Korean-Japanese Planet Search Program: Substellar Companions around Intermediate-Mass Giants
Masashi Omiya, Inwoo Han, Hideyuki Izumiura, Byeong-Cheol Lee, Bun'ei Sato, Kang-Min Kim, Tae Seog Yoon, Eiji Kambe, Michitoshi Yoshida, Seiji Masuda, Eri Toyota, Seitaro Urakawa, Masahide Takada-Hidai
aa r X i v : . [ a s t r o - ph . E P ] J a n Korean-Japanese Planet Search Program:Substellar Companions aroundIntermediate-Mass Giants
Masashi Omiya ∗ , Inwoo Han ∗ , Hideyuki Izumiura †, ∗∗ , Byeong-Cheol Lee ∗ ,Bun’ei Sato ‡ , Kang-Min Kim ∗ , Tae Seog Yoon § , Eiji Kambe † , MichitoshiYoshida ¶ , Seiji Masuda k , Eri Toyota †† , Seitaro Urakawa ‡‡ and MasahideTakada-Hidai §§ ∗ Korea Astronomy and Space Science Institute, 61-1 Hwaam-dong, Yuseong-gu, Daejeon 305-348,South Korea, email: [email protected] † Okayama Astrophysical Observatory, National Astronomical Observatory of Japan, Asakuchi,Okayama 719-0232, Japan ∗∗ Department of Astronomical Science, The Graduate University for Advanced Studies, ShonanVillage, Hayama, Kanagawa 240-0193, Japan ‡ Tokyo Institute of Technology, 2-12-1-S6-6 Ookayama, Meguro-ku, Tokyo 152-8550, Japan § Department of Astronomy and Atmospheric Sciences, Kyungpook National University, Daegu702-701, South Korea ¶ Hiroshima Astrophysical Science Center, Hiroshima University, Higashi-Hiroshima, Hiroshima739-8526, Japan k Tokushima Science Museum, Asutamu Land Tokushima, Itano-gun, Tokushima 779-0111, Japan †† Kobe Science Museum, 7-7-6 Minatojimanakamachi, Chuo-ku, Kobe, Hyogo 650-0046, Japan ‡‡ Bisei Spaceguard Center, Japan Spaceguard Association, 1716-3 Okura, Bisei-cho, Ibara,Okayama 714-1411, Japan §§ Liberal Arts Education Center, Tokai University, 1117 Kitakaname, Hiratsuka, Kanagawa259-1292, Japan
Abstract.
A Korean-Japanese planet search program has been carried out using the 1.8m telescopeat Bohyunsan Optical Astronomy Observatory (BOAO) in Korea, and the 1.88m telescope atOkayama Astrophysical Observatory (OAO) in Japan to search for planets around intermediate-mass giant stars. The program aims to show the properties of planetary systems around such starsby precise Doppler survey of about 190 G or K type giants together with collaborative surveys of theEast-Asian Planet Search Network. So far, we detected two substellar companions around massiveintermediate-mass giants in the Korean-Japanese planet search program. One is a brown dwarf-masscompanion with 37.6 M J orbiting a giant HD 119445 with 3.9 M ⊙ , which is the most massive browndwarf companion among those found around intermediate-mass giants. The other is a planetarycompanion with 1.8 M J orbiting a giant star with 2.4 M ⊙ , which is the lowest-mass planetarycompanion among those detected around giant stars with > M ⊙ . Plotting these systems oncompanion mass vs. stellar mass diagram, there seem to exist two unpopulated regions of substellarcompanions around giants with 1.5–3 M ⊙ and planetary companions orbiting giants with 2.4–4 M ⊙ .The existence of these possible unpopulated regions supports a current characteristic view that moremassive substellar companions tend to exist around more massive stars. Keywords: stars: planetary systems, brown dwarfs, techniques: radial velocities
PACS:
NTRODUCTION
Many of the over 380 exoplanets discovered to date orbit solar mass (0.7–1.5 M ⊙ )stars. This observational bias occurs because the main targets of previous Dopplerspectroscopy-based exoplanet searches have been solar-type stars. These studies haverevealed a variety of their planetary systems (e.g. Butler et al. 2; Udry & Santos 24)whose statistical properties are now used to constrain planet-formation models (e.g. Ida& Lin 6). However, only about 35 and 25 planets were detected around intermediate-mass (1.5–5 M ⊙ ) and low-mass ( < M ⊙ ) stars in surveys of evolved G-K type (sub)giants and K-M type dwarfs, respectively, so far. The planetary systems around such starsare not well understood. Clarifying the relationship between stellar mass and planetarysystem is important for understanding general planet formation because planet formationshould depend on the properties of the protoplanetary disks, which be affected by thecondition of host stars.Evolved intermediate-mass (sub) giant stars are suitable targets for Dopplerspectroscopy-based planet searches because these stars have low surface activityand their spectra exhibit many sharp absorption lines. Although the number of substellarcompanions found orbiting such stars is still small, some planetary system propertieshave begun to emerge. For example, the masses of planets and their host stars arecorrelated: more massive substellar companions tend to exist around more massive stars(e.g. Lovis & Mayor 17). This correlation suggests that the mass ranges of the browndwarf desert depend on the host star mass [18]. The planet occurrence rate also dependson the host star mass: the planet frequency around 1.5–2 M ⊙ giant stars is higher thanthat around lower-mass stars [9]. Moreover, the semi-major axes of planetary systemsseem to be correlated with host star: the range for planets orbiting intermediate-massgiant stars exceeds 0.6 AU , and the range for planets around solar-type stars is largethan 0.02 AU (e.g. Sato et al. 22). These properties of substellar systems aroundintermediate-mass giant stars seem to be not similar to those around solar-type stars. Aunified understanding of planetary systems over a wide range of host star masses willprovide valuable insights into the dependence of substellar systems on the central starsand into general plant formation mechanisms. KOREAN-JAPANESE PLANET SEARCH PROGRAM
