Skip to main content

Full text of "DTIC ADA492761: Common Proper Motion Companions to Nearby Stars: Ages and Evolution"

See other formats


The Astrophysical Journal, 687:566-578, 2008 November 1 

© 2008. The American Astronomical Society. All rights reserved. Printed in U.S.A. 


COMMON PROPER MOTION COMPANIONS TO NEARBY STARS: AGES AND EVOLUTION 

V. V. Makarov 

Michelson Science Center, Mail Stop 100-22, California Institute of Technology, 770 South Wilson Avenue, Pasadena, CA 91125 

AND 

N. Zacharias and G. S. Hennessy 

US Naval Observatory, 3450 Massachusetts Avenue, NW, Washington, DC 20392-5420; wm@caltech.edu 
Received 2008 March 13; accepted 2008 July 1 


ABSTRACT 

A set of 41 nearby stars (closer than 25 pc) is investigated which have very wide binary and common proper motion 
(CPM) companions at projected separations between 1000 and 200,000 AU. These companions are identified by astro- 
metric positions and proper motions from the NOMAD catalog. Based mainly on measures of chromospheric and X-ray 
activity, age estimation is obtained for most of 85 identified companions. Color-absolute magnitude diagrams are 
constructed to test whether CPM companions are physically related to the primary nearby stars and have the same age. 
Our carefully selected sample includes three remote white dwarf companions to main-sequence stars and two systems 
(55 Cnc and GJ 111 PC) of multiple planets and distant stellar companions. Ten new CPM companions, including three 
of extreme separations, are found. Multiple hierarchical systems are abundant; more than 25% of CPM components are 
spectroscopic or astrometric binaries or multiples themselves. Two new astrometric binaries are discovered among nearby 
CPM companions, GJ 264 and HIP 59000, and preliminary orbital solutions are presented. The Hyades kinematic group 
(or stream) is presented broadly in the sample, but we find few possible thick-disk objects and no halo stars. It follows from 
our investigation that moderately young (age Si 1 Gyr) thin-disk dwarfs are the dominating species in the near CPM 
systems, in general agreement with the premises of the dynamical survival paradigm. 

Subject headings: binaries: general — stars: kinematics 


1. INTRODUCTION 

Components of wide stellar binaries and common proper mo¬ 
tion pairs have been drawing considerable interest for many years. 
Despite the increasing accuracy of observations and the growing 
range of accessible wavelengths, the origin of very wide, weakly 
bound, or unbound systems remains an open issue. Lepine & 
Bongiomo (2007) estimated that at least 9.5% of stars within 
100 pc have companions with projected separations greater than 
1000 AU. The renewed interest was boosted by the detection of a 
dearth of substellar mass companions in spectroscopic binaries 
and by the attempts to account for the missing late-type members 
of the near solar neighborhood. 

The main objective of this paper is to investigate a well-defined 
set of nearby stars in very wide CPM pairs and to discover new 
pairs, possibly with low-mass companions. The secondary goal of 
our investigation is to establish or estimate the age and evolu¬ 
tionary status of bona fide companions using a wide range of 
available astrometric and astrophysical data. The origin and status 
of wide CPM systems is still a mystery, because most of them are 
likely unbound or very weakly bound and are expected to be eas¬ 
ily disrupted in dynamical interactions with other stars or mo¬ 
lecular clouds (§ 3). The nearest stars to the Sun, a Cen and 
Proxima Cen, form a wide pair which may be on a hyperbolic 
orbit (Anosova & Orlov 1991). It is expected that such systems 
should be mostly young, or belong to moving groups, remnant 
clusters or associations, but this has not yet been demonstrated 
on a representative sample. It is not known if the companions 
formed together and have the same age. We combine age-related 
parameters and data, including color-absolute magnitude dia¬ 
grams (§ 4), chromospheric activity indices (§ 6.1), coronal X-ray 
luminosity (§ 6.2), multiplicity parameters (§ 7), and kinematics 
(§ 8) to shed light on this problem. 


Previous investigations in the field are too numerous to be listed, 
but a few papers in considerable overlap with this study should 
be mentioned. Poveda et al. (1994) published a catalog of 305 
nearby wide binary and 29 multiple systems. They discussed the 
importance of moving groups for separating different species of 
wide binaries and tentatively assigned 32 systems to the Hyades 
stream (called a supercluster, following Eggen’s nomenclature) 
and 14 to the Sirius stream. Salim & Gould (2003) undertook a 
comprehensive revision of the Luyten catalog for approximately 
44% of the sky, drastically improving precision of epoch 2000 
positions and proper motions, and supplying the stars with NIR 
magnitudes from 2MASS. This allowed Gould & Chaname 
(2004) to estimate, for the first time, trigonometric parallaxes of 
424 common proper motion companions to Hipparcos stars with 
reliable parallaxes. This extrapolation of parallaxes to CPM com¬ 
panions is justified for high proper motion pairs where the rate 
of chance alignments is small. There is significant overlap be¬ 
tween the sample investigated in this paper and the catalog of 
Gould & Chaname (2004), although we did not use the latter as 
a starting point for our selection. We are also employing the par¬ 
allax extrapolation technique for dim companions not observed by 
Hipparcos when constructing color-magnitude diagrams in this 
paper. 

2. SELECTING CPM SYSTEMS 

Our selection of candidate CPM systems was based on the Naval 
Observatory Merged Astrometric Dataset (NOMAD; Zacharias 
et al. 2004b), 1 which provides an all-sky catalog of astrometric 
and photometric data. NOMAD includes astrometric data from the 
UNSO-B (Monet et al. 2003), UCAC2 (Zacharias et al. 2004a), 

1 At http://www.nofs.navy.mil/nomad. 


566 



Report Documentation Page 

Form Approved 

OMB No. 0704-0188 

Public reporting burden for the collection of information is estimated to average 1 hour per response, including the time for reviewing instructions, searching existing data sources, gathering and 
maintaining the data needed, and completing and reviewing the collection of information. Send comments regarding this burden estimate or any other aspect of this collection of information, 
including suggestions for reducing this burden, to Washington Headquarters Services, Directorate for Information Operations and Reports, 1215 Jefferson Davis Highway, Suite 1204, Arlington 

VA 22202-4302. Respondents should be aware that notwithstanding any other provision of law, no person shall be subject to a penalty for failing to comply with a collection of information if it 
does not display a currently valid OMB control number. 

1. REPORT DATE 

01 NOV 2008 2 ' REPORT TYPE 

3. DATES COVERED 

00-00-2008 to 00-00-2008 

4. TITLE AND SUBTITLE 

Common Proper Motion Companions to Nearby Stars: Ages and 

Evolution 

5a. CONTRACT NUMBER 

5b. GRANT NUMBER 

5c. PROGRAM ELEMENT NUMBER 

6. AUTHOR(S) 

5d. PROJECT NUMBER 

5e. TASK NUMBER 

5f. WORK UNIT NUMBER 

7. PERFORMING ORGANIZATION NAME(S) AND ADDRESS(ES) 

US Naval Observatory,3450 Massachusetts Avenue, 

NW,Washington,DC,20392-5420 

8. PERFORMING ORGANIZATION 

REPORT NUMBER 

9. SPONSORING/MONITORING AGENCY NAME(S) AND ADDRESS(ES) 

10. SPONSOR/MONITOR'S ACRONYM(S) 

11. SPONSOR/MONITOR'S REPORT 
NUMBER(S) 

12. DISTRIBUTION/AVAILABILITY STATEMENT 

Approved for public release; distribution unlimited 

13. SUPPLEMENTARY NOTES 

14. ABSTRACT 

15. SUBJECT TERMS 

16. SECURITY CLASSIFICATION OF: 17. LIMITATION OF 

___ ABSTRACT 

18. NUMBER 19a. NAME OF 

OF PAGES RESPONSIBLE PERSON 

a. REPORT b. ABSTRACT c. THIS PAGE Same OS 

unclassified unclassified unclassified Report (SAR) 

13 


Standard Form 298 (Rev. 8-98) 

Prescribed by ANSI Std Z39-18 





COMMON PROPER MOTION COMPANIONS 


567 


Hipparcos (ESA 1997), Tycho-2 (Hog et al. 2000) catalogs, and 
the Yellow Sky data set (D. Monet, private communication), sup¬ 
plemented by BVR optical photometry, mainly from USNO-B, 
and JHK near-IR photometry from 2MASS. This catalog covers 
the entire magnitude range from the brightest naked eye stars to 
the limit of the POSS survey plates (about 21st magnitude). The 
largest systematic positional errors are estimated for the Schmidt 
plate data used in the USNO-B catalog, with possible local offsets 
up to about 300 mas. Systematic errors in UCAC2 and 2MASS 
are much smaller (Zacharias et al. 2006). Thus, possible systema¬ 
tic errors of proper motions in NOMAD for the entries taken from 
the USNO-B catalog can be as large as 10 mas yr 1 , and for faint 
UCAC2 stars up to about 5 mas yr~ 1 . Internal random errors of 
proper motions are given for all stars in the NOMAD catalog; 
these are typically 5 to 10 mas yr -1 for faint stars. 

We used the following four criteria to select candidate CPM 
systems for this investigation: (1) at least one of the components 
should be listed in both Hipparcos and NOMAD; (2) the Hip¬ 
parcos parallax of the primary component should be statistically 
reliable and greater than 40 mas (distance less than 25 pc); (3) at 
least one companion to the primary is found in NOMAD within 
1.5° at epoch J2000.0, whose proper motion is within a tolerance 
limit of the primary’s proper motion; (4) the companion should 
be clearly visible in both DSS and 2MASS surveys, and be listed 
in 2MASS with J, H, and K s magnitudes. Some 1200 stars from 
the Hipparcos Catalogue with a parallax of 40 mas or larger were 
selected as initial targets. The tolerance limit was set at 8 mas yr~ 1 
if the difference of the primary’s Hipparcos and Tycho-2 proper 
motions was larger than this value in at least one of the coordinate 
components, and at 15 mas yr~* otherwise. In addition, the dif¬ 
ference in proper motion between the primary and the candidate 
companion was required to be within the 3 a formal error on the 
NOMAD proper motion. These fairly strict limits removed from 
our analysis some known or suspected nearby pairs, for example, 
the nearest stars Alpha and Proxima Cen, which differ in proper 
motion by more than a hundred milliarcseconds per year. The re¬ 
sulting list of candidates was inspected by eye to exclude numer¬ 
ous erroneous NOMAD entries. In this paper we consider only 
CPM systems with projected separations greater than 1000 AU. 

Table 1 lists 41 CPM systems, including 2 resolved triple sys¬ 
tems. Alternative names are given for all companions, giving pref¬ 
erence to Hipparcos numbers, various Luyten designations, and 
Gliese-Jahreiss identifications. The sources of J2000.0 ICRS po¬ 
sitions are Hipparcos and NOMAD. The VI photometry comes 
mostly from (Bessel 1990; Weis 1991, 1993, 1996; Koen et al. 
2002; Reid et al. 2002; Rosello et al. 1987) and for several stars 
from our own observations. 

3. DYNAMICAL SURVIVAL AND ORIGINS 

Very wide stellar systems of low binding energy encounter 
other stars and molecular clouds as they travel in the Galaxy, and 
these dynamical interactions are the main cause for stochastic evo¬ 
lution of their orbits and, in most cases, eventual disruption. Ana¬ 
lytical considerations of dynamical evolution and survival of wide 
systems is limited to two asymptotic approximations, those of very 
distant and weak (but frequent) interaction, and “catastrophic” 
encounters at small impact parameters, which are rare but can be 
disruptive. We discuss in this paper the second kind of interac¬ 
tions. According to Weinberg et al. (1987) catastrophic encoun¬ 
ters are defined as those with impact parameters b < 7>bf, where 
6 bf is defined, in analogy with Fokker-Planck diffusion coeffi¬ 
cients, as the critical impact parameter at which the expected var¬ 


iance of total energy is equal to a certain fraction of the total energy 
squared: 


a 


2 

A E 


= cE 2 . 


