Download Discovery of Eclipsing Binary with the Longest Known Period

Discovery of an Unusual Very Bright Eclipsing Binary MASTER OT
J095310.04+335352.8 with the Longest Known Period
V. Lipunov1, D. Denisenko1, E. Gorbovskoy1, V. Afanasiev2, D. Makarov2, A. Vinokurov2, V.
Krushinsky3, R. Jansen4, A. Tatarnikova1, N. Tiurina1, P. Balanutsa1, A. Tlatov5, V. Senik5,
Yu.Sergienko6, V. Yurkov6, A.Gabovich6
1
Lomonosov Moscow State University, Sternberg Astronomical Institute, Russia
Special Astrophysical Observatory of Russian Academy of Sciences, Nizhniy Arkhyz, Russia
3
Ural Federal University, Yekaterinburg, Russia
4
Dutch Astronomical Data Mining Initiative, The Netherlands
5
Kislovodsk Solar Station of Pulkovo Observatory, Russia
6
Blagoveshchensk Educational University, Russia
2
We report the discovery of the eclipsing binary MASTER OT J095310.04+335352.8 with extreme parameters. The orbital
period P=69.1 yr is more than 2.5 times longer than that of Epsilon Aurigae. The light curve is characterized by an
extremely deep total eclipse (with a depth of more than 4.5m) of symmetrical shape and a total duration of 3.5 yr. The
eclipse is essentially gray. The spectra acquired with the Russian 6-m BTA telescope both at minimum and maximum light
correspond mainly to an M1III-type red giant, but the spectra taken in the bottom of eclipse show small traces of a
sufficiently hot source. The observed properties of this variable can be explained as those of a red giant eclipsed by a large
cloud of small particles surrounding the unseen secondary companion, most likely a protoplanetary disk observed nearly
edge-on.
Subject: stars: binaries – stars: individual: MASTER OT J095310.04+335352.8 (TYC 2505-672-1)
1. INTRODUCTION
MASTER Global Robotic Net consists
of five identical MASTER-II observatories
(twin 0.40-m f/2.5 reflectors + 4Kx4K CCD,
which give us a 8 square degrees field of view),
operating in Russia from East to West MASTER-Amur, MASTER-Tunka, MASTERUral, and MASTER-Kislovodsk (Lipunov et.al.
2010, Kornilov et al. 2012, Gorbovskoy et al.
2013), as well as MASTER-SAAO, installed at
the South African Astronomical Observatory in
December, 2014. We also have MASTER Very
Wide-field cameras (MASTER-VWFC) in
Argentina with 800 square degrees field of
view and a limiting magnitude of 15m.
The typical limiting magnitudes on the
single images taken in survey mode range from
19-20m (60-sec exposures) to 20-21.5m (180-sec
images). All images are reduced in real-time
mode using our own unique software developed
by the MASTER team over the last 10 years.
This unique software is the key
MASTER feature that allows us to work both in
alert and survey mode, discovering optical
transients (GRB counterparts and another OTs)
in real-time. The fully robotic MASTER
software schedules current observations
depending on the weather and Sun ephemeris,
automatically points to the GRB alerts from
spacecrafts, automatically processes images in
real-time mode, and detects all new sources of
different nature (both stationary and moving) in
real-time mode.
The object is detected as a transient if it
appears in at least two images in the same
night, and there is no match within 5" circle in
either USNO-B catalog or MASTER reference
images. Besides the conventional transients the
MASTER database contains objects of the type
named Flare Star. The formal criterion for Flare
Star detection is a change in the magnitude by
more than 2.2 from its catalog value. Most of
the MASTER survey images are obtained in the
white light (unfiltered) to increase the limiting
magnitude. The corresponding internal
photometric magnitudes can be described fairly
well by the equation W=0.8*R2+0.2*B2, where
R2 and B2 are the 2nd epoch DSS red and blue
magnitudes, respectively, adopted from USNOB1.0 catalog (Monet et al., 2003).
2. MASTER DISCOVERY
The optical source MASTER OT
J095310.04+335352.8 (also known as TYC
2505-672-1) was originally discovered by
MASTER Global Robotic Net auto detection
system (Lipunov et.al. 2010) as an optical "antitransient" (Denisenko et al., 2013).
The primary goals of MASTER Global
Robotic Net are rapid response to gamma-ray
bursts (GRB) in alert mode and discovery of
optical transients in survey mode.
