Masses and angular momenta of contact binary stars

Abstract

Results are presented on component masses and system angular momenta for over a hundred low-temperature contact binaries. It is found that the secondary components in close binary systems are very similar in mass. Our observational evidence strongly supports the argument that the evolutionary process goes from near-contact binaries to A-type contact binaries, without any need of mass loss from the system. Furthermore, the evolutionary direction of A-type into W-type systems with a simultaneous mass and angular momentum loss is also discussed. The opposite direction of evolution seems to be unlikely, since it requires an increase of the total mass and the angular momentum of the system. Comment: 4 pages, 4 figures, accepted in MNRAS

Full text

Page 1

arXiv:astro-ph/0606715v1 29 Jun 2006
Mon. Not. R. Astron. Soc. 000, 000–000 (2006)Printed 5 February 2008 (MN LATEX style file v2.2)
Masses and angular momenta of contact binary stars
K. D. Gazeas⋆and P. G. Niarchos⋆
Department of Astrophysics, Astronomy and Mechanics, Faculty of Physics, University of Athens, GR-157 84, Zografos, Athens, Greece
Accepted 2006 April 13. Received 2006 April 12; in original form 2006 March 24
ABSTRACT
Results are presented on component masses and system angular momenta for over
a hundred low-temperature contact binaries. It is found that the secondary components
in close binary systems are very similar in mass. Our observational evidence strongly
supports the argument that the evolutionary process goes from near-contactbinaries to
A-type contact binaries, without any need of mass loss from the system. Furthermore,
the evolutionary direction of A-type into W-type systems with a simultaneous mass
and angular momentum loss is also discussed. The opposite direction of evolution
seems to be unlikely, since it requires an increase of the total mass and the angular
momentum of the system.
Key words: binaries: close - binaries: eclipsing - stars: evolution
1 INTRODUCTION
In his two seminal papers, Lucy (1968a,b) not only showed
that two stars can exist in an envelope of a common equipo-
tential and thus resolved the overall hitherto unexplainable
properties of W UMa-type stars, but also very clearly indi-
cated that such contact binaries must have very dissimilar
components. As pointed out by Hazlehurst (1970), evolution
can create this dissimilarity.
The evolutionary state of contact binary stars remains
unclear because their spectra cannot be analyzed for abun-
dances due to the extreme broadening and blending of spec-
tral lines. Indirect information though, such as their progres-
sively increasing numbers with age in old open (Rucinski
1998) and globular (Rucinski 2000) clusters, as well as
the kinematic characteristics (Guinan & Bradstreet 1988;
Bilir et al. 2005), very strongly suggest an advanced age of
> 2 Gyr.
Recently, Stepien (2004) has developed a model with
the currently less massive component being the more evolved
one. Such a model is conceptually very close to that used to
explain the semi-detached Algols. In his model, the current
secondary (less-massive) components must be in some cases
very low in mass to explain systems like AW UMa or SX Crv
(Rasio 1995).
In this Letter we present a summary of results on the
component masses (for 112 systems) and system angular
momenta (for 93 systems) for low-temperature contact bi-
naries. The sample used was collected mainly from the list of
contact binaries defined by Kreiner et al. (2003). Half of the
⋆e-mail: kgaze@physics.auth.gr; pniarcho@phys.uoa.gr
systems have solutions published in the frame of the W UMa
project (papers I-VI) (Kreiner et al. 2003; Baran et al.
2004; Zola et al. 2004; Gazeas et al. 2005; Zola et al. 2005;
Gazeas et al. 2006). All the rest were collected from the lit-
erature, the physical parameters of which have been deter-
mined accurately, using photometric light curves and ra-
dial velocity measurements for both components. We also
included the near-contact binaries (NCBs) and the de-
tached close binaries (DCBs) listed in Tables 2 and 3 of
Yakut & Eggleton (2005), in order to compare their physi-
cal parameters with those of our sample.
We had to exclude the cases, where the third light and
the low-amplitude light variation give unreliable solutions,
such as V2150 Cyg, V899 Her, HT Vir, BL Eri and GO Cyg.
In some cases, close binaries in triple systems have led to
spurious solutions and for this reason they were excluded
from our sample. Other cases with third light contribution
do not seem to produce any problem. Such systems, which
are members in multiple systems, have better geometrical
configuration and usually have very good spectroscopic de-
termination of the third light contribution, allowing accurate
determination of the orbital and physical parameters. Two
systems, V351 Peg and V402 Aur were very probably in-
correctly classified as W-type binaries, although the shape
of their light curves does not support such a classification.
Both systems have equal minima in their light curves, mak-
ing it difficult to distinguish them from each other. Since
their physical parameters (masses and periods) were closer
to those of A-type binaries, they were classified as A-type
systems.
Only recently, good spectroscopic data has become
available for more than a hundred contact binaries. Since

