Документ взят из кэша поисковой машины. Адрес оригинального документа : http://star.arm.ac.uk/~ambn/367b.ps
Дата изменения: Thu Dec 13 17:05:45 2001
Дата индексирования: Mon Oct 1 20:22:03 2012
Кодировка:
Astronomy & Astrophysics manuscript no.
(will be inserted by hand later)
Variability in the extreme helium star LSS 5121 ?
V. M. Woolf 1 , R. Aznar Cuadrado 1 , G. Pandey 2 , and C. S. Je ery 1
1 Armagh Observatory, College Hill, Armagh BT61 9DG, Northern Ireland
email: vmw@star.arm.ac.uk, rea@star.arm.ac.uk, csj@star.arm.ac.uk
2 Department of Astronomy, University of Texas, Austin, TX 78712, USA
email: pandey@astro.as.utexas.edu
Abstract. We report a photometric and spectroscopic study of the hot extreme helium star LSS 5121. We
found photometric variability, but no period was evident in its periodogram. This is consistent with the previous
proposal, based on spectral line variations, that LSS 5121 is a non-radial pulsator similar to other hot extreme
helium stars.
Key words. stars: chemically peculiar { stars: oscillations { stars: variables { stars: individual: LSS 5121
1. Introduction
Extreme helium stars (EHes) are luminous blue stars with
spectra that indicate surfaces composed almost entirely of
helium, with traces of carbon and nitrogen and very little
(usually  1 per cent) hydrogen. Their temperatures fall
in the range 8000K  T  32 000K. The EHes are of in-
terest because, among other things, some of them pulsate
or are otherwise variable, they have very unusual chemical
compositions, and they are quickly evolving, with the radii
and temperature changes of some being measurable over
a period of a decade or less (Je ery et al. 2001). Although
their evolutionary history is still not completely certain,
it is evident that they are post AGB stars evolving to be-
come white dwarfs. EHes evolution may be linked with
other hydrogen de cient star classes such as RCrB stars
and helium subdwarf B stars (He sdBs). It may be that a
sizable fraction of stars go through an EHe phase.
Pulsations in EHes are excited through iron group (Z-
bump) opacities (Saio 1993). The pulsations in some EHes
allow accurate estimates to be made of their physical prop-
erties, including their mass and radii (Lynas-Gray et al.
1984, 1987, Woolf & Je ery 2000). The study of pulsations
in stars with atmospheres not dominated by hydrogen (e.g.
EHes and He sdBs) provides additional constraints on pul-
sation theory. Fewer than 30 EHes are known (Je ery et al.
1996) and few of those have been studied in any detail. The
physical properties and the variability of additional EHes
Send o print requests to: V. M. Woolf
? Based on observations made at the University of Texas
McDonald Observatory, Fort Davis, Texas and the Jacobus
Kapteyn Telescope operated on the island of La Palma by the
Isaac Newton Group in the Spanish Observatorio del Roque de
los Muchachos of the Instituto de Astro sica de Canarias.
must be determined if a statistical study of the group is
to have meaning.
LSS 5121 is a hot extreme helium star. The tempera-
ture of LSS 5121 has been estimated as 28 300 K (Heber
et al. 1986) using UBV color measurements and 29 772 K
(Je ery et al. 2001) by comparing optical photometry and
IUE ultraviolet spectra with the calculated theoretical
ux distributions. Earlier spectroscopic observations by
Lawson and Kilkenny (1998) demonstrated that its ab-
sorption lines vary in strength on a timescale of a few
days. The line variations are similar to those observed in
the (T  20 kK) EHe BD 9 ф 4395, a non-radial pulsator
(Je ery & Heber 1992). The spectrum of LSS 5121 is very
similar to that of the (T  29 kK) EHe HD 160641 (Heber
et al. 1986), which has been found to pulsate non-radially
with several pulsation periods which range between 8.4
hours and 1.77 days (Lynas-Gray et al. 1987). The spec-
tral similarity indicates similar temperatures, gravities,
and chemical compositions. The gravity of LSS 5121 has
not been determined, but it is clear from its spectrum
that its lies between those of the EHes HD 160641 and
LS IV+6 ф 2. As BD 9 ф 4395, LSS 5121 and HD 160641 all
lie near or above the Z-bump instability boundary (Saio
& Je ery 1999) it would be reasonable to suppose that
LSS 5121 may also pulsate. Given the similarity of the
physical properties of LSS 5121 and HD 160641, one might
also expect LSS 5121 to pulsate non-radially. EHes with
physical properties like those of LSS 5121 tend to vary
on timescales of hours to days, rather than minutes or
months. Our observations were designed to detect vari-
ability on the expected timescales.