The absolute majority of MASTER
optical transients (OTs) are flares of various
types (optical counterparts of GRBs,
supernovae, QSO flares, dwarf novae, UV Ceti
type flare stars). As of April, 2015, MASTER
system has discovered 800 optical transients
(see
http://observ.pereplet.ru/MASTER_OT.html).
The
unusual
MASTER
OT
J095310.04+335352.8 object has met the Flare
Star criterion on MASTER-Amur and
MASTER-Kislovodsk images acquired in the
end of 2012 – beginning of 2013. The
previously unremarkable star TYC 2505-672-1
(R.A. = 09h53m10s.00, Decl. = +33°53'52''.7,
V=10.71,
B=12.51)
=
2MASS
J09531000+3353527 (J=7.61, H=6.78, K=6.57)
faded by ~4m from its ordinary level. Figure 1
compares MASTER-Amur images acquired in
2011 (maximum) and 2013 (deep fading), and
Table 1 lists selected photometry obtained by
MASTER telescopes.
Table 1. MASTER photometry of MASTER
OT J095310.04+335352.8 in 2011-2013.
Obs. Date
Magnitude
Observatory
and time
and filter
2011-02-03
10.34W
MASTER15:15:47
Amur
2011-03-14
10.03V
MASTER10:43:18
Amur
2011-04-07
10.44W
MASTER12:30:22
Amur
seasonal gap due to solar conjunction
2011-11-03
12.62W
MASTER23:44:28
Kislovodsk
2011-11-04
12.64W
MASTER00:17:08
Kislovodsk
2011-11-09
12.99R
MASTER17:42:27
Amur
2011-11-20
13.10R
MASTER16:58:19
Amur
2011-11-29
13.27R
MASTER16:47:27
Amur
2011-12-09
13.44R
MASTER16:14:25
Amur
2012-01-08
13.85R
MASTER13:37:51
Amur
2012-03-01
14.06R
MASTER12:50:18
Amur
2012-03-17
14.31V
MASTER14:40:17
Amur
2012-04-02
13.81W
MASTER18:09:36
Amur
seasonal gap due to solar conjunction
2012-01-08
13.85R
MASTER13:37:51
Amur
2012-03-01
14.06R
MASTER12:50:18
Amur
2012-03-17
14.31V
MASTER14:40:17
Amur
2012-04-02
13.81W
MASTER18:09:36
Amur
seasonal gap due to solar conjunction
2012-09-29
13.98W
MASTER20:02:30
Amur
2012-10-25
14.03W
MASTER19:23:53
Amur
2012-12-19
14.01W
MASTER15:46:57
Amur
2013-01-07
13.77W
MASTER16:46:19
Amur
2013-02-01
13.82W
MASTER16:51:19
Amur
The spectral type of and estimated distance to
TYC 2505-672-1 was reported (Pickles and
Depagne, 2010) to be M2III and 1179 pc,
respectively. Combined with the UCAC4
proper motion (3.5, -7.4) mas/yr (Zacharias et
al., 2013) this corresponds to a tangential
velocity of 46 km/s, which is reasonable. The
Galactic coordinates of the source are (191.7,
+51.3) and extinction toward it is negligible.
This formally translates into a ~900 pc height
above the Galactic plane, which appears too
large for a variable red giant star.
The behavior of the object was puzzling. At
first it was supposed to be a new R CrB type
variable (red supergiant ejecting carbon clouds
at irregular intervals of time). However, the
typical dimming phase of R CrB stars lasts
from 1 week to 1 month, whereas it took TYC
2505-672-1 more than 3 months to fade from
10m to 14m. The star was nearly constant in
1999-2000 during the observations by NSVS
(ROTSE-I) project. No previous fadings were
detected on nine Palomar plates taken from
1955 to 1996 and on 31 Palomar/NEAT images
taken during 9 different nights in 2001-2003.
The object was not detected in 1RXS and IRAS
bright source catalogs. Follow up spectroscopic
and photometric observations of TYC 2505672-1, as well as archival searches for the
historical fading episodes were requested in
ATel 4784.
3.SPECTROSCOPY AND PHOTOMETRY
The first spectrum of MASTER OT
J095310.04+335352.8 in eclipse (Afanasiev et
al., 2013) was obtained on Feb. 10, 2013 with
the Russian 6-m BTA telescope (SAO RAS,
Nizhniy Arkhyz, Karachay-Cherkessia) +
SCORPIO instrument (Afanasiev V. and
Moiseev A., 2005). The spectrum was
consistent with that of an M1III-type red giant
without unusual absorption lines, but with Hα
in emission (FWHM=320 km/s). The observed
wavelength of Hα line was 6561.0±0.1 Å
corresponding to the blue shift of -82 km/s. A
number of weak FeI and NaI emission lines
were also present with typical FWHM of ~150180 km/s.