( 1 ) 


The following proportionality relations are derived from (Weinberg 
et al. 1987) for the rates of catastrophic encounters: 


r cat oc nae x ( 

/ 1 ' 

\ Viel, 

^ (Z>BF ^ a )i 

(2) 

r cat oc «a32U 

-(1/2) 

(6bf ^ a), 

(3) 


where n is the perturber number density, a is the semimajor axis 
of the binary system, (1/F re i) is the mean reciprocal relative ve¬ 
locity of encounters. 

The first limiting case, 7 >bf ^ a, corresponds to encounters 
with individual stars, while the second, 6 bf » a, is a suitable ap¬ 
proximation for encounters with dense cores of molecular clouds. 
Note that in the latter case, the rate of high-energy interactions is 
independent of the relative velocity. 

The rate of disruptive interactions for both scenarios is pro¬ 
portional to the number density of perturbers n. It becomes im¬ 
mediately clear that the rate of disruption of very wide binaries 
is drastically different for the three major dynamical constituents 
of the Galaxy, the thin disk, the thick disk, and the halo. 

The halo stars populating the outer, spherical component of 
the Galaxy have by far the largest velocities when they happen to 
travel in our neighborhood. The mean velocity with respect to 
the local standard of rest is directly related to the dispersions of 
velocity components {a v , ay, a>). The “pure” halo, according 
to Chiba & Beers (2000), is characterized by a prograde rotation 
of ~ 30 to 50 km s _1 and a dispersion ellipsoid of (ay, ay, 
aw) = (141 ± 11,106 ± 9, 94 ± 8)kms _1 . The vertical veloc¬ 
ity component has immediate dynamical implications for wide 
binaries. The number density of molecular clouds, as well as of 
field stars is nonuniform in the vertical dimension, with a cusp at 
z = 0. Wide binaries from the halo cross the densest part of the 
disk very quickly and spend most of their time hovering far from 
the plane where the density of perturbers is much lower. On the 
contrary, the thin-disk stars spend most of the time within the 
densest parts of the Galaxy, oscillating with small amplitudes 
around its midplane. These dynamical differences has dramatic 
implications for the typical survival time of very wide binaries. 
We can quantify the differences in the following way. 

According to the numerical simulations of galactic motion in 
Makarov et al. (2004), the vertical oscillation is harmonic to first- 
order approximation, with a period 


TV(fzo) — PyH 


i + T ^ 1 - 45 


3.29M) 


(4) 


where v z q is midplane vertical velocity in kilometers per second and 
Pv,h = 77.7 Myr is the asymptotic hannonic period at r-o —> 0. 
This equation holds within ±0.5% for 0 < v z o < 21 km s _1 . An¬ 
other useful equation relates the maximum excursion from the Ga¬ 
lactic plane with the midplane velocity: 

Zmax = 12.044|l>-o|. (5) 

Assuming typical midplane velocities to be equal to vertical dis¬ 
persion estimates from Torra et al. (2000) for young stars, Famaey 



TABLE 1 

Examined CPM Double and Multiple Systems within 25 pc 


R.A. Decl. 

HIP/Name Alt. Name (J2000.0) (J2000.0) Sep. /i a cosS n e n[cr n ] V V - I J H K 

(1) (2) (3) (4) (5) (6) (7) (8) (9) (TO) (11) (12) (13) 


473. GJ 4 A 00 05 41.0129 +45 48 43.491 879 -154 85.10[2.74] 8.23 1.77 6.10 6.82 5.26 

428. GJ 2 00 05 10.8882 +45 47 11.641 328.1 870 -151 86.98[1.41] 9.97 2.53 6.70 6.10 5.85 


4849. GJ 3071 AB 01 02 24.5721 +05 03 41.209 340 221 46.61[1.61] 8.15 6.20 5.68 5.51 

WD 0101 + 048. GJ 1027 01 03 49.9093 +05 04 30.840 1276.0 320 232 14.10 0.34 13.50 13.40 13.42 


4872. GJ 49 01 02 38.8665 +62 20 42.161 730 89 99.44[1.39] 9.56 1.97 6.23 5.58 5.37 

V388 Cas. GJ 51 01 03 19.8653 +62 21 55.930 294.7 732 80 13.78 3.32 8.61 8.01 7.72 


5799. GJ 9045 A 01 14 24.0398 -07 55 22.173 124 278 41.01[0.89] 5.14 0.54 4.40 4.02 4.06 

LTT 683 . GJ 9045 B 01 14 22.4332 -07 54 39.232 49.1 123 272 7.83 0.83 6.40 6.02 5.88 


9749. GJ 9070 A 02 05 23.6559 -28 04 11.032 340 422 44.37[1.97] 10.96 1.69 7.99 7.35 7.16 

LTT 1097. GJ 9070 B 02 05 24.6587 -28 03 14.570 58.0 324 422 12.82 2.49 8.80 8.26 8.04 


14286. GJ 3194 A 03 04 09.6364 +61 42 20.988 721 -693 43.74[0.84] 6.67 5.39 512 5.03 

LTT 1095. GJ 3195 B 03 04 43.4407 +61 44 08.950 263.4 718 -698 12.55 2.27 8.88 8.33 8.10 


14555. GJ 1054 A 03 07 55.7489 -28 13 11.013 -339 -120 55.5[2.5] 10.24 1.70 7.24 6.58 6.37 

LTT 1477. GJ 1054 B 03 07 53.3793 -28 14 09.650 66.5 -336 -112 13.09 2.31 9.35 8.78 8.52 


15330. GJ 136 03 17 46.1635 -62 34 31.160 1338 649 82.51[0.54] 5.53 0.71 4.46 4.04 3.99 

15371. GJ 138 03 18 12.8189 -62 30 22.907 309.2 1331 647 82.79[0.53] 5.24 0.68 4.27 3.87 3.86 


17414. GJ 9122 A 03 43 52.5624 +16 40 19.272 155 -320 58.09[1.98] 9.96 1.65 7.05 6.41 6.25 

17405. GJ 9122 B 03 43 45.2490 +16 40 02.138 106.5 159 -313 61.40[2.37] 10.81 1.94 7.53 6.91 6.69 


21482. V833 Tau 04 36 48.2425 +27 07 55.897 232 -147 56.02[1.21] 8.10 1.60 5.95 5.40 5.24 

WD 0433+270. NLTT 13599 04 36 44.8902 +27 09 51.594 124.0 226 -153 15.81 0.80 14.60 14.23 14.14 


22498. DP Cam 04 50 25.0911 +63 19 58.624 219 -195 42.59[17.78] 9.83 1.22 7.55 6.95 6.80 

G 247-35. 04 50 21.6640 +63 19 23.420 42.1 210 -194 12.72 2.15 9.20 8.59 8.36 


25278. GJ 202 05 24 25.4633 +17 23 00.722 250 -7 68.2[0.9] 5.00 4.43 4.03 4.04 

25220. GJ 201 05 23 38.3810 +17 19 26.829 707.2 253 -5 69.8[1.5] 7.88 1.25 5.88 5.38 5.23 


34065. GJ 9223 A 07 03 57.3176 -43 36 28.939 -103 389 61.54[1.05] 5.27 a 0.73 4.41 3.99 4.04 

34069. GJ 9223 B 07 03 58.9171 -43 36 40.857 21.1 -99 383 66.29[6.81] 6.86 0.83 5.46 5.08 4.94 

34052. GJ 264 07 03 50.24 -43 33 40.7 184.8 -93 395 58.89[0.94] 8.69 a 1.35 6.45 5.83 5.70 


42748. GJ319A 08 42 44.5315 +09 33 24.114 216 -634 74.95[13.82] 9.63 1.86 6.69 6.05 5.83 

GJ 319 C 08 42 52.2287 +09 33 11.157 114.6 224 -616 11.81 2.39 8.12 7.49 7.28 

43587. GJ 324 A 08 52 35.8111 +28 19 50.947 -485 -234 79.80[0.84] 5.96 4.77 4.26 4.01 

LTT 12311. GJ 324 B 08 52 40.8393 +28 18 59.310 84.1 -488 -234 13.14 3.00 8.56 7.93 7.67 


46843. GJ 9301 A 09 32 43.7592 +26 59 18.708 -148 -246 56.35[0.89] 7.01 5.58 5.24 5.12 

NLTT 22015 . GJ 9301 B 09 32 48.2450 +26 59 43.864 65.0 -142 -243 10.36 9.86 9.47 


47620. GJ 360 09 42 34.8429 +70 02 01.989 -671 -269 85.14[3.18] 10.58 2.20 6.92 6.33 6.08 

47650. GJ 362 09 42 51.7315 +70 02 21.892 88.8 -669 -264 86.69[2.24] 11.24 2.41 7.33 6.73 6.47 


49669. GJ 9316 A 10 08 22.3106 +11 58 01.945 -249 5 42.09[0.79] 1.35 1.67 1.66 1.64 

GJ 9316 B 10 08 12.7970 +11 59 49.078 176.0 -244 12 8.11 1.00 6.42 5.99 5.88 

50564. GJ 9324 10 19 44.1679 +19 28 15.290 -230 -215 47.24[0.82] 4.80 4.04 3.94 4.02 

NLTT 23781 . 10 14 53.8493 +20 22 14.590 5220.6 -232 -212 16.48 10.81 10.20 9.99 


59000. GJ 9387 12 05 50.6574 -18 52 30.916 -19 -320 44.41[1.51] 9.95 1.57 7.42 6.79 6.62 

NLTT 29580. 12 05 46.6407 -18 49 32.240 187.6 -4 -314 16.23 3.32 11.20 10.63 10.32 


59406. GJ 3708 A 12 11 11.7583 -19 57 38.064 -216 -184 78.14[2.80] 11.68 2.33 7.89 7.36 7.04 

NLTT 29879. GJ 3709 B 12 11 16.95 -19 58 21.9 85.2 -203 -188 12.62 2.51 8.60 8.01 7.74 


61451. GJ 1161 A 12 35 33.5525 -34 52 54.901 -228 -134 46.19[0.91] 7.87 1.07 5.95 5.44 5.26 

LTT 4788 . GJ 1161 B 12 35 37.7821 -34 54 15.309 95.8 -219 -128 11.76 2.39 8.15 7.58 7.30 

63882. GJ 3760 13 05 29.8783 +37 08 10.635 -304 -202 43.18[6.95] 10.62 8.22 7.61 7.35 

NLTT 33194. 13 11 24.2045 +37 24 37.197 4342.8 -305 -193 11.89 11.36 11.10 




















































COMMON PROPER MOTION COMPANIONS 


569 


TABLE 1 —Continued 

R.A. Decl. 

HIP/Name Alt. Name (J2000.0) (J2000.0) Sep. ^ a cos 6 \x b n[<r n ] V V - / J H K 

(1) (2) (3) (4) (5) (6) (7) (8) (9) (10) (11) (12) (13) 

65083. LTT 5136 13 20 24.9410 -01 39 27.026 129 -251 48.18[3.05] 11.61 1.94 8.39 7.79 7.53 

LTT 5135. 13 20 12.5595 -01 40 40.980 199.8 129 -257 13.41 2.39 9.57 8.98 8.78 


65877. GJ 515 13 30 13.6398 -08 34 29.492 -1107 -475 55.50[3.77] 12.39 -0.01 12.62 12.68 12.74 

LTT 5214. 13 30 02.8247 -08 42 25.530 502.3 -1102 -472 14.33 3.04 9.60 9.05 8.75 


66492. NLTT 34715 13 37 51.2257 +48 08 17.079 -234 -139 45.66[2.72] 9.77 1.60 6.94 6.34 6.14 

NLTT 34706. GJ 520 C 13 37 40.4407 +48 07 54.169 110.4 -225 -136 14.47 2.73 10.12 9.59 9.30 


71914. GJ 9490 A 14 42 33.6486 +19 28 47.219 -254 -154 44.54[2.57] 9.1 l a 1.33 6.60 5.97 582 

71904. LTT 14350 14 42 26.2580 +19 30 12.694 135.0 -261 -177 38.62[2.01] 10.08 1.43 7.45 6.80 6.66 


75718. GJ 586 A 15 28 09.6114 -09 20 53.050 73 -363 50.34[1.11] 6.89 0.87 5.44 5.05 4.89 

75722. GJ 586 B 15 28 12.2103 -09 21 28.296 52.2 82 -356 48.06[1.14] 7.54 0.91 5.99 5.55 5.46 


79607. GJ 9550 A 16 14 40.8536 +33 51 31.006 -266 -87 46.11 [0.98] 5.70 0.80 3.95 3.35 4.05 

79551. GJ 9549 16 13 56.4533 +33 46 25.030 632.3 -264 -84 43.82[5.69] 12.31 3.06 8.60 8.00 7.75 


82817. GJ 644 AB 16 55 28.7549 -08 20 10.838 -829 -879 174.22[3.90] 9.02 2.33 5.27 4.78 4.40 

82809. GJ 643 16 55 25.2251 -08 19 21.274 72.1 -813 -895 153.96[4.04] 11.74 2.63 7.55 7.06 6.72 

LHS 429. GJ 644 C 16 55 35.2673 -08 23 40.840 231.2 -810 -872 154.5[0.7] b 16.85 b 4.54 b 9.78 9.20 8.82 