Page 2

2K. D. Gazeas and P. G. Niarchos
-0.7?-0.6?-0.5?-0.4?-0.3?-0.2?-0.1?0.0?
0.0?
0.5?
1.0?
1.5?
2.0?
2.5?
3.0?
M=0.5Mo?
M=1Mo?
?
?
Mass?
logP?
M?1?
M?2?
low-q M?1?
low-q M?2?
Figure 1.
all 112 contact binaries in our sample. The masses of the primary
components are plotted with full circles, while those of secondaries
with open circles. Triangles represent the low-q systems, with
secondary components of very small mass. The absence of systems
with periods between the values -0.30 and -0.23 (0.5 and 0.6 days)
is obvious. Note that the masses of the secondaries are between
the limits of 0 − 1M⊙, while the masses of primaries gradually
increase, almost proportional to logP.
The mass distribution for the two components of
even small-mass secondaries can be observed spectroscopi-
cally in contact systems, most objects have been analyzed as
double-lined binaries bypassing any need of inferences based
on single-lined data or solely on sometimes highly unreliable
photometric elements (particularly mass ratios).
In our study we consider component ”1” as the more
massive one. Our assumption is based on the double-lined
spectroscopic observations, where the mass ratio is taken as
q = M2/M1 < 1.
2 MASSES
In Figure 1 we present the distribution of the component
masses versus the orbital period. It seems that the secondary
components in all systems are very similar in masses, regard-
less of the orbital period. Masses of secondaries are between
the values of 0 and 1 M⊙. The mean value of the mass of
the secondaries is 0.45M⊙, while the primaries have masses
between 0.5 and 2.5M⊙ (only HV UMa has M1 = 2.8M⊙).
The same pattern of distribution appears when the masses of
A and W-type systems are plotted in separate figures (Fig-
ures 2a and 2b). In this case, the average mass of A-type
secondaries is equal to 0.41M⊙ and that of W-type equal to
0.49M⊙. It is remarkable to see the similarity of the mass dis-
tribution of secondary components with that of white dwarfs
(Madej et al. 2004). According to Stepien (2004), mass ex-
change is taking place in the majority of contact binaries
and the secondaries are helium-rich objects. In this case the
masses of the secondary components are expected to be sim-
ilar or smaller than those of white dwarfs, which could have
been grown as cores of isolated stars.
Another interesting feature shown in Figure 1 as well
as in the upper and lower panels of Figure 2 is that there
-0.7?-0.6?-0.5?-0.4?-0.3?-0.2?-0.1?0.0?
0.0?
0.5?
1.0?
1.5?
2.0?
2.5?
3.0?
?
?
Mass (A-type)?
logP?
A-type M?1?
A-type M?2?
low-q M?1?
low-q M?2?
-0.7?-0.6?-0.5?-0.4?-0.3?-0.2?-0.1?0.0?
0.0?
0.5?
1.0?
1.5?
2.0?
2.5?
3.0?
?
?
Mass (W-type)?
logP?
W-type M?1?
W-type M?2?
Figure 2. The mass distribution of the two components of 60
A-type contact binaries (upper panel) and 52 W-type contact
binaries (lower panel) of our sample. Triangles represent the low-
q systems, as in Figure 1.
is a total absence of systems with periods between 0.5 and
0.6 d. This gap is rather caused by the selection effect of
our sample of contact binaries. Many contact binaries with
equal minima and periods close to 12 h are difficult to be
observed and can be mistaken as pulsating variables, with a
period of 6 h (i.e. β Lyrae variables), or remain unclassified.
A recent study (Rucinski 2002) has also showed that many
contact binaries are still undetected.
In our sample, all W-type systems have orbital periods
shorter than 0.5 d, while the A-type systems can have all
possible periods in the range considered, with a small pref-
erence in large values.
Seven systems (CK Boo, FP Boo, SX Crv, GR Vir,
TZ Boo, AW UMa and FG Hya) (especially AW UMa,
SX Crv and TZ Boo) are plotted with triangles in all fig-
ures, as they are low-q systems with very low-mass secon-
daries (M2 < 0.17M⊙). In these systems the rotational an-
gular momentum is mostly ”absorbed” from the primary
component and plays a significant role on the total angular

