Download Mass loss in semi-detached binaries

Survey
yes no Was this document useful for you?
   Thank you for your participation!

* Your assessment is very important for improving the workof artificial intelligence, which forms the content of this project

Document related concepts

Planetary nebula wikipedia , lookup

P-nuclei wikipedia , lookup

Nucleosynthesis wikipedia , lookup

White dwarf wikipedia , lookup

Standard solar model wikipedia , lookup

Astronomical spectroscopy wikipedia , lookup

Hayashi track wikipedia , lookup

Stellar evolution wikipedia , lookup

Star formation wikipedia , lookup

Main sequence wikipedia , lookup

Transcript
M A S S L O S S IN S E M I - D E T A C H E D
BINARIES*
K. D . A B H Y A N K A R
Centre of Advanced Study in Astronomy, Osmania University, Hyderabad, India
(Received 7 June, 1983)
Abstract. Pre-Main-Sequence contracting objects, post-Main-Sequence expanding stars and mass-losing
components of semi-detached systems all occupy more or less the same region in the conventional
H-R-diagram. We make a transformation to variables A (logL) and A (log Te), where A is the difference
between the observed quantity, logL or log T~, and the value of that quantity which a star of the same mass
would have on the empirical Main Sequence. It is demonstrated that a plot between the new variables clearly
separates the mass-losing stars from other objects which is essentially an effect of the increasing abundance
of helium relative to hydrogen.
1. Introduction
H-R-diagram is a powerful tool for studying the evolution of stars mainly because of
the general validity of the Russell-Vogt theorem for stars in hydrostatic and thermal
equilibrium. Given the total mass and run of chemical composition within the star, one
can obtain a unique configuration with specific values of effective temperature and
luminosity which fix the position of the star in the H-R-diagram. The zero-age Main
Sequence is characterised by homogeneous chemical composition throughout the star
and energy production by hydrogen burning in the core. Since it is possible to calculate
the change of chemical composition caused by thermonuclear reactions one can trace
the track of the star in the H-R-diagram as it evolves away from the Main Sequence.
In general, the star expands and the effective temperature drops which causes it to move
towards right into the giant and subgiant region of the H-R-diagram. There is also an
increase in luminosity due to the increase in mean molecular weight caused by the
conversion of hydrogen into helium in the central core.
The same region of giants and subgiants is also occupied by the pre-Main-Sequence
contracting stars which are not in strict hydrostatic equilibrium. They are distinguished
from the post-Main-Sequence objects by other characteristics such as: surrounding
nebulosities, infrared excess, presence of emission lines in spectra, flare activity, and
overabundance of lithium.
There is a third group of objects which also occupy the giant and subgiant region of
the H-R-diagram, viz. the evolved components of semi-detached binaries (Kopal,
1956). It is now well-established that they have lost a large amount of their original mass,
particularly from their hydrogen rich envelopes. As it is desirable to distinguish them
from the other two groups we give here a method of doing so by means of a transformation of variables.
* Paper presented at the Lcmbang-Bamberg IAU Colloquium No. 80 on 'Double Stars: Physical Properties
and Generic Relations', held at Bandung, Indonesia, 3-7 June, 1983.
Astrophysics and Space Science 99 (1984) 355-361. 0004-640X/84/0992-0355501.05.
9 1984 by D. Reidel Publishing Company.
356
K. D. ABHYANKAR
2. T h e C h o i c e o f V a r i a b l e s
The clue to the choice of variables comes from R CMa-systems which contain
