Äîêóìåíò âçÿò èç êýøà ïîèñêîâîé ìàøèíû. Àäðåñ îðèãèíàëüíîãî äîêóìåíòà : http://hea-www.harvard.edu/PINTofALE/papers/cs10_drakej1/drakej1.ps Äàòà èçìåíåíèÿ: Fri Nov 3 03:21:41 2000 Äàòà èíäåêñèðîâàíèÿ: Tue Oct 2 08:13:22 2012 Êîäèðîâêà: Ïîèñêîâûå ñëîâà: comet

Cool Stars, Stellar Systems and the Sun
ASP Conference Series, Vol. 154, 1998
eds. R. A. Donahue and J. A. Bookbinder
The Coronal Metallicity of the RS CVn Binary HR 1099
Jeremy ``Unbiased'' Drake and Vinay Kashyap
Harvard­Smithsonian Center for Astrophysics, 60 Garden Street,
Cambridge MA 02138
Abstract:
We have derived the abundance of iron relative to hydrogen in the corona
of the active RS CVn binary HR 1099 (V711 Tau) using EUVE spectroscopic
observations and additional X­ray constraints provided by ASCA observations.
We have compared the results with simple two temperature optically thin plasma
model fits to an ASCA spectrum of this star. Results from the two methods are
in reasonable agreement. Existing photospheric studies also yield a similar Fe
abundance of [Fe/H]¸ \Gamma0:4, suggesting at face value an absence of any coronal
abundance anomaly in HR 1099 involving Fe. However, as with nearly all of
the RS CVn's, existing photospheric abundances are perhaps questionable and
demand further scrutiny.
1. Summary
We have determined the coronal metallicity of the very active RS CVn binary
HR 1099 (V711 Tau) using two methods:
1. Comparing the Fe line to continuum ratio in a high quality EUVE spectrum
co­added from different observations to yield an aggregated exposure of
230,000 s;
2. Plasma model fits to ASCA pulse height distributions. The results indicate
an iron abundance in the corona of HR 1099 of [Fe/H]= \Gamma0:4 \Sigma 0:2.
The highest temperature of significant coronal emission measure in active
RS CVn stars such as HR 1099 exceeds the formation temperature of the hottest
bright spectral line formed in the bandpass of the EUVE spectrometer (Fe xxiv
192 š A). Thus, in order to constrain this higher temperature plasma emission
measure distribution and continuum flux contribution in the EUV bandpass,
additional contraints from X­ray observations are required. We used the ASCA
spectrum to place constraints on the hot tail of the EM distribution (log T ? 7:3)
by fitting the continuum to determine the hottest significant plasma tempera­
ture. This technique can be extended to low S/N spectra by comparing the
EUVE Deep Survey count rate to the predicted count rate based on the EUVE
EM distribution calculated for different values of metallicity.
CD--1014

Coronal Metallicity of HR 1099 CD--1015
2. Analysis and Results
The analysis and results are described in the following figures and their captions.
Figure 1. The 2T ­fit to the ASCA spectrum (MEKAL) with variable
global metallicity. Care is needed not to arrive at unrealistic results.
The fits are unstable to free parameters such as the intervening in­
terstellar medium neutral hydrogen column density NH , and to an
additional gaussian component added to either mimic additional spu­
rious or actual missing lines in the plasma radiative loss model. The
2T fit parameters derived here (reduced ü 2 = 2:4) with NH fixed at
1:35 \Theta 10 18 are Z = 0:52 \Sigma 0:02 or [Fe/H]= \Gamma0:3; T1 = 0:87 keV,
Norm= 1:3; T2 = 2:27 keV, Norm= 6:2.
3. Conclusions
The coronal abundance derived here is similar to the photospheric iron abun­
dance of [Fe/H]=­0.6 derived for the secondary star (which dominates the EUV
and X­ray emission) by Randich, Giampapa & Pallavicini (1994). At face value,
this suggests no coronal abundance anomaly is present. However, the iron abun­
dances derived for the primary is solar, [Fe/H]=0.0! This Fe dichotomy is com­
mon to other RS CVn's, which in general also appear more metal poor than
their ages and galactic locations might suggest. Either: i) The photospheric
abundances are incorrect; or ii) The RS CVn's pose a major puzzle for stellar
evolution theory.