83599. GJ 653 17 05 13.7781 -05 05 39.220 -921 -1128 89.70[28.71] 10.09 2.13 6.78 6.19 5.97 

83591. GJ 654 17 05 03.3941 -05 03 59.428 184.5 -917 -1138 92.98[1.04] 7.73 1.35 5.52 4.94 4.73 


86036. 26 Dra 17 34 59.5940 +61 52 28.394 277 -526 70.98[0.55] 5.23 4.24 3.88 3.74 

86087. GJ 685 17 35 34.4809 +61 40 53.631 737.5 264 -514 70.95[1.09] 9.97 1.81 6.88 6.27 6.07 


93899. GJ 745 B 19 07 13.2039 +20 52 37.254 -481 -333 112.82[2.41] 10.76 2.09 7.28 6.75 6.52 

93873. GJ 745 A 19 07 05.5632 +20 53 16.973 114.2 -481 -346 115.91[2.47] 10.78 2.09 7.30 6.73 6.52 


97295. GJ 9670 A 19 46 25.6001 +33 43 39.351 19 -446 47.94[0.54] 4.96 0.53 4.05 3.98 3.83 

97222. LTT 15766 19 45 33.5520 +33 36 06.055 792.3 23 -449 49.09[1.43] 7.68 5.81 5.32 5.25 

LTT 15775. GJ 9670 B 19 46 27.5446 +33 43 48.894 25.8 25 -438 8.58 1.13 6.64 6.12 6.00 


98204. GJ 773 19 57 19.6421 -12 34 04.746 -94 -513 52.92[1.48] 9.31 1.36 6.82 6.20 6.02 

NLTT 48475 . 19 57 23.8000 -12 33 50.260 62.6 -92 -518 15.36 3.39 10.21 9.65 9.32 


98767. GJ 777 A 20 03 37.4055 +29 53 48.499 683 -524 62.92[0.62] 5.73 4.55 4.24 4.08 

LTT 15865. GJ 777 B 20 03 26.5799 +29 51 59.595 178.0 687 -530 14.38 3.03 9.55 9.03 8.71 


102409. GJ 803 20 45 09.5317 -31 20 27.238 261 -345 100.59[1.35] 8.75 2.07 5.81 5.20 4.94 

102141. GJ 799 20 41 51.1537 -32 26 06.730 4680.0 280 -360 97.80[4.65] 10.33 2.92 5.44 4.83 4.53 


109084. GJ 4254 22 05 51.2986 -11 54 51.022 -266 -175 46.70[7.86] 10.15 7.22 6.60 6.40 

LP 759-25. 22 05 35.7280 -11 04 28.820 3030.9 -274 -162 11.66 11.05 10.72 


113602. NLTT 9310 23 00 33.4015 -23 57 10.309 190 -345 49.15[3.03] 11.57 1.95 8.25 7.67 7.41 

113605. NLTT 9315 23 00 36.5922 -23 58 10.657 74.5 195 -346 49.36[3.19] 11.61 1.98 8.26 7.66 7.42 


115147. V368 Cep 23 19 26.6320 +79 00 12.666 201 72 50.65[0.64] 7.54 5.90 5.51 5.40 

LSPM J2322+7847.... 23 22 53.8733 +78 47 38.810 959.1 210 64 16.18 3.62 10.42 9.84 9.52 

116215. GJ 898 A 23 32 49.3998 -16 50 44.308 344 -218 71.70[1.36] 8.62 1.28 6.24 5.61 5.47 

116191. GJ 897 23 32 46.5991 -16 45 08.395 338.3 382 -186 89.9[7.3] 10.43 2.24 6.71 6.09 5.86 


Notes.— Units of right ascension are hours, minutes, and seconds, and units of declination are degrees, arcminutes, and arcseconds. Col. (1): HIP number or name. 
Col. (2): Alternative name. Col. (3): Right ascension. Col. (4): Declination. Col. (5): Separation on the sky in arcseconds. Cols. (6) and (7): Proper motion in mas yr -1 . 
Col. (8): Parallax and its error in mas. Col. (9): V magnitude. Col. (10): V — I color. Col. (11): J magnitude. Col. (12): H magnitude. Col. (13): K magnitude. 
a Our photometric observations. 
b Photometry and parallax from Dahn et al. (2002). 

et al. (2005) for thin-disk giants, and Chiba & Beers (2000) for the fined as the fraction of an oscillation period when the star is within 

thick disk and halo, we estimate characteristic midplane veloci- 100 pc of the plane, f(\z\ < 100). 

ties, maximum vertical excursions, periods of oscillation, and the The halo binaries cross the thin disk so quickly that their chances 

fraction of lifetime spend in the dense part of the Galaxy for these to encounter a perturber (a field star or a molecular core) are rela- 

four dynamical components (Table 2). The latter parameter is de- tively slim. Thus, generic binaries of very low binding energy can 











































570 


MAKAROV & HENNESSY 


Vol. 687 


TABLE 2 

Vertical Motion of Galactic Components 


All -max U 

Component (kms -1 ) (pc) (Myr) f(\z\ < 100) 


Thin disk (young). 6 72 79 1.00 

Thin disk (giants). 18 217 86 0.31 

Thick disk. 35 422 109 0.16 

Halo. 94 1130 305 0.07 


probably survive for a long time in the halo. However, these ob¬ 
jects are rare in the solar neighborhood because of the intrinsic 
low number density, and none seem to be present in our sample. 
A typical thick disk binary may also stay intact much longer than 
a young disk binary, because it spends at least 6 times less time in 
the high-density midplane area. As far as encounters with stars 
are concerned, the difference in the time of survival can be even 
greater, because the average reciprocal velocity of encounter en¬ 
ters equation (2). Most of the interactions of thick-disk binaries 
with thin-disk perturbers will be rapid, further reducing the rate 
of disruptive events. 

We can expect from this analysis that the distribution of very 
wide binaries and common proper motion pairs in age should be 
bimodal. Young CPM pairs in the thin disk, despite the higher rate 
of catastrophic interactions, can survive in significant numbers to 
this day. This kind of binary should be especially prominent if in¬ 
deed most of the new stars are bom in loose comoving groups 
such as the Lupus association of pre-main-sequence stars (Makarov 
2007b). 

4. COLOR-ABSOLUTE MAGNITUDE DIAGRAMS 

Figure 1 represents the joint M Ks versus V — K s color-absolute 
magnitude diagram for all resolved CPM companions listed in 
Table 1 that have V and JHK magnitudes. We assumed in con¬ 
structing this diagram zero extinction for all stars, and we applied 
the Hipparcos parallaxes determined for primary stars to their 
CPM companions, unless the latter have independent trigono¬ 
metric parallax measurements. Known unresolved binary or mul¬ 
tiple stars are marked with inscribed crosses. A zero-age main 
sequence (ZAMS) and a 16 Myr isochrone at solar metallicity 
(Z = 0.001; from Siess et al. 2000) are drawn with thin lines, and 
the empirical main sequence from Henry et al. (2004) with a thick 
dashed line. Some of the interesting stars discussed later in this pa¬ 
per are labeled and named. Mutual positions of primary and sec¬ 
ondary CPM companions are shown with dotted lines only for 
pairs with white dwarf companions. 

The diagram shows that most of normal stars lie on or slightly 
above the main sequence. This confirms that the fainter CPM 
companions are probably physical. We find that many of the com¬ 
ponents lying close to the 16 Myr isochrone (top solid line) are 
known visual, astrometric or spectroscopic binaries, which ac¬ 
counts for their excess brightness. For example, the primary com¬ 
ponent of the CPM pair HIP 66492 and NLTT 34706 is a resolved 
binary with a period P = 330 yr, semimajor axis a = 2.13", and 
eccentricity e = 0.611 (Seymour et al. 2002). Formally, the joint 
magnitude can be as much as 0.75 brighter (in the case of twin 
companions) than the magnitude of the primary star. A number 
of components in Figure 1 lie significantly outside the upper en¬ 
velope of unresolved binaries defined by the empirical main se¬ 
quence minus 0.75 mag. Gross photometric errors (in particular, 
in V for faint M dwarfs) cannot be completely precluded, but we 
suspect that most of these outlying stars should be either very 
young or unresolved multiple stars. 



Fig. 1. — Joint color-absolute magnitude diagram of CPM pairs in Mk s vs. 
V — K s axes. The ZAMS and 16 Myr isochrones are drawn from (Siess et al. 
2000), both for Z = 0.001. The three white dwarfs of our sample are connected 
with their M dwarf companions by dotted straight lines. The thicker dashed line 
indicates the empirical main sequence for field dwarfs from Henry et al. (2004). 
Known unresolved binary companions of all kinds are marked with crosses in¬ 
scribed in circles. 

The CPM pair of HIP 61451 and LTT 4788 is an emphatic ex¬ 
ample of overluminous stars whose origin remains an unresolved 
issue. They match the 16 Myr isochrone on the HR diagram very 
well. The A' v -band excess for these companions is 0.7 and 1.0 mag, 
respectively. In the literature we found no indication of binarity 
for either star. The primary component can still be binary with 
an almost twin companion, but the secondary should be at least 
triple to account for the near-infrared excess if it is a normal (old) 
M2.5 dwarf. On the basis of the kinematics of HIP 61451 and its 
excess luminosity, Eggen (1995) included it in his list of the 
Pleiades supercluster, which is synonymous with the Local Young 
Stream (Makarov & Urban 2000). This may indicate an age be¬ 
tween 1 and 125 Myr. Furthermore, HIP 61451 is a moderate and 
very soft X-ray emitter (Table 4), which may be expected of a 
post-T Tauri star. On the other hand, its level of chromospheric 
activity, at logR(j K = —4.601 (Gray et al. 2006), is not impres¬ 
sive, corresponding to an activity age of ~1 Gyr (see § 6). We 
propose that a careful investigation of the M-type CPM compan¬ 
ion LTT 4788 should resolve the mystery of this system. 

Figure 2 shows a color-absolute magnitude diagram of some 
selected CPM components discussed below in more detail, on My 
versus V — K s axes. Each star is identified with its Hipparcos num¬ 
ber or other name. The two thin lines show the ZAMS (bottom) and 
the 16 Myr isochrone (top) from the models by Siess et al. (2000) 
both forZ = 0.001 and zero extinction. The thicker dash-dotted 
line is the empirical main sequence for field stars from (Reid & 
Cruz 2002). In this plot the CPM components are connected with 
thin dashed lines to show their relative position. Again, trigono¬ 
metric parallaxes of the primary components were assumed for 
faint companions with unknown distances. 

5. NEW CPM PAIRS 

We report 10 new possible CPM companions and 8 new CPM 
systems, including 3 at extremely large separations, identified by 
our search procedure (§ 2). Table 3 gives the Washington Double 
Star catalog (WDS; Mason et al. 2001) identifications for the 
primaries of known systems and indicates new systems and wide 
companions. The original discoverer references and other catalog 








No. 1, 2008 


COMMON PROPER MOTION COMPANIONS 


571 



Fig. 2. —Color-absolute magnitude diagram of selected CPM companions in 
My versus V — K s axes. The zero-age main sequence (ZAMS) and 16 Myr iso¬ 
chrone are drawn from Siess et al. (2000), both for Z = 0.001. Components of 
CPM pairs are connected by dashed straight lines. The thicker dot-dashed line in¬ 
dicates the empirical main sequence for field dwarfs from Reid & Cruz (2002). 

identifications can be found in WDS. In this section we discuss 
four probable new systems with peculiar characteristics, which 
may be interesting to pursue with additional photometric and spec¬ 
troscopic observations. 