secondaries of very low mass. In the case of R CMa itself the evolved secondary has
a mass of 0.2M o (Sahade, 1963; Radhakrishnan et aL, 1983) which corresponds to a
Main-Sequence star of spectral type M5 that would have a luminosity 3.5 x 10- 3 Lo
and effective temperature of 2800 K. However, its actual luminosity of 0.55L o and
effective temperature of 4500 K make it appear like a Main-Sequence star of spectral
type G8. Thus the secondary of R CMa is both overluminous and hotter for its mass;
this is a characteristic of increased abundance of helium relative to hydrogen (cf.
Schwarzschild, 1958; p. 143). Since the evolved components of semi-detached sytems
are supposed to have lost a large portion of their hydrogen envelope exposing the inner
.5
I
t
I
I
I
'
I
0-5
_
+/..
~
I
_
0 3s
/~"
\ :,~
\
:,o.~o
/t
~
I
/[
-
-
0.298
'.2.o ~
\.~.o/
\
\ \.~.;'. \ '~.~._C~
"
~
\
/
It
16L i,.~,
,,
o
J
<3
-;.
-
0
"\.
0.46\
0-250
\dj/)(
~.o
l o! ........ .'-.
"'"""
"
"" "
\
0.5~
~0.35
-2
-
~
-
0 ~ 0 - 9 9 Sirius B
-3
'
Fig. 1.
I
",1.2
,
1
+0.8
i
I
+0.Z,
Alog (1"e )
0
I
I
-0.4
E v o l u t i o n a r y t r a c k s of various objects in the A l o g L - A log T e plot; see text for explanation.
MASS LOSS IN SEMI-DETACHED BINARIES
357
helium-rich core they should show both the increase in luminosity as well as effective
temperature.
We choose, therefore, the following variables:
(i) A (logL) = observed logL - logL for a star of the same mass on the empirical
Main Sequence.
(ii) h (log Te) = observed log Te - log Te for a star &the same mass on the empirical
Main Sequence.
Figure 1 shows the following groups of stars in the A (logL) - A (log Te) plane:
(a) Empirical Main Sequence is represented by just one point viz. the origin.
(b) The zero-age Main Sequence based on calculations of Iben (1965) forX = 0.708,
Z = 0.02 is shown by the S-shaped curve near the origin. The difference from empirical
Main Sequence arises because at the upper end it contains somewhat evolved stars while
at the lowest end it contains stars which have just reached the zero-age Main Sequence.
The square surrounding ZAMS is a measure of uncertainty in the estimation of our
variables.
(c) Tracks (small dashed curves) of pre-Main-Sequence contracting single stars of
various masses (marked on the curves) according to Iben (1965).
(d) Tracks (long-dashed curves) of post-Main-Sequence single stars of various
masses (marked on the curves) due to Iben (1964).
(e) Track of a rapidly mass accreting star of mass 1.5M o in a binary (continuous
curve marked I) according to Neo et al. (1977); the mass of the star at each stage is
marked on the track.
(f) Track of the mass-losing component in a binary with original masses
M~ = 1.8Mm, M 2 = 0.7M e (continuous curve marked II) as calculated by Refsdal and
Weigert (1969); here also the mass at various stages is marked on the track.
(g) Evolutionary tracks of helium white dwarfs in close binaries (continuous curves
marked (III) according to Webbink (1975)); mass of the star is indicated on each curve.
(h) Helium Main Sequence (dash dot curve) due to Hansen et aL (1972); numbers
of this curve give the mass of the helium star.
(i) A typical white dwarf, Sirius B, in the lower part.
It is seen that the pre-Main-Sequence contracting stars and post-Main-Sequence
evolving stars both occ,upy a region to the right &the origin. The rapidly mass accreting
component of a binary first moves slightly to the left due to the heating of the
photosphere by the infalling material, but it also eventually moves to the right as the
envelope grows in size. This track also shows that a star cannot accrete much of the
mass lost by its companion. Most of the mass lost by the evolving component leaves
the system through the outer Lagrangian points.