In 2005, we started a joint planet search program between Korean and Japanese re-searchers to search for planets around GK-type giant stars using a precise Doppler tech-nique with using the 1.8-m telescope at BOAO and the 1.88-m telescope at OAO [18].This survey program is an extended version of the ongoing OAO planet search program[21] and part of an international collaboration among researchers from Korea, China andJapan (an East-Asian Planet Search Network, EAPS-Net; Izumiura 8). The collaborationaims at clarifying the properties of planetary systems around intermediate-mass stars by Recently, a close-in planet with a semi-major axis of 0.081 AU was found orbiting an intermediate-masssubgiant with mass of 1.68 M ⊙ [10]. urveying more than 800 GK giants for planets at OAO, BOAO, the Xinglong station(China), and the Subaru Telescope.For the Korean-Japanese planet search program, we selected about 190 target starsfrom the Hipparcos catalog based on the following criteria: color-index 0.6 < B − V < − < M v <
2, declination d > − ◦ , and visual magnitude 6.2 < V < BOES Observations and Analysis
Radial velocity observations at BOAO are carried out with the 1.8-m telescope andthe BOAO Echelle Spectrograph (BOES; Kim et al. 14), a fiber-fed high resolutionechelle spectrograph. We place an iodine (I ) cell in the optical path in front of the fiberentrance of the spectrograph [13] for precise wavelength calibration and use a 200- m mfiber, obtaining a wavelength resolution R = l / D l ∼ absorption lines, for radial velocity measurements. We alsomake use of Ca II H lines at around 3970 Å as chromospheric activity diagnostics. Radialvelocity analysis is performed using the spectral modeling technique described in Satoet al. [20], which is based on the method of Butler et al. [3], and improve and optimizefor BOES data analysis [18]. We employ the extraction method described in Sato et al.[20] to prepare a stellar template spectrum from stellar spectra taken through the I cell(star+I spectra). The technique allows us to achieve a long-term Doppler precision of14 m s − over 5.5 yr. HIDES Observations and Analysis
Radial velocity observations at OAO are carried out with the 1.88-m telescope andHIgh Dispersion Echelle Spectrograph (HIDES; Izumiura 7) attached to the Coudéfocus of the telescope. We use an I cell placed in the optical path in front of the slitof the spectrograph [11] as a precise wavelength calibrator. We set the slit width to 200 m m (0.76"), providing a spectral resolution of 63000. Until November 2007, we hadtaken star+I spectra with the 5000 Å to 6200 Å wavelength range, and had taken stellarspectra at the same wavelength range without the I cell for abundance analysis. Sincethe HIDES CCD system was upgraded to a three CCD mosaic in December 2007, weobtain simultaneous spectra in the full 3750 Å to 7550 Å wavelength range. We use the5000 Å to 5900 Å wavelength ranges of star+I spectra for radial velocity measurements.Echelle data reduction is performed using the IRAF software package in the standardmanner. Stellar radial velocities are derived from the I -superposed stellar spectrumsing the spectral modeling techniques detailed in Sato et al. [20], giving a Dopplerprecision of less than 8 m s − over 5.5 yr. TWO SUBSTELLAR COMPANIONS AROUND MASSIVEINTERMEDIATE-MASS GIANTS
Up to now, we have been monitoring the radial velocities of sample stars using 1.8-mBOAO telescope and 1.88-m OAO telescope for 5.5 yr and identified many candidatestars with large radial velocity variations. And then, two massive intermediate-massstars with a periodic variation caused by a substellar companion are found among them:a brown dwarf-mass companion orbiting HD119445 [18] and a planetary companionorbiting a giant star.We monitored the radial velocity of HD 119445 for 2.3 years from the beginning ofthe survey at both observatories. The observed radial velocities of HD 119445 are shownin figure 1. The best-fit Keplerian orbit derived from both the BOAO and OAO data hasa period P = ± K = ± − , andan eccentricity e = ± − . This values is comparable to the typical radial velocity scatter (10–20 m s − ) ofthe G-type giants [21]. Adopting a stellar mass M = 3.9 ± M ⊙ for HD 119445, weobtained a semi-major axis a = 1.71 ± M sin i = 37.6 ± M J for a brown dwarf-mass companion [18].A large radial velocity variation in the giant star with a planetary companion wasfound in the early BOAO radial velocity survey and we made intensive follow-upobservations of the star at BOAO and OAO. The observed radial velocities of the starare shown in figure 2. A best-fit Keplerian orbit derived from both the BOAO and OAOvelocity data by a least square fit has a period P = K = − , and an eccentricity e = − for BOAO and OAO data. This values is comparable to the typical radialvelocity scatter of the G-type giants [21]. Adopting a host star’s mass M = 2.4 (2.0–2.6) M ⊙ , which was estimated from evolutionary track and fundamental stellar parameters of L = 43 L ⊙ , T eff = 4861 K, and [Fe/H] = 0.15, we obtained a semi-major axis a = 0.77AU and a minimum mass M sin i = 1.7 M J for a planetary companion. DISCUSSION
