, were very small. Only one star had large proper motion, 31.0 NTT(01/02) ± ± µα = 1 24mas and µδ = -236 22mas per annum, consistent 30.9 with the well known Hipparcos proper motion of HD75289 30.8 (µα = -20.50 ± 0.49mas/yr and µδ = -227.68±0.44mas/yr). It is 30.7 30.6 clear that this star is a co-moving companion of the RV planet [arcsecs] separation host star. We therefore denote this star as HD75289B. 30.5 30.4 2MASS However, the 2MASS limit is approximately 2.5 magnitudes 30.3 brighter than the NTT limit, hence the motion of the faint com- 0.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0 4.5 5.0 panion candidates can be investigated only with a second epoch time[yr] NTT observation (see 2MASS and SofI/NTT image in Fig.1). 273.6 273.4 273.2 0.3 273.0 2MASS 272.8 0.2 272.6 272.4

0.1 PA [°] 272.2

0.0 272.0 NTT(01/02) 271.8 271.6 NTT(12/02) MDC[''] PM_DEC -0.1 271.4 -0.5 0.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0 -0.2 time[yr]

-0.3 Fig. 3. Proper motion of HD75289B for all three epochs. We -0.3 -0.2 -0.1 0.0 0.1 0.2 0.3 measure the distance betweenR1 andHD75289B (top)andthe PM_RA [''] position angle PA of HD75289B measured from R2. See Fig.1 Fig. 2. Result of the astrometry obtained by comparing two for the stars R1 & R2. Due to the motion of HD75289B rel- NTT images from epochs 01/02 and 12/02. The formal proper ative to the reference stars both values are changing following motion of all detected objects around HD75289A are shown the predicted curves (straight lines) for a co-moving compan- in the diagram. All of them have negligible proper mo- ion to HD752898A. The astrometric uncertainty is illustrated tions, similar in size as the astrometric uncertainty (∼20mas) with dotted lines. hence they are very slowly moving background objects. Only HD75289B (bottom with error bars) shares the proper mo- tion of HD75289A (square) which is well known from the Hipparcos astrometry. The proper motion of the bright primary 4. Photometry star A is not measured in our NTT images because of satura- Table1 gives the of HD75289A&B in tion. JHKS derived by 2MASS, together with our derivation of HSofi=11.224±0.040.Our result is consistent with the 2MASS photometry. From the known spectral type of HD75289A Due to the large number of stars in the NTT FOV (see Fig.1) and hence, its expected intrinsic B-V color (B-V=0.58 from several non-moving background stars are detected and the Kenyon & Hartmann (1995)) and its published B-V color (from proper motion can be determined with a precision in the or- Hipparcos B-V = 0.578±0.003mag), we find that interstellar ∼ der of 20mas (see Fig.2). Due to PSF saturation the proper absorption is negligible, as expected for nearby stars. With motion of HD75289A cannot be measured accurately in both J-K from 2MASS and color to temperature conversion from NTT images, but can be calculated for the given epoch differ- Kenyon & Hartmann (1995)we can derivea Te f f between 5800 ence with Hipparcos data of the , yearly proper to 6380K (J-K = 0.355±0.029) for the primary, which is con- motion and equatorial coordinates (square in Fig.2). sistent with its published spectral type G0V (Udry et al. 2000). In addition we illustrated the proper motion of HD75289B We obtained 3210 to 3860K (J-K = 0.907±0.047) for the com- over all three epochs with two reference stars R1 and R2 (see panion, hence spectral type between M0 and M5. We used the Fig.1 and Fig.3). 2MASS color transformations of Carpenter (2001) to convert 4 M. Mugrauer et al.: A low-mass stellar companion of the planet host star HD 75289

J-KS from 2MASS to J-K of Bessel & Brett which is similar to The companion and the primary star were both located on the Johnson. slit, and spectra of both objects were taken simultaneously. Te f f of HD75289A is well known, hence a black body function with the given T (6030K) can be used to determine the re- Table 1. Photometry for HD75289A&B. The 2MASS Point e f f sponse function of the spectrograph, which is needed to ob- Source catalog yield apparent JHK magnitudes which are con- S tain relative flux calibrated spectra of both objects. In Fig.4 we firmed in H with our SofI/NTT images. show the relative flux calibrated spectra of HD75289A&B. The continuum of the companion is much flatter than the pri- band mA mB J 5.346±0.019 11.750±0.036 mary continuum, consistent with a cooler . From a H 5.187±0.031 11.181±0.031 black body fit on the continuum of the HD75289B we deter- HSofi - 11.224±0.040 mine its Te f f to be in the range between 3250K and 3500K, KS 5.012±0.020 10.879±0.027 hence spectral type M2 to M5.