The track of the mass losing component is, however, quite distinct in that it moves
up and to the left which corresponds to an increase in luminosity as well as surface
temperature. This is obviously the effect of the decrease in the hydrogen content by mass
loss from the envelope. The star eventually moves towards the helium Main Sequence
and ultimately becomes a white dwarf. We thus see that the track of a mass-losing star
in the A ( l o g L ) - A (log T~) plane is markedly different from those for pre-Main-
358
K . D . ABHYANKAR
Sequence and post-Main-Sequence single stars which have not lost mass during the
course of evolution.
3. Components of Semi-Detached Systems
Figure 2 is an enlarged version of Figure 1 with components of semi-detached systems
plotted therein. The open symbols represent the more massive primaries while the filled
+/.
!
~
0
1
0"265
. 2 6
I
5
S Cnc~01.298
+3
- - 0.32 mT LMi
S Eqt 9
9
Eri \
0'4 0.2
2~
0-31 9
Sg~" S Ve[ 9
AL Geml0"l~
_
I'W AndQ0.
0.3 9
Prim.
Sec.
Hot semi- det.
Cool s e m i - d e f .
Algol s y s t e m s
R CMa systems
RS CVn systems
0
II
~
o
o
z~
0.54 mmRZCnc
9
9
9
1.2
f
/
/
~RY
Aqr
Type of binary
/
Per
9
Latin0.6 ~RY Cem
0.2|R CMa RT1-6
~\
0,8"
/
/
I
--%+2
.
8"-'-~"---~.p.
i~
J
I
/I
_A
I
~o
I
<2
\
O.rO
U Her I 1 \
+1
1.4 9 ~1,073 /
\
\
s"
zvu,'
•1"30
\
2"31
4:s d~1.7
0
--
-
I
+0.3
I
+0.2
I
+0-1
r
,.,
I '&
~
qRW O~m,l"8 1.7~SS Sg,;
L f
-- - -
2-1 D)./U Sgr b.7-
'~gd
2.5?
+0.4
1-2
Rs vo,\- /
_-~-~--f
06.5
I
0
A,L,
/
--~,s.0
71
3'0
;~
,,
/
z.z
- - -D-I;
r73.2
I
- 0"1
I
-
0"2
I
- 0.3
I
-0"4
A log (l'e)
Fig. 2.
Position of the components of" semi-detached systems in the A |ogL - A log T~ plot; see text ['or
explanation.
MASS LOSS IN SEMI - DETACHE D BINARIES
359
symbols denote the less massive secondaries; the numbers indicate masses of the two
components. The following groups of binaries are shown in Figure 2:
(a) Hot semi-detached systems (vertical rectangles) listed by Popper (1980).
(b) Cool semi-detached systems (horizontal rectangles) listed by Popper (1980).
(c) Algol-systems (circles) listed by Popper (1980).
(d) R CMa-systems (squares) listed by Cester et al. (1979). The data for R CMa itself
is ours (Radhakrishnan et al., 1983); in the case of other stars the more massive star
was assumed to be a Main-Sequence star of observed spectral type for deriving the mass
of the secondary.
(e) Six RS CVn-systems for which data are taken from Popper (1961, RS CVn),
Arnold etal. (1979, SS Cam), Jakate etal. (1976, SZ Psc), Popper (1956, Z Her),
Mardirossian et al. (1980; WW Dra) and Chambliss (1976; AR Lac).
It is clearly seen that the secondary components of most semi-detached systems
occupy the region of mass losing stars in which the hydrogen content is considerably
reduced. We would like to draw attention to TT Hya in which Kulkarni and Abhyankar
(1981) had considered the secondary to be in pre-Main-Sequence contraction phase.
From its position in Figure 2 it is clear that the star has definitely lost mass. Hence, its
undersize nature which is confirmed by Kaul and Abhyankar (1982), has to be attributed
to the second phase of contraction after considerable loss of mass from the envelope
through the outer Lagrangian points by normal evolutionary process.
It may be noted that the secondaries of semi-detached systems invariably lie below
the tracks of helium white dwarfs of same masses as calculated by Webbink (1975). This
underluminosity indicates that they have lost mass so rapidly that the shell burning of
hydrogen is also getting suppressed.
4. Conclusions
We have demonstrated that the A (logL) - A log(Te) plot separates the mass losing
evolved components of semi-detached systems from the pre-Main-Sequence contracting
stars and post-Main-Sequence expanding single stars. It is to be noted that only a small