HIP 109084 is a rather nondescript M0 dwarf at approximately 
20 pc. This star has an uncertain parallax in the Hipparcos catalog 
with a formal error of 7.9 mas, most likely affected by unresolved 
binarity. According to Gizis et al. (2002), the Ha line is in ab¬ 
sorption (EW = —0.55 A), hence the Ha lower limit on age is 
150 Myr (§ 6). Its alleged CPM companion LP 759-25, as one of 
the nearest and latest M dwarfs, has drawn more interest. Phan- 
Bao & Bessell (2006) estimate a spectroscopic distance of 18 pc 
for this star. At a projected separation of 65,000 AU, this may be 
one of the widest known CPM pairs, but more accurate astro- 
metric information is required to verify the physical connection 
between these stars. 

The K3 dwarf HIP 4849 at 21 pc from the Sun is a binary re¬ 
solved by Hipparcos and with speckle interferometry (Fabricius 
& Makarov 2000b; Balega et al. 2006). Its inner companion is 
probably a K8 dwarf orbiting the primary at a = 465 mas with a 
period of 29 yr. We propose that this binary system has a distant 
CPM companion, the DA5 white dwarf WD 0101+048. The pro¬ 
jected separation between the CPM components is 27,000 AU. 
The white dwarf companion is itself a binary star, having a spec¬ 
troscopically detected close DC white dwarf companion (Maxted 


et al. 2000). The center-of-mass radial velocity of the WD pair is 
63.4 ± 0.2 km s _1 , whereas Nidever et al. (2002) determine a 
radial velocity of 22.17 km s~ 1 for the primary K dwarf. The ra¬ 
dial velocity for the WD companion is likely to include the gravi¬ 
tational redshift, which may account for the apparent difference 
with HIP 4849. We estimate an age of 1.3 Gyr for HIP 4849 from 
a chromospheric activity index of logRjj K = —4.661 given by 
Gray et al. (2003), whereas the cooling age for WD 0101+048 
is 0.63 ± 0.07 Gyr, not including its main-sequence lifetime 
(Bergeron et al. 2001). 

The pair of stars HIP 50564 and NLTT 23781 is remarkable not 
least because of its extreme separation (5230", or 111,000 AU on 
the sky). Other interesting properties of this system are discussed 
in § 6.2. 

The CPM pair of HIP 22498 (DP Cam) and G247-35 is sepa¬ 
rated by “only” 1000 AU in the sky projection, and it is surprising 
it has not been identified as such before. The primary component, 
a K7 dwarf, is listed as eclipsing binary in the catalog of eclipsing 
stars ( Maikov et al. 2006). Its Hipparcos parallax is very poor 
even for a “stochastic” solution, indicating an unresolved type 
of binarity; however, both this binary and its distant M-type com¬ 
panion lie on the main sequence in Figures 1 and 2. Very little is 
known about G 247-35, apart from the photometric observations 
in (Weis 1988). 

6. ACTIVITY AND AGES 

6.1. Chromospheric Activity 

The so-called Ha limit relation tells us that there is a certain 
age in the evolution of M dwarfs of a given mass (or V — 1 color) 
when the ubiquitous chromospheric activity, related by emission 
in the Ha line, disappears and the stars transform from dMe to 
normal inactive dwarfs (Gizis et al. 2002). The empirical relation, 
fairly well defined on open clusters, can be written as 

log Age HQ = 0.952(F-/ c +6.91). (6) 

This formula should be used with caution because recent stud¬ 
ies of M dwarf activity based on large samples of stars selected 
from the Sloan Digital Sky Survey indicate that the activity life¬ 
time versus spectral-type relation is strongly nonlinear (West 
et al. 2008), with a steep ascent between M3 and M5. This ab¬ 
rupt change may be related to the transition from partially con¬ 
vective to fully convective stellar interiors. Most of the latest 
M dwarfs in the solar neighborhood are active, but an age-activity 
correlation is still evident at spectral type M7 where the fraction of 
chromospherically active stars declines with the distance from the 
Galactic plane (West et al. 2006). This relation can be used to dif¬ 
ferentiate the oldest late-type M dwarfs, although exact calibration 


TABLE 3 

WDS Identifications and New Pairs 


HIP 

WDS 

HIP 

WDS 

HIP 

WDS 

HIP 

WDS 

HIP 

WDS 

473. 

. 00057+4549 

4849 

New 

4872 

New 

5799 

New 

9749 

02053-2803 

14286. 

. 03042+6142 

14555 

03079-2813 

15371 

03182-6230 

17414 

03439+1640 

21482 

04368+2708 

22498. 

. 04503+6320 

25278 

New 

34065 

07040-4337 

42748 

08427+0935 

43587 

08526+2820 

49669. 

. 10084+1158 

46843 

09327+2659 

47620 

09427+7004 

50564 

New 

59000 

New 

59406. 

. 12113-1958 

61451 

12356-3453 

63882 

13055+3708 

65083 

13203-0140 

65877 

13303-0834 

66492. 

. 13379+4808 

71914 

14426+1929 

75718 

15282-0921 

79607 

16147+3352 

82817 16555- 

-0820 (231" comp, is new) 

83591. 

. 17050-0504 

86036 

17350+6153 

93899 

19072+2053 

97295 

19464+3344 (792" comp, is new) 

98204 

19573-1234 

98767. 

. 20036+2954 

102409 

20452-3120 

109084 

New 

113602 

New 

115147 

23194+7900 


116215. 23328-1651 













572 


MAKAROV & HENNESSY 


Vol. 687 


TABLE 4 

X-Ray Luminosities 


HIP/Name 

(1) 


Lx 

(2) 


HR1 

(3) 

115371. 

0.058 

± 

0.013 

-0.91 

61451. 

0.11 

± 

0.04 

-0.86 

102409. 

5.59 

± 

0.11 

-0.07 

102141. 

3.38 

± 

0.10 

-0.19 

14555. 

4.62 

± 

0.27 

-0.27 

59000. 

0.71 

± 

0.24 

+0.20 

116215. 

0.13 

± 

0.03 

-0.56 

116191. 

1.41 

± 

0.09 

-0.28 

75722. 

0.43 

± 

0.07 

-0.40 

82817. 

1.09 

± 

0.07 

-0.26 

5799. 

0.74 

± 

0.12 

-0.01 

25278. 

0.16 

± 

0.03 

-0.48 

25220. 

2.75 

± 

0.11 

-0.12 

50564. 

1.09 

± 

0.38 

+0.22 

46843. 

1.49 

± 

0.09 

-0.19 

21482. 

8.20 

± 

2.38 

-0.04 

97222. 

0.087 

± 

0.023 

-0.85 

LTT 15775 . 

0.048 

± 

0.06 

-0.29 

79607. 

46.1 

± 

0.6 

+0.06 

473. 

0.050 

± 

0.014 

-0.42 

66492. 

0.076 

± 

0.028 

-0.74 

86036. 

0.47 

± 

0.02 

-0.48 

86087. 

0.028 

± 

0.005 

-0.58 

V388 Cas. 

0.20 

± 

0.02 

-0.19 

47650. 

0.18 

± 

0.03 

-0.40 

115147. 

10.6 

± 

0.2 

-0.10 


Notes. —Col. (1): HIP number or name. Col. (2): X-ray 
luminosity in units of 10 29 ergs s” 1 . Col. (3): Hardness ratio 
HR1 fromRO&47: 


is currently problematic because of the lack of independent age 
estimates. 

A widely used means of age estimation is provided by the em¬ 
pirical relation between the level of chromospheric activity as 
measured from the index of Ca n lines. The equation used in 
this paper, 

log Age HK = (-2.02 ± 0.13) log - (0.31 ± 0.63), (7) 

was derived by (Soderblom et al. 1991) for the Sun, Hyades, and 
UMa Group. We utilize these relations in Table 5 to estimate (very 
roughly) the ages and age limits for several late-type components. 

6.2. X-Ray Activity 

The binary and multiple systems under investigation in this 
paper are so wide that the observed ROSAT sources can be unam¬ 
biguously identified with individual components. Table 4 lists all 
the components identified by us in the ROSAT Bright Source and 
Faint Source catalogs (Voges et al. 1999, 2000). The hardness 
ratios EIR1 in this table are from the ROSAT catalogs, while the 
X-ray luminosities, in units of 10 29 ergs s , are computed from 
the specified count rates, hardness ratios, and Hipparcos paral¬ 
laxes. Most of the faint sources, with Lx < 1, are very soft, with 
HR1 closer to — 1. They are similar in X-ray activity to the qui¬ 
escent Sun, or slightly exceed it. The vast majority of weak nearby 
dwarfs are likewise soft, indicating low coronal activity (Elunsch 
et al. 1999). Normal M-type dwarfs have significantly smaller 
X-ray luminosities than G- and K-type stars. Indeed, most of 
the X-ray-weak systems include a K-type primary, and a few 
F-type primaries, whereas numerous M-type wide companions 


are not detected by ROSAT. A few notable M-type emitters should 
be mentioned. 

The star FI1P 14555 is a flare M0 dwarf with a Hipparcos par¬ 
allax of II = 52 ± 5 mas. This poor parallax determination, in 
addition to a great deal of confusion associated with this multiple 
system, is related to a failed component solution in Hipparcos, 
based on the wrong assumption that FIIP 14555 and TUP 14559 
(at separation 30.3", position angle 101°) form a physical pair at 
the same distance from the Sun. Fabricius & Makarov (2000a) 
resolved the Hipparcos data for this system using more accurate 
initial assumptions and obtained a parallax IT = 55.2 ± 2.5 mas 
and a proper motion p = (—339, —121) ± (2.5,2.2) mas yr~' 
for FIIP 14555, which is quite close to the original solution, but a 
II = 8.8 ± 9.4 mas and p = (—18, —37) ± (10, 9) mas yr~' for 
HIP 14559. Thus, these stars are certainly optical companions. 
LTT 1477 is probably a real, albeit more remote, CPM compan¬ 
ion. The outstanding X-ray brightness of HIP 14555 finds ex¬ 
planation in the observations by Gizis et al. (2002), who find it to 
be a double-lined spectroscopic binary (SB2) with a remarkable 
surface velocity of rotation v sin i = 30 km s We are dealing 
with a typical extremely active M dwarf in a multiple system: a 
short-period spectroscopic binary with a remote companion. 

The star HIP 47650 is an M3 flare star and a member of the 
Hyades stream according to Montes et al. (2001). Nidever et al. 
(2002) determined a “stable” radial velocity of +6.6 km s 1 for 
this star, precluding a detectable spectroscopic companion. We 
should therefore consider the possibility that this star is young. 
Both HIP 47650 and its brighter companion HIP 47620 lie sig¬ 
nificantly above the empirical main sequence in Figure 1. These 
stars are brighter in Mk s than the empirical main sequence from 
Henry et al. (2004), by 0.71 and 0.76 mag, respectively. Figure 2 
shows that both components are also brighter than their field 
counterparts in My versus V — K s axes as well, but by a smaller 
amount (0.49 mag in both cases). These photometric data suggest 
a large K-band excess, probably due to a young age similar to the 
age of the Pleiades. The substantial amount of X-ray radiation 
from HIP 47650 is accompanied by pronounced chromospheric 
activity. According to Wright et al. (2004) its average 5-value of 
Ca ii chromospheric activity of 3.2 is outside and above the nor¬ 
mal range where calibrated indices R(j K can be estimated. What 
remains puzzling is that two stars of similar mass in a binary sys¬ 
tem can be so different in chromospheric and coronal activity: HIP 
47650 is a dMe star with EWh b = 2.87 A (Rauscher & Marcy 
2006), whereas HIP 47650 has no emission in Ha (Gizis et al. 
2002). Since the difference in V — Iq between the components is 
only 0.2 mag, employing formally the age criterion in § 6.1 places 
the system in very narrow brackets of age just above 1 Gyr. How¬ 
ever, it seems unlikely that both chromospheric and X-ray activity 
in the more massive companion HIP 47620 waned so abruptly; the 
transition from dMe to normal M dwarfs is probably protracted 
and statistically uncertain. This binary system indicates that the 
evolution of surface rotation, which is a crucial factor in solar- 
and subsolar-mass dwarfs, may take different courses even for 
coeval, nearly identical stars. 