MgI OH AlI CO SiI M2+V

5. Spectroscopy HD75289B

M3V To confirm the spectral type of the companion we obtained IR normalized flux spectra of HD75289A&B in June 2003 with SofI in spectro- HI scopic mode. We used long slit spectroscopy with a slit-width 5600 5800 6000 6200 6400 6600 of one arcsec and the red grism covering the wavelength range wavenumber[cm-1 ] from 1.53 to 2.52 µm. The dispersion was 10.22Å per pixel M2+V with a IR HgCdTe detector in the large field mode (288mas 12 6-3 4-2 3-1 2-0 CO 13CO Ca/Fe Na g Al Mg pixel scale). The resolving power is λ/∆λ ≈ 588. M3V HBr Background subtraction was obtained by nodding between two positions along the slit, as well as by a small jitter around those HD75289B two positions, to avoid that individual pixels see always the ) ) ) same part of the sky. Eighteen individual spectra, each with )

normalizedflux an integration time of 30s, were averaged, i.e. a total integra- M6V ) tion time of 9min. All images were flat fielded with a standard dome flat and wavelength calibrated with a Xe lamp. We used standard IRAF routines for background subtraction, flat field- 4200 4400 4600 4800 ing and averaging all individual spectra. wavenumber[cm-1 ]

Fig. 5. Normalized H and K band spectra of HD75289B, com- pared with spectra of GJ411 (M2+V), GJ725 (M3V) and Wolf359 (M6V) from Meyer et al.(1997/98). HD75289A

Fig.5 shows the normalized H and K band spectra of HD 75289 B. The most striking -sensitive feature in ′ the H band is the second-overtone CO band head [ν, ν ] at HD75289B 6177cm−1, which is found in the spectra of K and M stars. The spectra of HD75289B shows CO molecular lines which are log(flux)+const. fainter than Mg (5844cm−1), typical for red dwarfs and coin- cident with the JHK absolute magnitudes derived in section4. Hydrogen line at 5950cm−1, which can be found only in stars earlier than K3, is not visible in the companion spectra and the Al feature at 5973cm−1 is strong as the Mg line. Both com-

1.6 1.8 2.0 2.2 2.4 parisons serve as evidence for a spectral type cooler than M1V. −1 −1 wavelenght[µm] The Si line at 6264cm is not apparent and Si at 6292cm is faint comparable with the OH (∆ν = 2) molecular features at Fig. 4. Relative flux of HD75289A&B. The drop at 1.85 µm 5920cm−1, typical for a spectral type M3V. is due to strong water absorption as well as some telluric lines. The strongest apparent lines in the companion spectrum in K From a black body fit, we derive a Te f f of the companion be- are from molecular bands of the first CO overtone extending tween 3250K and 3500K, coincident with a spectral type M2 from 4360cm−1 to the low frequency side of the spectrum. In to M5. addition, 13CO at 4260cm−1 and H Br γ at 6297cm−1 are not M. Mugrauer et al.: A low-mass stellar companion of the planet host star HD 75289 5

2 apparent, all argue for a dwarf cooler than K2V. The CO bands 1.1 are a bit stronger than the Ca/Fe (4415cm−1) as well as Na HD75289A (4530cm−1) atomic features and Ca/Fe is weaker than Na. The 1.0 −1 −1 4 0.9 Al line at 6720cm is faint but the Mg line at 4750cm is 0.8 missing, typical for spectral types cooler than M3V. 0.7 0.6 The detected features in the spectrum of HD75289B in H and 6 K, the black body fit of the continuum,and the companionJHK 0.4 [mag]

colors, are all consistent with a spectral type of M2V to M5V, H M i.e. Te f f is in the range of 3240 to 3580K. 8 0.2