fraction of the mass lost by the evolved component can be accomodated on the
companion. Hence, most of the mass is lost from the system through the outer
Lagrangian points, and the more massive primary has essentially retained its original
character with only a small increase in its mass. Since the evolved secondary must have
been originally the more massive component the quantity 1 - q, where q = m2/m 1 is the
mass ratio, is a measure of the minimum fraction of mass lost by it. Hence, there should
be a correlation between the mass ratio and the state of evolution of the secondary.
In order to find this correlation we consider the differential evolution of the two
components of the semi-detached system in the h (logL) - A (log Te) plane. Defining
5 = A (secondary) - A (primary) we have plotted 8 (A logL) vs 5 (A log Te) in Figure 3
for all the systems shown in Figure 2. We find a linear correlation (line A B ) between
these quantities such that systems with small value of q, which lie at the upper end of
the line AB, show larger differential evolution by mass loss as compared to the systems
360
K . D . ABHYANKAR
2-\
'
RZ Cnc "" 0.12
+3
T
S
s cnc 9
I
O.OB
9 XZ Sgr
LMI 10.11
_
tAS
\
"o.,o
Eri
.o.LV ,
~10'22
W And
~
O'i41R
0,40 1 R T
+2
Lac
CMa
~
T"T Hya
I0
""~Y
Gern9 23
297
9 RY Per
0.16
3
J
v
0.22
D / 3 Per
~+1
V 356
9 RW Gem
0-28
Sgr
0.47
V Pup
I
u Her
0,53
SZ Psc
0
--
9
5$ Cam
I
+ 0-Z,
t
+ 0-3
I
W Dra 9 0"31
0 37
Vul
z vul.o ~2_L.__~, __
mAA-RS CVn
WW D r a A l &
B
Z Her
I
I
I
+ 0.2
+ 0'1
- 0-1
-0.2
[~ tog T~]
Fig. 3. Differential evolution of the components of the semi-detached systems; q = m2/m L values are
marked for each system.
with larger values of q which tie at the lower end. The RS CVn-systems show the least
differential evolution by mass loss and that also along the axis b (A logL) = 0 as is
expected from their almost equally massive detached components.
The correlation of overluminosity with q was noted long ago by Struve (1954). We
now find that the excess of effective temperature is also similarly correlated with q. As
the evolved secondary loses mass with attendant reduction in the value of q, its hydrogen
content is also reduced and the star becomes both overluminous and hotter for its
residual mass.
References
Arnold, C. N., Hall, D. S., Montle, R. F., and Stuhlinger, T. W.: 1979, Acta Astron 29, 243.
Cester, B., Giuricin, G., Mardirossian, F., Mizzetti, M., and Milano, L.: 1979, Astron. Astrophys. Suppl. 36,
273.
Chambliss, C. R.: 1976, Publ. Astron. Soe. Pacific 88, 762.
Hansen, C. J., Cox, J. P., and Herz, M. A.: 1972, Astron. Astrophys. 19, 144.
MASS LOSS IN SEMI-DETACHEDBINARIES
361
Iben, I., Jr.: 1964, Astrophys. J. 140, 1631.
Iben, I., Jr.: 1965, Astrophys. J. 142, 993.
Jakate, S., Bakos, G. A., Fernie, J. D., and Heard, J. F.: 1976, Astron. J. 81,250.
Kaul, J. and Abhyankar, K. D.: 1982, J. Astrophys. Astron. 3, 93.
Kopal, Z.: 1956, Ann. Astrophys. 19, 298.
Kulkarni, A. G. and Abhyankar, K. D.: 1981, J. Astrophys. Astron. 2, 119.
Mardirossian, F., Mezzetfi, M., Cester, B., and Giuriein, G.: 1980, Astron. Astrophys. Suppl. 39, 73.
Neo, S., Miyaji, S., and Nomoto, K.: 1977, Publ. Astron. Soc. Japan 29, 249.
Popper, D. M.: 1956, Astrophys. J. 124, 196.
Popper, D. M.: 1961, Astrophys. J. 133, 148.
Popper, D. M.: 1980, Ann. Rev. Astron. Astrophys. 18, 115.
Radhakrishnan, K. R., Sarma, M. B. K., and Abhyankar, K. D.: 1983, Astrophys. Space Sci. 99, 229 (this
volume).
Refsdal, S. and Weight, A.: 1969, Astron. Astrophys. 1, 167.
Sahade, J.: 1963, Ann. Astrophys. 26, 80.
Schwarzschild, M.: 1958, Structure and Evolution of the Stars, Princeton University Press, Princeton.
Struve, O.: 1954, Mem. Soc. Roy. Liege 14, 236.
Webbink, R. F.: 1975, Monthly Notices Roy. Astron. Soc. 171,555.