Figure ??? shows the red part of an after eclipse
spectrum. It is a pure M1-type giant SED. Note
that Hα and the NaI doublet are in absorption
although they were in emission during eclipse.
The emission components of the lines may
possibly also be present, but their intensities are
near noise level due to the very strong spectrum
of the red giant.
Following
the
MASTER
discovery
announcement, the new variable star was added
to AAVSO international database and Variable
Star Index (Watson et al., 2007), initially as a
suspected R CrB star. As a result, the recovery
from the deep fading in 2013–2014 was traced
quite well by several visual observers and
observers using CCD's with BVRI filters. The
data have shown the gradual increase of flux in
all photometric bands without a color change
(gray eclipse). As of October, 2014 after the
yearly gap due to the solar conjunction the star
has returned to its pre-eclipse level of V=10.8m,
B=12.5m. The second-epoch spectrum taken
with BTA also showed M1III spectral type.
However, the Hα line has completely
disappeared in the red continuum that has
increased ~50-fold.
4.HISTORICAL ECLIPSE
On June 23, 2014 the 3rd release of digitized
Harvard College Observatory plates (Grindlay
et al.) was published covering Galactic latitudes
from +60 to +45. One of the authors (R.J.) has
checked the position of MASTER OT
J095310.04+335352.8 on the DASCH project
website http://dasch.rc.fas.harvard.edu/ and
found an additional eclipse of this star in 1942–
1945. The comparison of DASCH images from
1968 (maximum) and 1945 (fading) is shown in
Fig. 2. The DASCH observations of this field
cover the time interval from 1890 to 1990 with
one significant gap in 1953-1967. Figure 3
shows the light curve of
MASTER OT
J095310.04+335352.8 = TYC 2505-672-1
comprised of Harvard plates, NSVS, Catalina
Sky Survey (Drake et al., 2013) and MASTER
observations. Magnitudes measured in various
photometric bands were reduced to the common
zero point using different color adjustments.
For example, B-band photographic magnitudes
from Harvard plates were corrected by –1.8m
and unfiltered (W) MASTER observations by
+1.45m, respectively. Thus, the mystery of this
object
was
solved.
MASTER
OT
J095310.04+335352.8 turned out to be a deeply
eclipsing binary star with an unprecedented
long interval between the eclipses.
mag
V0695 Cyg EA/GS/
D
3784.3
10.36
3.73
–
3.89
V
V0383 Sco EA/GS+
SRC
4875.9
13.35
10.6
–
16.3:
V
Gamma Per EA/GS
5346
14.64
2.91
–
3.21
V
V0381 Sco EA/GS+
SRC
6545
17.92
11.63
V–
16.0
p
VV Cep
EA/GS+
SRC
7430
20.34
4.8 –
5.36
V
NSV 6116
(α Com)
EA:
9443
25.85
4.32
–?V
Eps Aur
EA/GS
9892
27.08
2.92
–
3.83
V
5.PERIOD DETERMINATION
Using the combined data from 1890 to 2014,
we determined the orbital period of this new
variable to be P=25245, which is equal to 69.1
yr. We rule out the possibility of a fraction
value of this period.
Taking the gap in 1965–1967 data and
the lack of accurate data in 1990–1992 into
account, the P/3 value is 23 years. But the
Harvard data in 1920–1922 and 1897–1899
show no fading of this star 23 and 46 years
before the 1942–1945 eclipse, thereby ruling
out this possible alias. Figure 4 shows the
folded light curve with the 69.1 yr period. Thus
MASTER OT J095310.04+335352.8 becomes
the eclipsing variable with the period at least
2.5 times longer than those of all previously
known objects with measured periods.
There are seven eclipsing variables with
known periods longer than 10 years in the
General Catalogue of Variable Stars (Samus et
al.). They are listed in Table 2 with their types,
periods (in days and years for convenience) and
magnitude ranges. EA classification in the
second column corresponds to the variable of
Algol type, GS subtype means the system with
one or both giant and supergiant components,
SRC refers to semiregular pulsations of a latetype supergiant. A colon (:) means that the
listed value is uncertain.
Table 2. Previously known eclipsing
variables with periods longer than 10 years
(3652.5 days).