LSS 5121 was too dim, mV = 13:3, for Lawson and
Kilkenny (1998) to test for photometric variations with the

2 V. M. Woolf, R. Aznar Cuadrado, G. Pandey, and C. S. Je ery: LSS 5121
Table 1. Observations
Telescope HJD 2 451 000
JKT 677.5626 { 677.7374
JKT 678.5567 { 677.7344
McD 0.8-m 706.8056 { 706.9503
McD 0.8-m 707.9320 { 707.9538
McD 0.8-m 709.7630 { 709.9532
McD 0.8-m 710.7639 { 710.9643
McD 0.8-m 711.7452 { 711.9635
McD 0.8-m 717.7015 { 717.9539
McD 2.1-m 709.7752 { 709.9544
McD 2.1-m 710.9034 { 710.9525
McD 2.1-m 711.7642 { 711.9420
0.5 m telescope used in their observations. Although pre-
liminary observations of LSS 5121 at the Jacobus Kapteyn
Telescope (JKT) in 1999 June showed probable variabil-
ity, LSS 5121 was not the primary target for the observ-
ing run, so the observations did not cover enough time for
the period search to be conclusive (Aznar Cuadrado et al.
2000). The observations we now report were carried out
in 2000 May and June to provide better measurements of
the star's variability.
2. Observations and data reduction
2.1. JKT photometry
Photometric measurements of LSS 5121 were performed
using the JAG-CCD camera on the 1-m JKT at the Roque
de los Muchachos Observatory, La Palma on the nights
of 2000 May 13 and 14. Observations were made using
the Cousins V lter. Pixels on the 2048  2048 chip cov-
ered 0.33 arcsec per pixel, giving a usable eld of about
10 arcmin. Each night the observations covered about 4.2
hours, with exposures being taken every 3 or 4 minutes.
2.2. McDonald photometry
LSS 5121 was observed on the nights of 2000 June 10,
11, 13, 14, 15, and 21 using the Prime Focus Corrector of
the McDonald Observatory 0.8-m telescope. Observations
were made using the Bessel V lter. The camera gives 1.3
arcsec per pixel using the 20482048 pixel CCD, giving a
46 arcminute eld. As LSS 5121 is in a fairly crowded eld,
we were able to reduce the portion of the chip read out,
thus reducing the readout time and disk space require-
ments, while still including enough comparison stars in the
observation. The photometric data from both McDonald
and JKT were obtained for use as di erential photometry.
The times covered by observations at JKT and McDonald
are listed in Table 1.
2.3. McDonald spectroscopy
Spectra of LSS 5121 were obtained on the nights of 2000
June 13, 14, and 15 using the Sandiford Echelle spectro-
graph (McCarthy et al. 1993) on the 2.1-m Otto Struve
Fig. 1. Di erential photometry from the JKT and McDonald
runs described in Table 1.
Telescope. Repeated 15 to 20 minute exposures were made
during the times listed in Table 1 to allow us to de-
tect changes in the spectrum due to pulsation. Spectra
typically had a signal to noise ratio between 7 and 11.
The wavelength range of the echelle orders overlapped, so
the spectra give continuous coverage between 4410 and
4960  A. We used a 1.65 arc sec slit width, which gives a
spectral resolution of =  40 000.
2.4. Data reduction
Photometric data were reduced using standard iraf pack-
ages to subtract bias and divide by ats. iraf's phot
package was used to do aperture photometry on the im-
ages. Stars in uncrowded parts of the eld and falling on
parts of the CCD without bad pixels or columns were cho-
sen for use as comparison stars for the di erential photom-
etry by visually inspecting the images.
Spectra were reduced using standard iraf packages
for bias and at eld correction, reducing echelle orders
to one dimensional spectra, and applying the wavelength
scale using thorium-argon arc spectra. Velocity corrections
for Earth's motion were found for each exposure using
rvcorrect and were applied using dopcor.
3. Analysis
3.1. Di erential photometry
Comparison stars were chosen for each night's data.
To be chosen, a star needed to be present on a good
part of the chip for every LSS 5121 image for the
night. The ux measured for LSS 5121 and the com-
bined ux measured for the comparison stars were
used to nd mag = mV(LSS 5121) mV (comparison).
Corrections were then made to put all the data on the
same scale, taking into account that di erent reference
stars were used on separate nights. Di erential photome-
try for the two observing runs are shown in Fig. 1.

V. M. Woolf, R. Aznar Cuadrado, G. Pandey, and C. S. Je ery: LSS 5121 3
Fig. 2. Radial velocities measured for LSS 5121.