By far the brightest X-ray source in our collection is the BY 
Dra-type binary HIP 79607 (TZ CrB, orbital period 1.14 days). 
This example confirms that short-period spectroscopic binaries 
with evolved or solar-type primaries are the most powerful emit¬ 
ters among normal stars, surpassing single pre-main-sequence 
stars in X-ray luminosity by a factor of a few (Makarov 2003). 
The impressive flare activity on this star was investigated in detail 
by Osten et al. (2000). Its distant companion HIP 79551 separated 
by at least 13,000 AU is an M2.5 dwarf without any signs of 
chromospheric activity; we surmise that it should be older than 





























No. 1, 2008 


COMMON PROPER MOTION COMPANIONS 


573 


TABLE 5 

Velocities, Moving Groups, Activity, and Ages 


HIP/Name 

U 

V 

W 

SKG 

References 

EW( Ho) 

log Age 

15330. 

-71 

-47 

+16 

Hercules? 

i 


9.2 

15371. 

-70 

-46 

+16 

Hercules? 

i 


9.4 

47620. 

-36 

-13 

-14 

Hyades 

2 

<0 

>8.7 

47650. 

-35 

-13 

-14 

Hyades 

2 

2.87 

<8.9 

79607. 

-7 

-29 

+9 



0.64 

>9.5 

21482. 

-39 

-17 

-2 

Hyades? 

3 

1.0 

>9.6 

102409. 

-10 

-17 

-10 

BETAPIC 

4 

2.2 

7.0 

102141. 

-9 

-16 

-11 

BETAPIC 

4 

10.9 

7.0 

25278. 

-37 

-15 

8 

Hyades 

2 


8.5 

25220. 

-38 

-14 

7 

Hyades 

2 

-0.76 

8.7 

116215. 

-13 

-21 

-10 



-0.59 

8.8 

116191. 

-13 

-21 

-10 



1.98 

< 9.0 

4872. 

-32 

-16 

6 

Hyades 

2 



43587. 

-37 

-18 

-8 

Hyades 

2 




References.— (1) Soubiran & Girard 2005; (2) Montes et al. 2001; (3) Eggen 
1993; (4) Makarov 2007a. 


3 Gyr (§ 6.1), setting a lower bound on the age of the primary 
component. The primary, a F6+G0 pair of dwarfs (Frasca et al. 
1997), has a visual companion at 5.9", orbital period 852.8 yr 
(Tokovinin et al. 2006). This inner companion (erl CrB) may be 
responsible for the tight spectroscopic pair via the Kozai cycle, 
if the original orbits were not coplanar (§ 7.1). In this case, the 
substantial age of the system estimated from the CPM compan¬ 
ion is consistent with the timescale of dynamical evolution. The 
vertical velocity component with respect to the local standard of 
rest is +16 km s , assuming a standard solar velocity of W = 
+7 km s _1 . This places the TZ CrB multiple system in the older 
thin disk (Table 2, § 3), whose constituents spend roughly one- 
third of their lifetimes in the dense part of the Galactic disk. Thus, 
survival of the wide companion for longer than 3 Gyr is plausible. 

The star HIP 21482 appears to be another example of an ex¬ 
tremely active BY Dra-type spectroscopic binary in a hierarchical 
multiple system (Tokovinin et al. 2006). The inner spectroscopic 
pair has a orbital period of 1.788 days and is already circularized 
and rotationally synchronized (Montes et al. 1997). Its helio¬ 
centric motion (Table 5) is similar to the Hyades stellar kine¬ 
matic group (SKG), except for the deviating, small W velocity. 
The star was even suspected of having originated in the Hyades 
open cluster, which would fix its age at 600 Myr; in particular, 
Eggen (1993) suggested that it could belong to the extended halo 
of evaporated stars around this cluster. The exceptional chromo¬ 
spheric activity of the inner pair at log = —4.057 at the very 
tail of the distribution observed for nearby field stars (Gray et al. 
2003), may also indicate a young age. However, the remote com¬ 
panion WD 0433+270 is a cool DC white dwarf, and therefore 
the system can hardly be young. Bergeron et al. (2001) estimated 
a T e s = 5620 ± 110 K and cooling age of 4.07 ± 0.69 Gyr, an 
order of magnitude older than the Hyades. 

The star HIP 115147 (V368 Cep) is one of the nearest post- 
T Tauri stars. It is mistakenly identified as a RS CVn-type binary 
in the SIMBAD database, although, contrary to the previously 
discussed objects of this type, it is not a spectroscopic binary. Both 
its secondary companion at 11" and the newly discovered tertiary 
CPM companion LSPM J2322+7847 (Makarov et al. 2007) lie 
significantly above the main sequence in optical and infrared col¬ 
ors. The probable age of this system is only 20-50 Myr, and the 
high rate of rotation of the primary (with a period of 2.74 days; 
Kahanpaa et al. 1999) is obviously due to its youth. The origin of 


this post-TT triple system is an open issue, a high-velocity ejection 
from the Ophiuchus SFR being one of the possibilities considered. 

The pair of outstanding T Tauri stars HIP 102409 (AU Mic) 
and HIP 102141 (AT Mic) epitomize the class of very young, 
active X-ray emitters. They may be as young as 10 Myr, and 
both display the whole complement of stellar activity indica¬ 
tors. AU Mic has a nearly edge-on debris disk, and its remark¬ 
able X-ray luminosity is probably nurtured by the high rate of 
rotation with a period of surface spots of 4.847 days (Hebb et al. 
2007). Its distant companion, AT Mic, is a flare M4.5 dwarf and 
an extreme UV source. Both stars lie significantly above the 16 Myr 
isochrone in Figure 1. AT Mic has a somewhat poorly investigated 
companion LTT 8182 at 3.8", position angle 218° which is missing 
in the 2MASS survey and omitted in Table 1. Its Ha emission is 
also remarkably high (EW = 9.3 A; Scholz et al. 2007). AT Mic 
and AU Mic are separated by more than 0.2 pc in the sky plane, one 
of the largest separations found in this paper, and it is unlikely the 
two stars could be gravitationally bound. They will inevitably part 
ways in the future, along with other members of the dispersed 
BETAPIC stream ( Makarov 2007a). 

Both components in the CPM pair HIP 25278 and 25220 
are prominent X-ray sources (Hiinsch et al. 1999). The primary 
GJ 202, an F8 V star, is, however, more than 10 times weaker than 
its K4 companion GJ 201. Both stars have been assigned to the 
Hyades SKG by Montes et al. (2001). HIP 25278 appears to be a 
single star of slightly subsolar metallicity with an estimated age of 
5.6 Gyr (Nordstrom et al. 2004). Takeda & Kawanomoto (2005) 
determine a slightly higher [Fe/H] = 0.05 and find a surprisingly 
high content of lithium (EW = 0.094 A). Another unexplained 
characteristic of this star is its position below the main sequence in 
Figure 1. The moderate X-ray activity is accompanied by a notice¬ 
able Ca ii chromospheric signature at logRjjj, = —4.38 and ro¬ 
tation P/sin / = 4.1 days (Reiners & Schmitt 2003). Using the 
above value for logR^ and equation (7), we obtain an age of 
0.3 Gyr (Table 4), significantly less than Nordstrom et al.’s es¬ 
timate, and roughly consistent with the age of the Hyades open 
cluster. The CPM companion GJ 201, an active K4 V star, has 
a logP[[ K = —4.452 (Gray et al. 2003), and hence, an age of 
0.48 Gyr. It appears to be spectroscopically single. Its lithium 
abundance is low, however (Favata et al. 1997). Furthermore, 
the Ha line is in absorption according to Herbst & Miller (1989) 
placing this star in the realm of inactive, regular dwarfs. The high 
level of X-ray activity in this stars remains a mystery, because it 
cannot be explained just by the relative youth. Indeed, the distri¬ 
bution of X-ray luminosity between the companions appears to be 
inverted to that observed in the Hyades cluster (Stem et al. 1995), 
in that the F8 primary companion is below the lower envelope of 
Zx for its Hyades counterparts, while the secondary component, 
GJ 201, is roughly a factor of 10 more luminous than the average 
K dwarf in the Hyades and is comparable in X-ray emission to the 
brightest nonbinary F8-G0 Hyades members. 

The stars HIP 116215 and 116191, of spectral types K5 and 
M3.5, respectively, have space velocities similar to the Local 
stream of young stars (Montes et al. 2001). They may be as young 
as the Pleiades. The primary component (GJ 898) is single and its 
X-ray luminosity is similar to the average value for the Hyades 
late K-dwarfs. The secondary (GJ 897 AB) is a resolved visual 
binary (Mason et al. 2002) with an orbital period of 28.2 yr and 
a semimajor axis of 0.59", which probably explains why this 
M dwarf lies significantly above the main sequence, while the 
primary is quite close to it. A log R' HK = —4.486 from Gray et al. 
(2003) for HIP 116215 implies an age of 560 Myr, again similar 
to the age of the Hyades. The Ha line is in absorption for the 

















574 


MAKAROV & HENNESSY 


Vol. 687 


primary, but prominently in emission for the secondary (Gizis et al. 

2002) . This fact can be used to estimate the boundaries of Ha-age 
(§ 6.1), which yields log (Age) £ [8.1,9.0], in good agreement 
with the logRy K -age estimate. The remaining difficulty in the 
interpretation of this system is the unusual strength of X-ray emis¬ 
sion from HIP 116191, by far surpassing the levels observed for 
this age and spectral type in the Hyades. One may suspect that one 
of the visual companions in this binary is an undetected short- 
period spectroscopic binary. 

The star HIP 46843 is likely another representative of young 
X-ray emitters. A log = —4.234 from Gray et al. (2006) yield 
an age of 175 Myr, in fine agreement with the rotational age esti¬ 
mate 164 Myr from Bames (2007). Its M5.5 companion GJ 9301 B 
is undetected in X-rays. The young age of this system is confirmed 
by the Lx in Table 4 for HIP 46843, which is only slightly smaller 
than the typical luminosity of Pleiades members (~3 x 10 29 ; 
Stauffer & Hartmann 1986) of this spectral type. GJ 9301 B is 
therefore one of the youngest late M dwarfs in the solar neigh¬ 
borhood. Note that SIMBAD mistakenly provides an uncertain 
estimate of M v from (Reid et al. 1995) as a V magnitude. 

The star HIP 50564 of spectral class F6IV is remarkably ac¬ 
tive in X-ray but is unremarkable chromospherically (log = 
—4.749; Gray et al. 2003) and depleted in lithium. The low de¬ 
gree of activity points at an age of 1.9 Gyr. On the other hand, this 
star is a 6 Scuti-type variable and a fast rotator, v sin i = 17 km s~ 1 . 
It has a solar iron abundance, [Fe/H] = 0.09 from (Nordstrom et al. 
2004) and a space motion typical of the local young stream, 
(U, V , W) = (—14, -26,-12) km s _1 . The key to the mystery 
of its X-ray activity may be in a short-period, low-mass com¬ 
panion; indeed, Cutispoto et al. (2002) mention that the star is 
“reported as SB1 ” (single-lined spectroscopic binary) without 
providing further detail. Its M5-type CPM companion NLTT 
23781 separated by at least 0.5 pc is one of the discoveries in this 
paper. It was cataloged in (Lepine & Shara 2005; Salim & Gould 

2003) , but otherwise, this interesting object completely escaped 
the attention of observers. Its location in the HR diagram (Fig. 1) 
above the main sequence indicates a young age or binarity. Thus, 
this extreme system represents a mystery in itself. If it is indeed 
1.9 Gyr old, how could it survive at this separation having spent 
all the time in the thin disk, and why the remote companion is 
overluminous? 