Star name Variabil
ity type
Period,
days
Notes: the orbital period of NSV 6116 (α Com),
which is a visual and double-lined
spectroscopic binary with the inclination very
close to 90°, is determined from long-term
optical observations. The initial calculation (by
Muterspaugh et al., 2010) showed the eclipses
of Alpha Com to be highly likely, with a
predicted closest projected approach around
2015 Jan. 24 using the period value of 9485.7d.
As the event approached, the new findings
showed the prediction to be in error. The
revised value of the period is 9443.1±3.0 days
(Muterspaugh et al., 2015). The eclipse has
likely occurred 2 months earlier than predicted,
around 2014 Nov. 20, and has not been
observed.
Period, Rang
Orbital periods of these seven stars and with
years
e,
MASTER OT J095310.04+335352.8 are
plotted in Fig. 5. This new MASTER variable
star stands out on this plot and it is clear, that its
period is exceptionally long.
5.PARAMETERS OF THE BINARY
SYSTEM
The mass of the eclipsing body is unknown
since it is entirely hidden by the surrounding
dust cloud. We assume for the simplicity of
calculation that the total mass of the system is
about 4 M. From the generalized 3rd Kepler's
law we then obtain the semi-major axis of about
33 AU and the orbital velocity of 14 km/s (3.0
AU/yr). The eclipse duration is 3.5 yr (0.05 P)
with the ingress lasting 1 yr, total phase 1.5 yr,
and egress again 1 yr. That gives us the total
size of the eclipsing body to be about 7.5 AU (1
billion km, 1014 cm or 750 solar diameters).
The nominal size of the M1III-type red giant is
60 times that of the Sun (0.6 AU in diameter).
Since the partial phases of the eclipse last about
a year rather than 0.2 yr, the eclipsing body
should have its density increasing towards the
center, as expected for the protoplanetary disk.
The height (thickness) of the eclipsing body
should exceed the diameter of the primary
M1III-type star (0.6 AU). So, the vertical to
horizontal size ratio may be as small as 1:10 in
case of edge-on geometry.
6.DISCUSSION
Taking into account the above mentioned data,
MASTER OT J095310.04+335352.8 is similar
to the famous eclipsing binary Epsilon Aurigae
(ε Aur), but more extreme. First of all, the
eclipse depth in ε Aur is five times smaller than
in J0953+3353, despite the fact that the central
star is hotter and more luminous (F0I). That
difference may be explained by the different
viewing geometry of the two binary systems.
Since the inclination of the orbital plane of
J0953+3353 is closer to 90° and the central star
is smaller (~60 R vs ~200 R), we see a
deeper eclipse. The second difference is the
eclipse shape. In J0953+3353 the eclipse is
symmetrical with a flat bottom, whereas in ε
Aur the ingress is shallower and egress is
steeper (Stencel, 2012). Third, Maslov et al.
(2014) noted that ε Aur has shallow secondary
minimum on IR light curves. Finally, there is a
rebrightening in the middle of eclipse in ε Aur
explained by the gap or a hole in the center of
the occulting disk supposedly swept by the
binary star inside (Stencel et al., 2008).
According to the recent infrared observations
with Herschel Space Observatory (Hoard et al.
2012), the cool dust disk in ε Aur has an
overall temperature about 550 K with the side
of the disk facing the bright F-type star being
warmer (1150 K).
There are two systems in Table 2 that have an
eclipse depth of about 5 magnitudes, similar to
that of MASTER OT J095310.04+335352.8.
However, these two stars (V381 Sco and V383
Sco) are EA/GS+SRC type variables. In these
systems semiregular pulsating stars (red
supergiants)
are
eclipsing
the
hotter
components. According to Galan et al., the
V383 Sco system consists of a pulsating Mtype supergiant that periodically obscures the
much more luminous F0I star. The eclipses in
V383 Sco have different depth in various
photometric bands, with the color index (V-I)
changing by more than 1.5m between the
maximum and minimum light. The principal
difference between J0953+3353 and such
systems is that the obscuring body of
J0953+3353 can not be self-luminous. As in
case of ε Aur, we are dealing with a dust disk
that absorbs the light of the central star(s)
resulting in the gray eclipse. Also, no
significant pulsations in maximum light are
present in the historical light curve of
J0953+3353 which also tells in favor of the
cold dust disk.