CD--1016 J.U. Drake and V. Kashyap
Figure 2. The DS light curve of HR 1099 for the three co­added
observations corresponding to the spectra used in our analysis. Each
time bin corresponds to one EUVE orbit, or about 5540 s of elapsed
time, and to between ¸ 500 and ¸ 2000 s of exposure time. The average
count rate is indicated by the green (dashed) line. Two or three flares
occur, but the combined spectrum is not flare dominated.

Coronal Metallicity of HR 1099 CD--1017
Figure 3. The EUVE spectrum with line identifications. We draw
attention to the likely presence of of many weak transition region lines
in the 80--140 š A range arising from n = 2--3 L­shell transitions in Ne,
Mg and Si.

CD--1018 J.U. Drake and V. Kashyap
Figure 4. The EM distribution outlined by the various lines due to
charge states of Fe ranging from Fe xxv-- xxiv.
While, in comparison to the photospheric state­of­the­art, stellar coronal
abundance estimates are currently slightly crude, especially those based on ra­
diative loss model fits to low resolution X­ray pulse height spectra (e.g., Fig­
ure 1), in the case of the RS CVn's we are currently in a situation where the
best possible coronal results are probably more reliable than existing photo­
spheric measurements. However, it should be remembered that in the coronal
case we are also often treating observations of both coronae of a binary system as
that of a single star. In the event that the two stars are at different evolutionary
phases, or even that abundance anomalies, in which coronal values differ from
photospheric, are present in the RS CVn's, the coronal abundances might differ
on the two components.
The manifold simplifications involved in the optically­thin, ionization and
excitation equilibrium approach to the analysis of what are highly complex and
dynamic coronae should also be born in mind. However, in cases where photo­
spheric studies are presented with problems, such as large rotational broadening
and composite spectra in the case of RS CVn and BY Dra binaries, and atmo­
spheric modeling difficulties in the case of active M dwarfs, the coronal regime
offers an alternative route to stellar abundances. Moreover, elements such as Ne
and Ar that are inaccessible in late­type photospheres are observable in coronae.
In order to verify whether or not true coronal abundance anomalies are
present on the RS CVn's, the photospheric abundances demand further atten­

Coronal Metallicity of HR 1099 CD--1019
Figure 5. Spot the continuum! Use your skill, judgement and subjec­
tive bias to estimate the coronal iron abundance of HR 1099. Answers
on a postcard to . . . The EUV continuum calculated from the observed
emission measure for different values of the Fe abundance. The ``true''
continuum, in my (JJD: subjective and biased) opinion, lies at about
Z = 0:4 solar, or [Fe/H]= \Gamma0:4.
tion. Indeed, we note that one or two new photospheric abundance studies
appear elsewhere in this volume. This is one direction we are pursuing in future
work.

CD--1020 J.U. Drake and V. Kashyap
Figure 6. (top panel): The metal abundances for RS CVn's and ac­
tive stars derived by Randich, Giampapa & Pallavicini (1994) and (bot­
tom panel): the abundances in the RS CVn's for which estimates were
made for both primary and secondary components (joined dots). The
solid line in the upper panel illustrates the mean abundance vs tem­
perature. In general this sample of active stars is too metal poor for
thin disk stars; moreover, the wildly different Fe abundances in pri­
mary and secondary components cannot be understood within current
stellar evolution theory if they were formed from the same protostellar
material. As also discussed by Randich et al., the abundances could be
affected by, e.g.: i) problems in deconvolving the composite spectra; ii)
spots; ii) in­filling by the chromosphere.

Coronal Metallicity of HR 1099 CD--1021
Acknowledgments. JJD was supported by the AXAF Science Center (un­
der NASA contract NAS8--39073) during the course of this research. VK was
supported by NASA grants NAG5--3173, NAG5--3189, NAG5--3195, NAG5--3196
and NAG5--3831.
References
reference Randich, S., Giampapa, M.S., & Pallavicini, R., 1994, A&A, 283, 893