The star HIP 59000 has a known CPM companion NLTT 
29580 separated by 4200 AU in the sky projection. Gray et al. 
(2006) report a substantial chromospheric activity of the primary, 
logR[j K = —4.341, which translates into a chromospheric age of 
0.29 Gyr. HIP 59000 is orbited by a low-mass companion, prob¬ 
ably a brown dwarf, for which we derive a first orbital solution in 
§ 7. This inner companion is not close enough to the primary 
(P ~ 5.1 yr) to account for the significant X-ray luminosity of 
the system. We think that either the primary is a yet-undetected 
short-period spectroscopic binary (in which case the astrometric 
companion may have a stellar mass), or the system is indeed 
fairly young. The remote CPM companion NLTT 29580, a M5.0 
star, is confirmed the photometric parallax from Reid et al. (2003) 
being in excellent agreement with the updated parallax of HIP 
59000 (45 mas). 

There is little doubt that the origin of the X-ray activity in HIP 
82817 is in the innermost component of this intriguing system of 
at least five stars, which drives the fast rotation of the second¬ 
ary. Indeed, the A component is orbited by a B component at 
Lab = 626 days, which is in fact some 50% more massive than 
the primary because it is a spectroscopic binary with a period 
Rb = 2.96553 days and a mass ratio of 0.9 (Mazeh et al. 2001), 
made of nearly identical M dwarfs. Both eccentricities are low, 


and the orbits are likely to be coplanar. The widest CPM com¬ 
panion LHS 429 is a M7 dwarf lying on the empirical main se¬ 
quence for late field dwarfs (Fig. 1). Mazeh et al. (2001) suggest 
an age of ^5 Gyr for the system. 

The F5 V star HIP 5799 and its G9 CPM companion GJ 9045 B 
are moderately metal-deficient ([Fe/H] = —0.3), kinematically 
belong to the thin disk population and have an estimated age of 
2.5 Gyr (Soubiran & Girard 2005). This age estimation is sup¬ 
ported by the moderate HK activity obtained by Gray et al. (2003) 
for the primary. The combination of a significant X-ray emission 
from the primary and the lack of such from the secondary, a mod¬ 
est rotational velocity of HIP 5799 (v sin/ = 4.4 km s Tokovinin 
1990) and the above age are puzzling. The peculiar location of 
HIP 5799 in the HR diagram (Fig. 1) much above the main se¬ 
quence and closer to the 16 Myr isochrone may give a clue. This 
star may be a yet undetected short-period spectroscopic binary 
seen almost face-on. 

The CPM pair HIP 86036 (=26 Dra) and HIP 86087 (=GJ 685) 
represents a rare case when both components are detected by 
ROSAT. Their X-ray luminosities differ by more than a factor of 
10 which may be the natural consequence of the difference in the 
sizes of their coronae, the primary being a GO V star and the dis¬ 
tant companion a M1 V dwarf. The primary is in fact a triple sys¬ 
tem where A component has a 76 yr orbiting B companion and a 
wide low-mass C companion at 12.2" (Tokovinin et al. 2006). 
Definitely, these resolved companions (not present in our sample) 
are not responsible for the enhanced X-ray emission from the in¬ 
ner system and we have to look for signs of a young age. Sur¬ 
prisingly, we find conflicting data. The primary star HIP 86036 is 
moderately metal-poor ([Fe/H] = — 0.18) and has an age of 8.4Gyr 
according to Soubiran & Girard (2005). Nordstrom et al. (2004) 
give an even older age of 11.5 Gyr for this star. However, the rota¬ 
tional age of the distant companion GJ 685 is only 435 ± 50 Myr 
at R rot = 18.6 days (Bames 2007). The Ha line is in absorption 
(EW = —0.4 A; Stauffer & Hartmann 1986), which only means 
that this Ml V star is probably older than 200 Myr. Another con¬ 
fusing detail comes from the Ca n HK line flux which is low for 
this type of star and the period of rotation (Rutten 1986). It is pos¬ 
sible that the fast rotation of GJ 685 is driven by extraneous agents, 
and the rotation age estimate is confused. To summarize, the origin 
of X-ray activity and the age of this system remains unknown. 

7. MULTIPLICITY 

At least 17 out of our 41 CPM systems contain inner binary or 
triple components. We have reasons to believe that some of the 
CPM components are still undiscovered binaries, especially those 
stars that are too luminous for their spectral type and age, and have 
enhanced rates of rotation and chromospheric activity. The rate of 
triple and higher order multiple systems among nonsingle stars 
in the Hyades is only 0.14 (Patience et al. 1998), significantly 
smaller than we find for CPM pairs (0.41). To some extent, the 
high-order multiplicity of very wide pairs can be explained by 
the higher mass of binary stars and therefore, better chances of 
survival in the course of dynamical interaction with other consti¬ 
tuents of the Galaxy. This argument may be particularly relevant 
for older CPM systems of extreme separations. On the other hand, 
there may be a more subtle reason for the abundance of hierarchical 
systems. The primary fragmentation of a prestellar molecular cloud 
and the secondary fragmentation during H 2 dissociation are likely 
to take place at two distinct hierarchical spatial scales (Whitworth 
& Stamatellos 2006). Of the three main models of low-mass star 
formation considered in that paper, the two-dimensional fragmen¬ 
tation triggered by supersonically colliding gas streams appears to 
be the most plausible scenario for wide companions in multiple 



No. 1, 2008 


COMMON PROPER MOTION COMPANIONS 


575 


systems. It predicts a wide range of initial orbital eccentricities 
and relative inclinations in such systems. 

Perhaps the system of CPM companions HIP 473 and 428 is 
the most important for empirical study of the Kozai-type evolu¬ 
tion of multiple systems. The latter star, an M2e dwarf, is known 
as the F components of the system ADS 48, where the primary 
star has a visual twin companion B (spectral type MO) separated 
by 6". The most interesting aspect of this system is that the mu¬ 
tual inclination of the B and F companions is ~80° according to 
the family of probable orbits computed by Kiyaeva et al. (2001). 
The eccentricity of the inner pair AB is probably between 0.2 
and 0.6. Therefore, ADS 48 may be a paragon of the Kozai evo¬ 
lution in progress, where the inner pair has not yet shrunk but re¬ 
mains in an elliptical orbit. Kiyaeva et al. (2001) note a probable 
inner tertiary companion, which may account for the total dy¬ 
namical mass higher by ^0.3 M 0 than what is expected from the 
spectral type. Furthermore, they note a slight variation in posi¬ 
tion of the A component with a period of 15 yr, possibly indi¬ 
cating another ~0.05 M @ companion. The A component lies on, 
or slightly below, the main sequence in Figure 2; thus, the hy¬ 
pothetical companions contribute little in the total luminosity. 
Anosova & Orlov (1991) pointed out that the probability of a hy¬ 
perbolic orbit for the F component appears to be greater than of 
an elliptical orbit. Such systems may be unstable in the long run. 
In the latter paper, it is proposed that ADS 48 is a member of the 
Hyades flow, of which we have quite a few representatives in our 
selection (Table 5). Stars in a kinematically coherent stream are 
more likely to be found in accidental slow passages near each 
other. The star HIP 428 is an emission-line M2 dwarf (Rauscher 
& Marcy 2006), indicating an upper limit on age of ~ 1 Gyr. This 
estimate is consistent with the upper envelope of the Hyades 
flow (Eggen 1998). We believe that Kiyaeva et al.’s suggestion 
that the F companion is physically bound to the AB pair with an 
orbital period of ~10 5 yr is more plausible in the light of recent 
astrometric data. 

The nearby star HIP 14555 (GJ 1054A), along with its op¬ 
tical companion HIP 14559, epitomizes the difficulties that arise 
in the reduction of Hipparcos data for visual multiple systems 
(§ 6.2). The improved solution for HIP 14555 from Fabricius & 
Makarov (2000a) is n = 55.5 ± 2.5 mas, ( p , a cos 8 , p, 6 ) = (—339, 
—121) mas yr -1 , which is close to the original results. The remain¬ 
ing inconsistency is that with the estimation by Henry et al. (2002) 
who inferred a distance of 12.9 pc based on their spectral type de¬ 
termination and the V magnitude specified in Hipparcos. This bi¬ 
ased estimate comes from the photometric data which seem to be 
too bright. Figure 2 depicts the HR diagram for both HIP 14555 
and the alleged CPM companion LTT 1477, with photometric 
data from (Weis 1993) and the same parallax of 55.5 mas as¬ 
sumed for both stars. The primary component lies significantly 
above the empirical main sequence and appears to match the 
16 Myr isochrone from (Siess et al. 2000). Despite the promi¬ 
nent Ha emission and X-ray activity, this star is not considered 
to be young. The apparent brightness excess is the consequence 
of unresolved binarity of HIP 14555. Indeed, according to Gizis 
et al. (2002), the star is double-lined spectroscopic binary (SB2). 

The star HIP 34052 (GJ 264) is the tertiary component of a 
well-known wide triple system, which also include the pair of 
solar-type stars GJ 9223 (A) and GJ 9223 (B), separated by 21" 
on the sky. By virtue of the high proper motion and brightness, the 
system has been included in the lists of nearby star for a long time, 
attracting considerable attention due to the possibility of testing 
the evolution of stellar gravity, temperature and chemical com¬ 
position in great detail. The spectroscopic investigation of com¬ 
ponents A and B by Chmielewski et al. (1991) found a common 


iron abundance of [Fe/H] = —0.27 ± 0.06 and effective tem¬ 
peratures 5870 ± 40 and 5290 ± 70 K, respectively. Somewhat 
different lithium abundances were determined for the two com¬ 
ponents, but both at the solar level or below it. These estimates, 
together with the Galactic orbit (eccentricity 0.31) and a negligible 
chromospheric activity from the Can lines, indicate an old system, 
probably representing the old disk. A theoretical ZAMS used by 
Chmielewski et al. (1991) adjusted to the location of the B compo¬ 
nent on a log Teff-Mboi diagram yielded a parallax of 68 mas. The 
trigonometric parallax of the system is close to 60 mas (Table 1). It 
may be suspected that the B component is too bright for the esti¬ 
mated Teff and metallicity. However, all three companions lie close 
to the main sequence with their photometric parameters in Table 1. 

Relatively little is known about the tertiary component, HIP 
34052. A robust astrometric solution was produced for this 
star in Hipparcos, without any indications of binarity or vari¬ 
ability. However, the Hipparcos proper motion (p, a cos 8 , p b ) = 
(—75.4,401.3) mas yr -1 differs significantly from the Tycho-2 
proper motion (—93.0, 395.3 mas yr -1 ; Hog et al. 2000). Since 
the latter is systematically more accurate in the presence of or¬ 
bital motion, Makarov & Kaplan (2005) included it in the list of 
astrometric binaries with variable proper motion. We further elab¬ 
orate on this star by applying a multiparameter orbital optimization 
algorithm designed for the Hipparcos Intermediate Astrometry 
Data (HIAD) (see, e.g., Makarov 2004). This algorithm, based on 
the Powell method of nonlinear iterative optimization, looks for 
the global minimum of the \ 2 statistics on abscissae residuals 
specified in the HIAD, corresponding to a certain combination of 
12 fitting parameters, including seven orbital elements and five 
astrometric corrections. The estimated orbital parameters are period 
P = 1501 days, inclination i = 180°, and apparent semimajor 
axis a o = 30.6 mas. The formal F-test on reduced \ 2 (0.933 after 
orbital adjustment) equals 1.0. The orbit is incomplete, because 
the period is longer than the time span of Hipparcos observations. 
Therefore, the orbital elements are fairly uncertain, and follow-up 
observations are needed to estimate the mass of the system. As¬ 
suming that the total mass of the system is 1.0 M 0 , the companion 
mass is only 0.2 M 0 , and the angular separation is about 150 mas 
(a = 2.6 AU). The companion may be possible to resolve with the 
Hubble Space Telescope ( HST ) or ground-based coronographic 
facilities. 

Astrometric binarity of the HIP 59000 (GJ 9387) K7 dwarf 
is advertised by its varying proper motion (Makarov & Kaplan 
2005). It is not a known spectroscopic binary; therefore, we at¬ 
tempted an unconstrained 12-parameter orbital solution for this 
star using the same algorithm described in the previous paragraph. 
A visual inspection of the HIAD data reveals that the orbital period 
is several years, and we are dealing with another incomplete orbit. 
As a consequence, the fitted parameters should be considered pre¬ 
liminary. We obtain a period P = 1854 days, apparent semimajor 
axis «o = 12 mas, To = JD 2,448,368, u = 61°, 12 = 53°, incli¬ 
nation i = 74°, and eccentricity e = 0.6. The updated parallax is 
n = 45.5 ± 0.7, which is close to the original Hipparcos parallax. 
The standard error of «o is about 2 mas, but the eccentricity is quite 
uncertain. Assuming a mass of 0.5 M 0 forthe visible primary, its ap¬ 
parent orbit on the sky leads to a total a = 2.34 AU and a second¬ 
ary mass of 0.063 M 0 . The expected radial velocity semiamplitude 
is 1.9 km s~ 1 . Thus, this newly discovered binary system contains 
a brown dwarf which may be only 290 Myr old (§ 6.1). 