The spectral energy distribution (SED) of the
system was quite similar on both our spectra
(despite the great difference in the continuum
fluxes). In eclipse as well as after eclipse the
main contributor to the SED is a red giant, and
its TiO bands are the most prominent
absorption features in the spectra. Note that we
neglect the interstellar reddening (because the
system is quite above the Galactic plane and
there is not any sufficient continuum
distortion).
The spectrum obtained in eclipse (February 10,
2013) includes the red region, allowing us to
estimate the spectral type of the partially
eclipsed red giant by means of [TiO]1 spectral
index ( S.J. Kenyon &T. Fernandez-Castro,
1987). The star is an M1-type giant. Note that
the SED in the red region is nearly the same as
that of a standard M-type1 giant, but in the blue
region the situation is quite different (see Fig.
6). The continuum fluxes are nearly 20 %
higher than the standard M1III SED. This
spectrum also exhibits some faint emission
lines against the TiO bands (see section
“Spectroscopy and photometry”).
The second spectrum was obtained after
eclipse. It doesn’t include the red region and we
therefore cannot trace the emission lines’
evolution. The main difference between our two
spectra is that the after eclipse the observed
SED is similar to standard M1 SED even in the
blue region (Fig. 7, Fig.8).
Let’s assume that the red giant has no sufficient
intrinsic variability, and neglect the limbdarkening effect. We subdivided the after
eclipse spectrum into 270 for continuum fluxes
near 5000 Å were the same as on the spectrum
in eclipse. We then subtract one spectrum from
another. As a result we obtained a SED of some
sufficiently hot source (see fig. N+2). Its
continuum flux grows from the visual to the
blue region. Note that the emission of this
source was visible only during eclipse (when
the contribution of the red giant to the total flux
was greatly diminished). Without eclipse the
SED is the pure contribution from the M1IIItype star. The SED of the hot component (Fig.
9) does not correspond to a SED of any
standard hot star. This may be due to the small
wavelength interval covered by this SED and
the bad signal/noise ratio in the blue part of the
spectrum. However, all numerical estimates of
the hot component properties are incorrect
without observations in the UV-range.
There is one more confirmation of the faint hot
component appearance during eclipse.
According to AAVSO data (Kafka S., 2015),
the color index B-V was bluer near the center
of the eclipse (B-V=1.0-1.2 mag) than during
egress and after eclipse when this color index
was nearly constant (B-V=1.6 mag).
A comparison of TYC 2505-672-1 with ε Aur
(another long period binary with similar
features on its light curves) leads us to conclude
that both systems have common features in
their SEDs. Although in the case of ε Aur
system we observe F2-type supergiant instead
of an M1-type giant, both systems demonstrate
faint traces of a hot component (imbedded in
the center of a huge dust disk). The existence of
the hot component (its spectral class is B1) for
ε Aur was proved by analyzing the UV-spectra
obtained by International Ultraviolet Explorer.
7.RESEARCH PROSPECTS
High-resolution spectroscopy (λ/Δλ~20000) is
needed to determine the component mass ratio
of this unique binary system. Due to the
exceptionally long orbital period, the changes
of radial velocity will become noticeable in
about 8 years (P/8) after the mid-eclipse time.
Precision photometric observations will also be
useful around orbital phase 0.5 (expected 2047)
to search for possible reflection effects. Far
infrared observations are encouraged to
determine the spectral contribution of the dust
disk and the size of its constituent particles.
Also, suspected variable stars, especially Mtype giants with RCB: and E: classification in
the NSV catalog should be checked in the
databases like DASCH for the past eclipses.
8.CONCLUSION
We have discovered an eclipsing binary system
with extreme properties: unprecedented long
period of 69.1 yr and very deep eclipse
reaching at least 4.5m in the V band, explained
by a cold dust disk. This discovery has been
possible by combining the unique search
capacities of modern MASTER robotic
telescope system and digitization of the archival
photographic plates covering more than a
century of observations. Our finding shows
once again that the sky is full of extraordinary
objects
awaiting
their
identification.
Combination of the state-of-art survey projects
and data mining in the numerous archives
stored at astronomical observatories worldwide
can bring many unexpected discoveries, even
among bright objects.
ACKNOWLEDGEMENTS
MASTER project is supported in part by the
Program of Development of Lomonosov
Moscow State University. This work was
partially supported by RFBR 15-02-07875. We
are thankful to several AAVSO observers who
have reported their BVRI photometry of
MASTER OT J095310.04+335352.8 to
AAVSO international database.