Their host stars, HD119445 and the giant star, have masses of 3.9 M ⊙ and 2.4 M ⊙ ,respectively. The brown dwarf-mass companion orbiting HD 119445 is most massivesubstellar companions, and the planetary companion orbiting the giant star is lowest-mass companion, among those discovered around massive intermediate-mass (1.9–5 M ⊙ ) stars. In figure 1, we plot masses of the companions detected within semi-majoraxis of 3 AU by precise Doppler surveys against their host star’s masses; solar-massstars (0.7 M ⊙ ≤ M < M ⊙ , open triangles ), intermediate-mass subgiants and giants IGURE 1.
Upper panel: radial velocities of HD 119445 observed at BOAO ( filled circles ) and OAO( open circles ). The solid line represents the Keplerian orbital curve. Lower panel: Residuals to the bestKeplerian fit.
FIGURE 2.
Upper panel: radial velocities of a giant star observed at BOAO ( filled circles ) and OAO( open circles ). The solid line represents the Keplerian orbital curve. Lower panel: Residuals to the bestKeplerian fit. M ⊙ ≤ M ≤ M ⊙ , filled circles ), intermediate-mass dwarfs ( open circles ), and HD119445 and the giant star ( stars ) (e.g. The Extrasolar Planets Encyclopadia ; updatedversion of figure 5 in Omiya et al. 18; this work). The detectable companion mass for agiven host star mass depends on the orbital separation of its companion and the radialvelocity jitter of the host star. Assuming that typical radial velocity jitters s for solar-mass stars, intermediate-mass subgiants (1.5–1.9 M ⊙ ) and giants (1.9–5 M ⊙ ) are ∼ − , ∼ − and ∼
20 m s − , respectively, we estimated the lower limits of companionmasses detectable by precise Doppler surveys around a solar-mass star and intermediate-mass subgiant and giant at 3 AU (sold lines in figure 1), corresponding to companionmasses that provide the amplitude of three times of typical radial velocity jitters. Wealso indicate detectable masses for these stars at 0.02 AU ( dotted lines ) and 0.6 AU ( dot-dashed lines ), corresponding to the semi-major axes of the known innermost planetsorbiting solar-type and intermediate-mass evolved stars.Two unpopulated regions of substellar companions orbiting intermediate-mass sub-giants and giants seem to exist in region (a) and (b) . A possible host star-companionmass correlation considered from unpopulated region (a) and (b) supports the currentview that more massive substellar companions tend to exist around more massive stars,that are derived from the results of planet searches around various mass stars (Lovis &Mayor 17; Hekker et al. 5).All of the brown dwarf-mass companions to intermediate-mass evolved stars werefound around those with ≥ M ⊙ and there seems to be a paucity of such companionsaround those with 1.5–2.7 M ⊙ (region (a) in figure 1). Considering the smaller numberof survey targets of ≥ M ⊙ (e.g., 35 % of the 300 OAO targets; Takeda et al. 23)compared with that of 1.5–2.7 M ⊙ , frequency of brown dwarf companions may becomehigher as stellar mass increases. This might favor gravitational instability in protostellardisks [19] rather than fragmentation of proto-stellar clouds [1] as the formation mech-anism of brown dwarf-mass companions because stellar systems with larger differencein mass between primary and secondary stars are more difficult to form by the lattermechanism [1].Also, there seems to be a possible paucity of lower-mass companions around2.4 ∼ M ⊙ stars, in particular, a lack of planetary companions around 3–4 M ⊙ stars(region (b) in figure 1). Although it is basically difficult to detect planets around such"noisy" stars with large intrinsic radial velocity variability ( s ∼
20 m s − ), planets withmass ≥ M J ( ≥ M J ) and a = a = s ∼
60 m s − ). Kennedy & Kenyon [12] predicted thatthe frequency of giant planets has a peak near 3 M ⊙ stars based on a core accretionscenario taking account of the movement of snow line along the evolution of accretionand the central stars. Moreover, if a formation mechanism works that invokes capturingof solid bodies migrating inward at the inner edge of the inactive magnetorotationalinstability-dead zone inside of the protoplanetary disk, gas giant planets could beformed efficiently at around 1 AU around intermediate-mass stars before the planetary We exclude a brown dwarf-mass companion orbiting a possible high mass giant HD 13189 ( M = ± M ⊙ ; Hatzes et al. 4) from the following discussion because of the large uncertainty in its host star’smass. IGURE 3.