The star HIP 75718 (GJ 586 A) is the primary in a system of 
at least four components (Tokovinin et al. 2006). The system is 
enshrouded in puzzles. The inner pair is both spectroscopic and 
astrometric (Duquennoy et al. 1992; Jancart et al. 2005) yielding 
a fairly detailed orbit. It consists of a K2 V dwarf (mass 0.74 M 0 ) 



576 


MAKAROV & HENNESSY 


Vol. 687 


and a later K dwarf (mass 0.49 M 0 ) in a 889.6 day orbit. The or¬ 
bit has an outstanding eccentricity of 0.9752 ± 0.0003, so that 
the separation between the companions at periastron is only about 
10 solar radii. The tertiary companion HIP 75722 (=GJ 586 B) is 
separated by 52" in the sky projection. It is another K2 V dwarf of 
the same mass as the primary of the inner pair (0.74 M 0 ). Hiinsch 
et al. (1999) assign the considerable X-ray flux detected by ROSAT 
to both A and B components, but in our opinion, it is the B com¬ 
ponent, surprisingly enough, that is responsible for the X-ray emis¬ 
sion (Table 4). The two companions are disparate in their Ca n line 
activity too, the A component being at log Ry K = —4.97, indicat¬ 
ing an old star, and B at log = —4.37 (Wright etal. 2004).For¬ 

mally, we would estimate the chromospheric age (§ 7.1) at 0.33 Gyr. 
Furthermore, the A and B components have different rates of ro¬ 
tation, /jot = 39.0 and 9.0 days, respectively. What could be 
the reason for the high activity and fast rotation of GJ 586 B? 
Tokovinin (1991) reported outlying radial velocity measures 
(spikes) for this star in otherwise constant series of observations 
and suggested that the B component can also be a high-eccentricity 
spectroscopic binary. If this is the case and the orbital period is of 
order a few days, the discrepant activity levels and age estimates are 
explained. However, Nidever et al. (2002) report a constant radial 
velocity from their extensive measurements. To confuse the mat¬ 
ter more, Nordstrom et al. (2004) specify a fairly low probability 
(0.285) of constant radial velocity from their 18 observations span¬ 
ning 6014 days. Finally, it is not clear whether the mysterious 
fourth component GJ 586 C (G 151-61) is physically associated 
with this triple or quadruple system. Its trigonometric parallax 
(Dahn et al. 1982) is tantalizingly close (n = 47 ±5 mas), but 
the proper motion is ~ 15% smaller. NOMAD supplies us with the 
following data for this star: position R. A. = 15 h 27 m 45.08 s , deck = 
—9°01'32.5" (J2000.0), and proper motion fj, = (30, —312) ± 
(2, 3) mas yr _1 . The available magnitudes are V = 15.41,/ = 
10.55, H = 9.92, and K s = 9.63. The smaller proper motion of 
this late M dwarf accounts for its absence in our NOMAD-based 
sample. Since the system appears to be genuinely old, it is doubt¬ 
ful that GJ 586 C can form a kinematic group with the brighter 
counterparts. 

7.1. Candidate Stars with Planets 

It is commonly accepted that planets can be present in binary 
stellar systems. The latest investigations in this area indicate that 
23% of candidate exosolar planetary systems also have stellar 
companions (Raghavan et al. 2006). Planets can form in stable 
circumbinary disks if the latter are large enough, so that the stel¬ 
lar binary and the distant planet form a dynamically stable hierar¬ 
chical system. In very wide CPM systems, we encounter a different 
hierarchical composition, when the remote tertiary companion has 
a stellar mass. Such planetary systems may be subject to the long¬ 
term oscillatory perturbations of inclination and eccentricity over a 
long time (several gigayears) because of the secular loss of or¬ 
bital energy known as the Kozai cycle. The eccentricity variation 
is especially important for the dynamical evolution of the inner 
planetary system. The Kozai-type variation is significant only if 
the tertiary companion has a different initial inclination from the 
inner orbit (Malmberg et al. 2007). If, for example, the initial in¬ 
clination of the tertiary is 76°, the planet will periodically de¬ 
scribe an orbit of e = 0.95. This high eccentricity entails very 
close periastron passages of the primary. Giant gaseous planets 
will be subject to the tidal friction at periastron passages quite 
similar to the mechanism suggested for stellar binaries (Kiseleva 
et al. 1998). The gradual loss of angular momentum may lock the 
planet on a high-eccentricity orbit, resulting in secular shrinkage 
of the orbit. The orbits of very short-period “hot Jupiters” 


(P <10 days) should be circularized similarly to tight spectro¬ 
scopic binaries. It is also important to note that the dynamical 
evolution due to the Kozai mechanism may be quite different for 
single planets and stable planetary systems even if the initial in¬ 
clination of the tertiary stellar companion is high. Innanen et al. 
(1997) point out that a system of four major solar system planets 
would remain stable and roughly coplanar in the presence of a 
distant companion on timescales much longer than the timescale 
of the Kozai cycle, owing to the mutual dynamical interaction 
between the planets. 

Our sample of CPM systems includes two candidate exoplanet 
hosts. The star HIP 43587 (GJ 324, 55 Cnc), which has a comov- 
ing companion LTT 12311 at a projected separation of 1050 AU, 
is a solar-type dwarf suspected of bearing a system of at least four 
planets (McArthur et al. 2004). One of them (55 Cnc d) is a super- 
Jupiter with a mass M sin/ = 3.9 Mj, aperiodof about5550days 
and a semimajor axis of nearly 6 AU. The other three suggested 
planets have masses between 0.037 and 0.83 Mj and periods rang¬ 
ing 2.8 to 44 days. The spectroscopically determined eccentrici¬ 
ties are all small (<0.1). There are a few conflicting clues about 
the age of the stellar components. The star 55 Cnc lies above the 
empirical main sequence by 0.55 mag according to (Butler et al. 
2006). Both this star and its companion GJ 324 B lie slightly 
above the main sequence in the Mk s versus V — K s diagram in 
Figure 1. The primary has a moderately enhanced metallicity 
[Fe/H] = 0.315, common among exosolarplanet hosts. However, 
the chromospheric activity of 55 Cnc is quite low, at logR(j K = 
—5.04 (Wright et al. 2004), which is in fact close to the mean 
chromospheric flux parameters for the most inactive field solar- 
type dwarfs (Gray et al. 2003). Wright et al. estimate a log (age) = 
9.81, and indeed, a similar age of 9.87 (7 Gyr) is obtained from the 
HK index. Montes et al. (2001) list 55 Cnc as a member of the 
populous Hyades stream (or kinematic group), based on its helio¬ 
centric velocity vector (see Table 5). In the light of recent investi¬ 
gations, the Hyades stream, originally believed to originate from 
the evaporating Hyades supercluster (Eggen 1993) of approxi¬ 
mately 700 Myr of age, incorporates stars of a wide range of age 
and chemical composition, indicating a curious phenomenon of 
dynamical alignment (Famaey et al. 2005). 

The star HIP 98767 (GJ 777 A) is similar to 55 Cnc in metal¬ 
licity ([Fe/H] = 0.213) and brightness excess (AM V = 0.66) ac¬ 
cording to Butler et al. (2006). Both this star and its distant CPM 
companion LTT 15865, separated by more than 2800 AU, lie 
slightly above the main sequence in Figure 1, but perfectly on the 
empirical main sequence in Figure 2. The primary star lies above 
the main sequence by 0.66 in absolute V magnitude according to 
(Butler etal. 2006) and is moderately metal-rich, [Fe/H] = 0.213. 
We do not know how to interpret these discordant photometric 
data, except to assume that there is some anomaly in the B and 
K bands. The primary is suspected to bear not just a single planet 
but a system of at least two planets (Vogt et al. 2005), a short-period 
HD 190360 c of mass M sin / « 0.06 M s and P = 17 days, and 
againa Jupiter-like HD 190360b of mass M sin / « 1.55 A/j and 
orbital period 2900 days. The eccentricities are «0 and 0.36, 
respectively. 

8. MOVING GROUPS AND STREAMS 

The solar neighborhood is permeated with SKGs, which are 
evident as number density clumps in the three-dimensional ve¬ 
locity space (Chereul et al. 1998). Since these streams are only 
loosely coherent kinematically and are not supposed to be grav¬ 
itationally bound, their existence poses a certain problem of 
dynamics. The Hyades stream figures prominently in our sample 
(Table 5). The Sun is located inside the Hyades stream today (but 



No. 1, 2008 


COMMON PROPER MOTION COMPANIONS 


577 


does not belong to it), so that any selection of the nearest stars 
will give preference to this SKG, as opposed to, for example, 
the Ursa Major SKG. Still, the large number of CPM systems in 
the Hyades stream is surprising for the following reason. Re¬ 
cent investigations indicate that the Hyades SKG is composed 
of stars and clusters of disparate ages and origins (Famaey et al. 
2005), contrary to the previous hypothesis that it is the result of 
dynamical evaporation of a massive open cluster. But if this stream 
is purely dynamical phenomenon, a kind of focusing taking hold 
of random unrelated objects, why do we find so many CPM pairs 
which are apparently generic? A dynamical agent sufficiently 
powerful to rearrange the local six-dimensional phase space of 
the Galaxy would probably accelerate the disruption of wide bi¬ 
naries rather than preserve them. Furthermore, the possible Hyades 
stream members present in our sample do not look like random 
field stars. Many of them have enhanced levels of chromospheric 
and X-ray activity indicative of moderately young age (~ 1 Gyr, 
roughly consistent with the age of the Hyades open cluster). On 
the other hand, the presence of weakly bound CPM pairs in the 
Hyades SKG is not consistent with the dynamical evaporation 
paradigm, because the latter assumes a dynamical relaxation 
and ejection event. A pair of M dwarfs like HIP 47620 and 47650 
(discussed in § 6.2) is unlikely to be thrown out of the Hyades 
cluster and remain intact. 

The star HIP 4872 and its distant companion V388 Cas (GI 51) 
are related to the Hyades stream by Montes et al. (2001). The latter 
companion is a well-known M5 flare dwarf of considerable X-ray 
luminosity (Table 4) and EUVactivity (Christian et al. 2001). A 
better age estimate can be obtained for the former companion, 
which is a flare Ml .5 dwarf. Rauscher & Marcy (2006) list this 
star as dMe with an Ha equivalent width of 2.0A. This yields 
an upper age limit of 280 Myr. The young age and the activity 
levels are consistent with this system being in the young core of 
the Hyades flow. 

The pair of CPM companions HIP 15330 and 15371 (£)' and 
( 2 Ret) is remarkable because both stars lie significantly below 
the ZAMS for Z = 0.01. They could be suspected to be metal- 
poor, but the iron abundance is only moderately low at [Fe/H] = 
—0.22 ± 0.05 according to del Peloso et al. (2000). The pair was 
originally assigned to the ( Her SKG, but since the latter star 
does not appear to belong to the moving group, it was renamed to 
( Ret SKG. Lately, Soubiran & Girard (2005) determined some¬ 
what smaller iron abundances (—0.34 and —0.30) for the two 
stars and assigned them to the Hercules SKG. The vertical veloc¬ 
ity is W = 16 km s , the maximum excursion from the plane 
z max = 0.31 kpc, and the eccentricity of the Galactic orbit e = 
0.26. Allen & Herrera (1998) propose to define the thick disk as 
either e > 0.3 or |z max | > 400 pc. Thus, the CPM pair in question 
does not qualify for the thick disk by any of these kinematic cri¬ 
teria. The origin of this system and its peculiar blueness remains 
an unresolved issue. 