REFERENCES
Afanasiev V. et al. 2013, Astronomer's
Telegram #4834
Afanasiev V., Moiseev A. 2005, Astron. Lett.,
Vol. 31, No. 3, p. 194
Afanasiev V., A. Moiseev 2011, Baltic
Astronomy, Vol. 20, p. 363-370
Gorbovskoy E. S. Et al. 2013, Astronomy
Reports, Vol. 57, Issue 4, pp.233-286
Denisenko D. et al. 2013, Astronomer's
Telegram #4784
Drake A. et al. 2013, Astronomer's Telegram
#4788.
Hoard J.D. et al. 2012, Astroph. J. Lett., Vol.
748, Issue 2, article id. L28.
Hack M. and Selvelli P.L. 1979, A&A, 75, 316,
1979
C. Galan et al. 2013, Astron. & Astroph., Vol.
550, id. A93
Grindlay et al. 2012, Proc. IAU Symp. No. 285,
p. 29
Kafka, S., 2015, Observations from the AAVSO
International Database, http://www.aavso.org
S.J. Kenyon and T. Fernandez-Castro A. 1987,
93, 938
V. Kornilov et al. 2012, Experimental
Astronomy, Vol. 33, Issue 1, pp.173-196, 2012
V. Lipunov et al. 2010, Advances in
Astronomy, Article ID 349171
http://www.hindawi.com/journals/aa/2010/3491
71/
I. A. Maslov, A. E. Nadjip, O. G. Taranova, A.
M. Tatarnikov, V. I. Shenavrin. 2014,
Astrophysics, Vol. 57, Issue 3, p. 370
D. Monet et al. 2003 Astron. J., Vol. 125, p.
984
M. Muterspaugh et al. 2010, Astron. J., Vol.
140, Issue 6, p. 1623
M. Muterspaugh et al. 2015,
http://arxiv.org/abs/1501.05639
A. Pickles, E. Depagne 2010, Publ. Astr. Soc.
Pac., Vol. 122, p. 1437
R. Stencel. Journ. of AAVSO 2012, Vol. 40, No.
2, p. 618
R. Stencel et al.2008, Astroph. J., Vol. 689,
Issue 2, p. L137, 2008.
Samus N.N., O. Durlevich, E. Kazarovets et al.
CDS, B/gcvs, 2007–2013.
Samus N. N., Durlevich O. V., al e., 2009,
VizieR Online Data Catalog, 1,
Samus, N. N.; Kazarovets, E. V.; Pastukhova,
E. N. 2009, Publ. Astron. Soc. Pac., 121, 13781385
C. Watson, A. Henden, A. Price. Journ. of
AAVSO, 2007, Vol. 35, No. 2, p. 414
N. Zacharias et al. 2013, Astron. J., Vol. 145,
Issue 2, article id. 44
FIGURES
Fig. 1. MASTER-Amur images of MASTER OT J095310.04+335352.8 = TYC 2505-672-1 in deep
fading (2013, left panel) and in maximum (2011, right panel).
Fig. 2. DASCH images of TYC 2505-672-1 at the maximum light (1968) and in eclipse (1945).
Fig. 3. The historical light curve of MASTER OT J095310.04+335352.8 = TYC 2505-672-1 combined
from various sources with the corresponding color corrections applied.
Fig. 4. The light curve of MASTER OT J095310.04+335352.8 folded with the 69.1 yr period. Phase
interval 0.95-1.05 is zoomed-in at the bottom showing the symmetric shape of the eclipse.
Fig. 5. Previously known eclipsing variables with orbital periods longer than 10 years compared to
MASTER OT J095310.04+335352.8.
Fig. 6.
Spectrum of MASTER OT J095310.04+335352.8. obtained during eclipse on February 10, 2013
(points connected with the thin line) and spectral energy distribution of a standard M1-type giant (zero
fluxes in the model energy distributions mark the spectral regions with strong telluric absorption
bands).
Fig. 7.
MASTER OT J095310.04+335352.8 spectrum obtained after eclipse on October 26, 2014
(points connected with thin line) and spectral energy distribution of a standard M1 giant (thick line).
Fig.8.
MASTER OT J095310.04+335352.8 spectrum obtained after eclipse on 2015-03-19. One can see the
red part of an after eclipse spectrum. Note that H\alpha and NaI doublet are in absorption although they
were in emission during eclipse. Emission components of the lines may also be present, but their
intensities are near noise level due to the very strong spectrum of the red giant.
Fig. 9.
Spectral energy distribution of the “hot component”.