Page 3

Contact binary stars3
-0.7?-0.6?-0.5?-0.4?-0.3?-0.2?-0.1?0.0?
0.0?
0.5?
1.0?
1.5?
2.0?
2.5?
3.0?
3.5?
4.0?
?
?
Total mass?
logP?
A-type ?
W-type ?
low-q CBs?
NCBs and DCBs?
Figure 3. The total mass distribution of 112 contact binaries
(full circles for A-type and open circles for W-type), 25 near con-
tact and 11 detached binaries (open squares). Triangles represent
the low-q contact binaries, as in Figure 1.
momentum of the system. According to Rasio (1995) these
very low-q systems cannot exist, since Jorb> 3Jspin.
There is a co-existence of A and W-type systems with
periods between 0.3 and 0.5 d. All systems with P < 0.3 d
are of W-type and all with P > 0.6 d are of A-type.
A very interesting feature is shown in Figure 3, where
the total mass of the above systems is plotted versus the pe-
riod. It is obvious that A-type systems are in general more
massive than W-type ones. For comparison, in the same
graph, we plotted the total mass of our CBs with the to-
tal mass of the short-period (< 1 d) NCBs and DCBs, taken
from Yakut & Eggleton (2005). One can see that most of
the NCB and DCB systems have total mass similar to that
of A-type contact binaries but larger than that of W-type
contact binaries.
These observational facts strongly support the evolu-
tionary progress from near-contact into contact configura-
tion of A-type, without the need of mass loss from the
system, as proposed by Yakut & Eggleton (2005). Further
progress in this direction, may transform the A-type sys-
tems into W-type, with a simultaneous mass loss, or can
lead the A-type systems to become binaries with extreme
small mass ratio. The opposite direction would require the
total mass to increase, which is unreasonable.
3 ANGULAR MOMENTA
In Figure 4 we present a plot of the orbital angular momen-
tum versus the orbital period. In this plot we see that A-type
systems generally tent to have larger angular momenta than
the W-type systems. On the top of each point in the graph,
a vertical line is added, representing the amount of the spin
angular momentum of the two components. The sum of the
spin and orbital angular momenta will give the total angular
momentum of each system.
In this way it can be seen that the angular momentum
of close binaries can be studied with either the orbital or the
-0.7?-0.6?-0.5?-0.4?-0.3? -0.2?-0.1?0.0?
0.0?
0.1?
0.2?
0.3?
0.4?
0.5?
0.6?
0.7?
0.8?
0.9?
1.0?
1.1?
1.2?
1.3?
1.4?
?
?
Orbital angular momentum?
logP?
A-type ?
W-type ?
low-q CBs?
Figure 4. The orbital angular momentum distribution of the 93
contact binaries of our sample (full circles for A-type and open
circles for W-type). Vertical lines above each point represent the
amount of the spin angular momentum of the two components,
which, if added to the orbital, will represent the total angular
momentum of each system. Triangles represent the low-q contact
binaries, as in Figure 1.
total angular momentum. The orbital angular momentum is
not affected from observational or modeling errors, but only
from masses and orbital periods. On the other hand, the
spin angular momentum is affected from the errors in mass,
radius and from the assumption we make for the radius of
gyration, which is still under investigation (Rasio 1995). For
example, a slightly smaller radius of gyration would shift the
seven low-q systems of our sample in a stable state.
Figure 4 shows that formation of W-type systems from
A-type is possible to be done, but the opposite direction
is not strongly supported. Evolution from A-type to W-
type systems seems to occure with a simultaneous mass
and angular momentum loss, unless the evolutionary direc-
tion, after the NCBs evolve to contact systems, follows two
separate tracks. However, we cannot exclude the possibil-
ity that some A-type systems with small angular momen-
tum have evolved from W-type systems, while others (with
large angular momentum) have evolved directly from NCBs
(Yakut & Eggleton 2005).
4 DISCUSSION
Do both A and W-type contact binaries have the same ori-
gin? Is the one type progenitor of the other? The above
questions are still open for investigation.
The main result of the present study, taking into ac-
count the masses and angular momenta of contact binaries,
is that the W-type systems cannot produce A-type binaries,
since angular momentum and mass cannot be added to a
system, but only lost from it. It seems more reasonable that
A-type systems evolve to W-type systems by loosing mass
and angular momentum. A similar evolutionary direction
from long to short period binaries (Bilir et al. 2005) showed
that systems with longer periods are kinematically young