9. DISCUSSION 

One of the most interesting results of this paper is that we find 
little, if any, presence of a thick-disk or halo population in the 
local sample of very wide binaries. The only CPM system that 
may belong to the thick disk is the WD+dM4.5 pair HIP 65877 
(DA3.5 white dwarf WD 1327-083) and LHS 353 (see Silvestri 
et al. 2002). This shows that even the widest pairs at separations 
greater than 1000 AU can survive for ~1 Gyr staying constantly 
in the thin disk of the Galaxy, despite numerous encounter and 
dynamical interaction events. This observation does not refute 
the dynamical analysis presented in § 3, because thick disk and 
halo stars are very rare in the solar vicinity, and our sample (based 


on bright stars in the Hippcircos catalog) is probably too small and 
incomplete to accommodate sufficient statistics. But if we boldly 
extrapolate this result to a wider part of the Galaxy, we con¬ 
clude that normal thin-disk, moderately young or very young, 
stars dominate wide CPM binary and multiple systems. Statis¬ 
tically, this is quite consistent with the estimation by Bartkevicius 
& Gudas (2002) on a larger sample of 804 Hippcircos visual sys¬ 
tems, who found that 92% of systems belong to the thin disk (and 
are mostly young to middle age), 7.6% to the thick disk, and much 
less than 1 % to the halo. Further inroads in this study can be made 
by collecting a larger volume-limited sample of very wide bina¬ 
ries and a comparison with a representative set of nearby field 
stars. We consider this paper as an initial step in this direction. 

Despite the considerable progress in recent years, chronology 
of solar-type stars is still in a rather pitiful state, and we find more 
evidence of this in the widely discrepant age estimates for a few 
CPM systems obtained with different methods. Although a sig¬ 
nificant fraction of CPM companions display enhanced chromo¬ 
spheric, X-ray, and EUV activity, only few are patently in the 
pre-main-sequence stage of evolution (e.g., AT and AU Mic), 
where these signs of activity and the high rate of surface rotation 
can be attributed to a very young age. The origin of activity in 
most of our CPM systems lies in short-period binarity of their 
components, i.e., in hierarchical multiplicity. An interesting con¬ 
nection emerges between the presence of wide companions and 
the existence of short-period binaries. The reason for abundant 
multiple systems may partly be purely dynamical, in that the 
chances of survival are higher for systems with an internal bi¬ 
nary because of the larger mass. An alternative astrophysical pos¬ 
sibility is that the original fragmentation of a star-fonning core 
takes place at various spatial scales and tends to produce multiple 
stellar systems, of which only hierarchical ones can survive for an 
appreciably long time. 

Apparently, the timescale of dynamical survival of wide com¬ 
panions (of order 1 Gyr) is sufficiently long compared to the time- 
scale of dynamical evolution of non-coplanar multiple systems 
(the Kozai cycle, § 7.1) for the latter to shape up the present-day 
systems. The existence of circularized spectroscopic binaries with 
periods less than a few days may be the direct consequence of the 
interaction with remote companions, followed by the tidal friction 
and loss of angular momentum (Eggleton & Kisseleva-Eggleton 
2006). Ultimately, the inner components will form a contact bi¬ 
nary and then merge. The existence of CPM multiple systems in a 
wide range of ages and separations will allow us to investigate this 
process in detail as it unfolds. Indeed, even in our sample of mod¬ 
est size we find examples of inner pairs of intermediate periods 
and large eccentricities, which are apparently evolving toward the 
tidally circularized state. The Kozai-type mechanism can affect 
the dynamical stability and composition of planetary systems. 
We find two stars in our sample with multiple planets (55 Cnc 
and GJ 777 A), and both have interesting dynamical properties 
very much unlike our solar system. 


The research described in this paper was in part carried out at 
the Jet Propulsion Laboratory, California Institute of Technology, 
under a contract with the National Aeronautics and Space Admin¬ 
istration (NASA). This research has made use of the SIMBAD 
database, operated at CDS, Strasbourg, France; and data products 
from the 2MASS, which is a joint project of the University of 
Massachusetts and the Infrared Processing and Analysis Center, 
California Technology Institute, funded by NASA and the Na¬ 
tional Science Foundation (NSF). 



578 


MAKAROV & HENNESSY 


Allen, C., & Herrera, M. A. 1998, Rev. Mex. AA, 34, 37 
Anosova, J. P., & Orlov, V. V. 1991, A&A, 252, 123 
Balega, I. I., et al. 2006, A&A, 448, 703 
Barnes, S. A. 2007, ApJ, 669, 1167 

Bartkevicius, A., & Gudas, A. 2002, Baltic Astron., 11, 153 

Bergeron, R, Legett, S. K., & Ruiz, M. T. 2001, ApJS, 133, 413 

Bessel, M. S. 1990, A&AS, 83, 357 

Butler, R. P„ et al. 2006, ApJ, 646, 505 

Chereul, E„ et al. 1998, A&A, 340, 384 

Chiba, M„ & Beers, T. C. 2000, AJ, 119, 2843 

Chmielewski, Y., et al. 1991, A&A, 247, 368 

Christian, D. J„ et al. 2001, AJ, 122, 378 

Cutispoto, G., et al. 2002, A&A, 384, 491 

Dahn, C. C„ et al. 1982, AJ, 87, 419 

-. 2002, AJ, 124, 1170 

del Peloso, E. F„ da Silva, L„ Porto de Mello, G. F. 2000, A&A, 358, 233 
Duquennoy, A., et al. 1992, A&A, 254, 13 
Eggen, O. J. 1993, AJ, 106, 1885 
. 1995, AJ, 110, 1749 
. 1998, AJ, 116, 284 

Eggleton, P. R, & Kisseleva-Eggleton, L. 2006, Ap&SS, 304, 75 
ESA. 1997, The Hipparcos and Tycho Catalogues (ESA SP-1200; Noordwijk: 
ESA) 

Fabricius, C„ & Makarov, V. V. 2000a, A&AS, 144, 45 

-. 2000b, A&A, 356, 141 

Famaey, B., et al. 2005, A&A, 430, 165 

Favata, F., Micela, G., & Sciortino, S. 1997, A&A, 322, 131 

Frasca, A., Catalano, S., & Mantovani, D. 1997, A&A, 320, 101 

Gizis, J. E., Reid, I. N„ & Hawley, S. H. 2002, AJ, 123, 3356 

Gould, A., & Chaname, J. 2004, ApJS, 150, 455 

Gray, R. O., et al. 2003, AJ, 126, 2048 

-. 2006, AJ, 132, 161 

Henry, T. J., et al. 2002, AJ, 123, 2002 

-. 2004, AJ, 128, 2460 

Herbst, W., & Miller, J. 1989, AJ, 97, 891 
Hebb, L., et al. 2007, MNRAS, 379, 63 
Hog E., et al. 2000, A&A, 355, L27 
Hiinsch, M., et al. 1999, A&AS, 135, 319 
Innanen, K. A., et al. 1997, AJ, 113, 1915 
Jancart, S., et al. 2005, A&A, 442, 365 
Kahanpaa, J., et al. 1999, A&A, 350, 513 
Kiseleva, L. G., et al. 1998, MNRAS, 300, 292 
Kiyaeva, O. V., et al. 2001, Astron. Lett., 27, 391 
Koen, C., et al. 2002, MNRAS, 334, 20 
Lepine, S., & Bongiomo, B. 2007, AJ, 133, 889 
Lepine, S., & Shara, M. 2005, AJ, 130, 1483 
Makarov, V. V. 2003, AJ, 126, 1996 

-. 2004, ApJ, 600, L71 

-. 2007a, ApJS, 169, 105 

-. 2007b, ApJ, 658, 480 

Makarov, V. V., & Kaplan, G. H. 2005, AJ, 129, 2420 

Makarov, V. V., Oiling, R. P., & Teuben, P. J. 2004, MNRAS, 352, 1199 

Makarov, V. V., & Urban, S. 2000, MNRAS, 317, 289 

Makarov, V. V., et al. 2007, ApJ, 668, LI55 

Maikov, O. Y., et al. 2006, A&A, 446, 785 

Malmberg, D„ Davies, M. B„ & Chambers, J. E. 2007, MNRAS, 377, LI 


REFERENCES 

Mason, B. D., et al. 2001, AJ, 122, 3466 
-. 2002, AJ, 124, 2254 

Maxted, P. F. L., Marsh, T. R., & Moran, C. K. J. 2000, MNRAS, 319, 305 
Mazeh, T„ et al. 2001, MNRAS, 325, 343 
McArthur, B. E., et al. 2004, ApJ, 614, L81 
Monet, D. G„ et al. 2003, AJ, 125, 984 
Montes, D„ et al. 1997, A&AS, 125, 263 

-. 2001, A&A, 379, 976 

Nidever, D. L., et al. 2002, ApJS, 141, 503 
Nordstrom et al. 2004, A&A, 418, 989 
Osten, R. A., et al. 2000, ApJ, 544, 953 
Patience, J., et al. 1998, AJ, 115, 1972 
Phan-Bao, N„ & Bessell, M. S. 2006, A&A, 446, 515 
Poveda, A., et al. 1994, Rev. Mex. AA, 28, 43 
Raghavan, D., et al. 2006, ApJ, 646, 523 
Rauscher, E„ & Marcy, G. W. 2006, PASP, 118, 617 
Reid, I. N„ & Cruz, K. L. 2002, AJ, 123, 2806 
Reid, I. N„ Hawley, S. L., Gizis, J. E. 1995, AJ, 110, 1838 
Reid, I. N„ Kilkenny, D„ & Cruz, K. L. 2002, AJ, 123, 2822 
Reid, I. N„ et al. 2003, AJ, 126, 3007 
Reiners, A., & Schmitt, J. H. M. M. 2003, A&A, 398, 647 
Rosello, G„ et al. 1987, A&AS, 67, 157 
Rutten, R. G. M. 1986, A&A, 159, 291 
Salim, S., & Gould, A. 2003, ApJ, 582, 1011 
Scholz, A., et al. 2007, ApJ, 662, 1254 
Seymour, D. M., et al. 2002, AJ, 123, 1023 
Siess, L., Dufour, E., & Forestini, M. 2000, A&A, 358, 593 
Silvestri, N. M„ Oswalt, T. D„ & Hawley, S. L. 2002, AJ, 124, 1118 
Soderblom, D. R., Duncan, D. K., & Johnson, D. R. H. 1991, ApJ, 375, 722 
Soubiran, C., & Girard, P. 2005, A&A, 438, 139 
Stauffer, J. R., & Hartmann, L. W. 1986, ApJS, 61, 531 
Stem, R. A., Schmitt, J. H. M. M„ & Kahabka, P. T. 1995, ApJ, 448, 683 
Takeda, Y„ & Kawanomoto, S. 2005, PASJ, 57, 45 
Tokovinin, A., et al. 2006, A&A, 450, 681 
Tokovinin, A. A. 1990, Pis’ma Astron. Zh., 16, 52 
. 1991, A&AS, 91, 497 

Torra, J., Fernandez, D., & Figueras, F. 2000, A&A, 359, 82 
Voges, W„ et al. 1999, A&A, 349, 389 

-. 2000, IAU Circ., 7432 

Vogt, S. S., et al. 2005, ApJ, 632, 638 

Weinberg, M. D„ Shapiro, S. L., & Wasserman, I. 1987, ApJ, 312, 367 
Weis, E. W. 1988, AJ, 96, 1710 
. 1991, AJ, 101, 1882 
. 1993, AJ, 105, 1962 
. 1996, AJ, 112, 2300 
West, A. A., et al. 2006, AJ, 132, 2507 
-. 2008, AJ, 135, 785 

Whitworth, A. P., & Stamatellos, D. 2006, A&A, 458, 817 
Wright, J. T., et al. 2004, ApJS, 152, 261 

Zacharias, N„ McCallon, H. L., Kopan, E., & Cutri, R. M. 2006, Extending the 
ICRF into the Infrared: 2MASS-UCAC Astrometry (26th IAU Meeting, 
Joint Discussion), 16 

Zacharias, N. Urban, S. E., Zacharias, M. I., Wycoff, G. L„ Hall, D. M., Monet, 
D. G., & Rafferty, T. J. 2004a, AJ, 127, 3043 
Zacharias, N„ et al. 2004b, BAAS, 205, 4815