Primary star masses versus masses of substellar companions orbiting within 3 AU.
Opentriangles , filled circles and open circles represent companions orbiting solar-mass stars, intermediate-massevolved stars (subgiants and giants) and intermediate-mass dwarfs, respectively. Stars represent the HD119445 and the giant star. Solid lines indicate the detection limits for the mass of companions orbiting at3 AU, corresponding to three times of typical radial velocity jitters s of 5 m s − for solar-mass stars (0.7 M ⊙ ≤ M < M ⊙ ), 7 m s − for intermediate-mass subgiants (1.5 M ⊙ ≤ M ≤ M ⊙ ) and 20 m s − for intermediate-mass giants (1.9 M ⊙ < M ≤ M ⊙ ). Dotted and dot-dashed lines indicate the detectionlimits for companions at 0.02 AU and 0.6 AU in solar mass stars and intermediate-mass evolved stars,respectively. Two regions devoid of substellar companions are denoted by (a) and (b). disks deplete [15]. Increasing the number of known massive planetary companionsaround massive intermediate-mass stars by further radial velocity surveys would beof great interest to understand the formation mechanisms of giant planets aroundintermediate-mass stars.
ACKNOWLEDGMENTS
This research was supported as a Korea-Japan Joint Research Project under the Japan-Korea Basic Scientific Cooperation Program between Korea Science and EngineeringFoundation (KOSEF) and Japan Society for the Promotion of Science (JSPS).
EFERENCES
1. Bate, M. R. 2000,
MNRAS , 314, 332. Butler, R. P., et al.
ApJ , 646, 5053. Butler, R. P., Marcy, G. W., Williams, E., McCarthy, C., Dosanjh, P., & Vogt, S. S. 1996,
PASP , 108,5004. Hatzes, A. P., Guenther, E. W., Endl, M., Cochran, W. D., Döllinger, M. P., & Bedalov, A. 2005,
A&A ,437, 7435. Hekker, S., Snellen, I. A. G., Aerts, C., Quirrenbach, A., Reffert, S., & Mitchell, D. S. 2008,
A&A ,480, 2156. Ida, S., & Lin, D. N. C. 2004,
ApJ , 604, 3887. Izumiura, H. 1999, in Proc. 4th East Asian Meeting on Astronomy, ed. P.S. Chen (Kunming: YunnanObservatory), 778. Izumiura, H. 2005, Journal of Korean Astronomical Society, 38, 819. Johnson, J. A., Aller, K. M., Howard, A. W., & Crepp, J. R. 2010a,
PASP , 122, 90510. Johnson, J. A., et al.
ApJL , 721, L15311. Kambe, E., et al.
PASJ , 54, 86512. Kennedy, G. M., & Kenyon, S. J. 2008,
ApJ , 673, 50213. Kim, K.-M., et al. et al.
PASP , 119, 105215. Kretke, K. A., Lin, D. N. C., Garaud, P., & Turner, N. J. 2009,
ApJ , 690, 40716. Liu, Y.-J., et al.
ApJ , 672, 55317. Lovis, C., & Mayor, M. 2007,
A§&A , 472, 65718. Omiya, M., et al.
PASJ , 61, 82519. Rice, W. K. M., Armitage, P. J., Bonnell, I. A., Bate, M. R., Jeffers, S. V., & Vine, S. G. 2003,
MNRAS ,346, L3620. Sato, B., Kambe, E., Takeda, Y., Izumiura, H., & Ando, H. 2002,
PASJ , 54, 87321. Sato, B., Kambe, E., Takeda, Y., Izumiura, H., Masuda, S., & Ando, H. 2005,
PASJ , 57, 9722. Sato, B., et al.
PASJ , 60, 53923. Takeda, Y., Sato, B., & Murata, D. 2008,