Page 4

4 K. D. Gazeas and P. G. Niarchos
(age 2 Gyr) in contrast to those with shorter periods (age 8
Gys).
An interesting result extracted from our plots is that
the secondary components in close binary systems are very
similar in mass. Scatter of the points in our plots is mainly
due to inaccurate photometric and spectroscopic solutions
and/or a possible undetected third light contribution. Since
the best of the available data is used in our sample, only
the third light could produce a problem and this is why
some solutions are excluded from our sample. A very recent
study of Pribulla & Rucinski (2006) about the formation of
contact binaries in multiple systems, suggests that a large
percentage of close binaries is formed into triple and multi-
ple systems. In such a case, a small amount of unreliability
is added to all the solutions, if they are affected from an
undetected third light.
Our observational evidence (Figures 1-4) strongly sup-
ports the argument that the evolutionary process is from
NCBs to A-type contact binaries, without any need of mass
loss from the system. The next step in this scenario may lead
either to a transformation of A-type to W-type systems with
a simultaneous mass and angular momentum loss, or to A-
type systems with extremely low mass ratio. These systems
will eventually merge into a single, fast-rotating object. The
opposite direction of evolution seems to be impossible, since
it requires an increase of the total mass and angular mo-
mentum of the system.
5 ACKNOWLEDGMENTS
The authors gratefully acknowledge Professor S. Rucin-
ski for his valuable help during the preparation of the
manuscript and for many insightful discussions and new
ideas about contact binary structure and evolution, as well
as the anonymous referee for useful suggestions, which im-
proved the article. This project was supported by the Spe-
cial Account for Research Grants No 70/3/7187 (HRAK-
LEITOS) and 70/3/7382 (PYTHAGORAS) of the National
and Kapodistrian University of Athens, Greece.
REFERENCES
Baran, A., Zola, S., Rucinski, S. M., Kreiner, J. M., Siwak,
M., Drozdz, M., 2004, Acta Astron., 54, 195 (Paper II).
Bilir, S., Karatas, Y., Demircan, O., Eker, Z., 2005, MN-
RAS, 357, 497
Gazeas, K. D. et al., 2005, Acta Astron., 55, 123 (Paper
IV).
Gazeas, K. D., Niarchos, P., Zola, S., Kreiner, J. M., Rucin-
ski, S. M., 2006, Acta Astron., 56, 127 (Paper VI).
Guinan, E. F. & Bradstreet, B. H. 1988, in Dupree A.K.,
Lago M.T.V.T., eds, NATO Adv. Sci. Inst., Ser.C, Vol.
241, Formation and Evolution of Low Mass Stars, Kluwer,
Dordrecht, p. 345
Hazlehurst, J. 1970, MNRAS, 149, 129
Kreiner, J. M. et al., 2003, A&A, 412, 465 (Paper I)
Lucy, L. B. 1968a, ApJ, 151, 1123
Lucy, L. B. 1968b, ApJ, 153, 877
Madej, J., Nalezyty, M., Althaus, L.G., 2004, A&A, 419, 5
Pribulla, T. & Rucinski, S. M., 2006, AJ, preprint
(astro-ph/0601610)
Rasio, F.A., 1995, ApJ, 444, L41
Rucinski, S. M., 1998, AJ, 116, 2998
Rucinski, S. M., 2000, AJ, 120, 319
Rucinski, S. M., 2002, PASP, 114, 1124
Stepien, K., 2004, in Dupree A.K., Benz A.O., eds, Proc.
IAU Symp. 219, Stars as Suns: Activity, Evolution and
Planets. Astron. Soc. Pac., San Fransisco, p. 967
Yakut, K., Eggleton, P.P., 2005, ApJ, 629, 1055
Zola, S. et al., 2004, Acta Astron., 54, 299 (Paper III).
Zola, S. et al., 2005, Acta Astron., 55, 389 (Paper V).