HD75289B

0.1 6. Discussion 10 0.08 HD75289 is a bright G0 dwarf (V=6.36mag) located at a dis- 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.0 ± ± tance of 28.94 0.47pc (distance module 2.308 0.036mag). J-K [mag] Its apparent JHK colors are typical for a G0V star at the given distance. Thus, the super-giant classification given in Simbad Fig. 6. The color-magnitude diagram with the isochrone for is invalidated, as pointed out by Udry et al. (2000). The same 5Gyrs from Baraffe et. al (1998) with [M/H]=0, mixing length group discovered an extrasolar planet with a minimum mass parameter α=1 and He abundance Y=0.25. The primary and of 0.42 MJup which revolves around its parent star in a nearly its companion are included in the diagram with their uncertain- circular orbit (e=0.054, a=0.046 AU). ties in magnitude and color. Masses are indicated as numbers HD75289B is clearly co-moving with HD75289A and the in solar masses. color-magnitude relation agrees with the assumption that both objects are at the same distance (Fig.6). With Baraffe et al. (1998) models, the JHK colors from Sect.3 and the given dis- tance module, we can derive the mass of HD75289B to be 0.135±0.003M⊙, see Fig.6. The system age is roughly 5Gyrs (Udry et al. 2000). Note that the age uncertainty of the pri- mary does not play an important role in the derivation of the mass of HD75289B, because the IR magnitudes for such low- mass stellar objects decrease very slowly from 1 to 10Gyrs. The given uncertainty of the companion mass is derived only form the magnitude errors. Inaccuracies of the used theoretical model were not considered here. With the derived companion mass (0.135 M⊙), the pri- mary mass (∼1M⊙) and the companion separation 621AU (21.465±0.023arcsecs) we can compute the expected RV vari- ation of the primary induced by the presence of the wide com- ∼ ∼ panion v 150m/s with an of 15000 years. Fig. 7. The limiting H magnitude versus separation from Although this is a large effect, the maximal yearly variation HD75289A for our NTT image shown in Fig.1 and 2MASS. ∼ of the RV is only 0.07m/s, too small to be detected in the The corresponding projected separation in AU is shown on the foreseen future. upper x-axis. Saturation occurs within 1.5arcsecs (43AU) (see Fig.7 shows the NTT detection limit which is 19mag in H vertical dashed line) hence a companion search is impossible and enables the detection of substellar companions down to there. The detection of all stellar companionsis feasible beyond MH=16.7mag around HD75289A (m≥44MJup according to 4.7arcsecs (136AU). The right y-scale shows the predicted ab- Baraffe et al. 2003). Objects up to 68arcsecs were observed solute H magnitudes for substellar objects from Baraffe et al. twice but no further co-moving companion could be identified. (2003) models for an age of 5Gyrs. The 3σ detection limit is Further stellar companions (m≥75MJup) can be ruled out for a 19mag in H, hence substellar companions down to 0.050 M⊙ projected separation from 136AU up to 1968AU. can be found beyond 15arcsecs (434AU).

Acknowledgements. We would like to thank the technical staff of the References ESO NTT for all their help and assistance in carrying out the observa- tions. Furthermore, we would like to thank M. Fern´andez, A. Seifahrt, Baraffe I., Chabrier G., Allard F., 1998, A&A 337, 403 A. Szameit and C. Broeg who have carried out some of the obser- Baraffe I., Chabrier G., Barman T. S., 2003, A&A, 402, 701 vations of this project. We made use of the 2MASS public data re- Carpenter J.M., 2001, AJ, 121, 2851 leases as well as the Simbad database operated at the Observatoire Eggenberger A., Udry S., Mayor M., 2004, A&A, 417, 353 Strasbourg. This work was partly supported by the Israel Science Gonzales G., Laws C., 2000, AJ, 119, 390 Foundation (grant no. 233/03) Kenyon S.J., Hartmann L.W., 1995, ApJS, 101, 117 6 M. Mugrauer et al.: A low-mass stellar companion of the planet host star HD 75289

Meyer M. R., Edwards S., Hinkle K. H., 1998, ApJ, 508, 397 Pravdo Steven H., Shaklan Stuart B., 1996, ApJ, 465, 264 Udry S., Mayor M., Naef D., 2000, A&A, 356, 590 Wallace L., Hinkle K., 1997, ApJS, 111, 445 Wilson, J.C., Kirkpatrick, J. D., Gizis, J. E., 2001, AJ, 122, 1989 Zucker. S., Mazeh T., 2002, ApJ, 568, 113