3.2. Radial velocities
Velocity shifts were measured using the cross correlation
package fxcor in iraf. The spectra were coadded to pro-
vide a template with a signal to noise ratio of about 35.
Only the two echelle orders covering 4460 to 4500  A and
4880 to 4930  A had features, notably the 4471.5, 4920.8
and 4921.9  A He i lines, strong enough for the cross cor-
relation to yield a velocity. We gave more weight to the
velocities found using the longer wavelength order, as the
order containing the 4471.5  A line had lower signal. Other
orders contained lines that may be useful for velocity de-
termination in studies using larger telescopes, but which
did not give reliable cross correlation peaks with our spec-
tra. The velocities thus found were applied and the result-
ing spectra were coadded to give a template with smaller
velocity smearing for a second iteration. Velocities found
for three nights are displayed in Fig. 2. We estimate un-
certainties in radial velocity determinations to be about
10 km s 1 . This is a rough estimate based on how accu-
rately we could determine the center of the cross correla-
tion peak.
4. Results
The uncertainties for the radial velocity data are large
compared to di erences between the calculated velocities:
we cannot say that radial velocity variations are obvious.
The di erential photometry shows evidence that LSS 5121
varies on a timescale of a few hours, however no period was
immediately apparent. The photometry measured at JKT
shows a magnitude change of about 0.04 mag over a period
of about 2.5 hours. The photometry measured the next
month at McDonald shows similar variation, though the
data are noisier. Uncertainties for the photometric data
varied with the weather, but in general we estimate uncer-
tainties of 0:003 mag for the JKT data and 0:015 mag
for the McD data. Noise was reduced by nding the mean
for the photometric data in 0.02 day bins (Figs. 3 & 4).
Tests with non-variable comparison stars with magnitudes
Fig. 3. Di erential photometry for LSS 5121. Small points are
measured photometry values. Large points indicate average
photometry in 0.02 day bins, with error bars showing standard
deviation of the mean in each bin. Top two panels are JKT
data. Bottom two are McDonald data. Error bars are smaller
than the large data points for the JKT data.
similar to LSS 5121 gave di erential photometry which did
not vary more than about 0.01 mag.
Periodograms and window functions were found for
the photometric and velocity data using dipso's pdgram
and pdgwin features (Scargle 1982). The velocity peri-
odogram has no major peaks, which is not surprising,
given the small number of data points. The photome-
try periodogram has no major peaks that are not also
in the data's window function (Fig. 5). The major peaks
in both are near multiples of the one day alias. Data were
phased to the periods indicated by minor peaks in the
periodogram not present in the window function, but this
provided no further evidence of periodic behavior. We can-
not rule out that LSS 5121 varies with a period of about
1 day or that it varies with longer periods and/or smaller
amplitudes than our observations could detect.

4 V. M. Woolf, R. Aznar Cuadrado, G. Pandey, and C. S. Je ery: LSS 5121
Fig. 4. McDonald di erential photometry. Symbols as in
Fig. 3.
Fig. 5. Periodogram (top panel) and window function (bottom
panel) for photometric data.
5. Discussion and Conclusions
Di erential photometry shows that LSS 5121 varies in
brightness, with changes of up to 0.04 mag over a span of
2.5 hours. LSS 5121 was previously known to have vary-
ing spectral lines. We have now shown that it varies in
Fig. 6. Instability boundary (solid line) in the log g log Te
plane for 0.7 M helium stars with X = 0:00 and Z = 0:01.
EHes are marked with circles. The properties of the EHe
LSS 5121 fall in the hatched area. Three He sdB stars are
shown by squares. Stars known to vary are indicated by lled
symbols.
brightness as well. Analysis of data from observations on
8 nights in 2000 May and June did not yield evidence of
any periodicity in the variations. Our search for variabil-
ity in radial velocity was inconclusive. A successful radial
velocity search will require a larger telescope to provide
higher signal to noise spectra.
Saio (1995) showed that the region of the HR diagram
where pulsations should occur overlaps part of the region
where EHes appear. Figure 6 shows the location of all
known EHes with known T e and log g, and the Fe-group
opacity instability boundary for 0.7 M helium stars in
the log g log T e plane. The EHes mentioned in the text
are labeled in the gure. Helium stars above the boundary
should be unstable against pulsation. The exact location
of the instability boundary depends on stellar mass and
chemical composition.
In general, the observed pulsation timescales in EHes
decrease with higher temperature (Saio & Je ery 1988).
Two EHes, LSS 3184 (BX Cir) and V652 Her pulsate
in their fundamental radial modes with well determined
periods of about 2.6 hours (Kilkenny et al. 1996, 1999).
Pulsations in the coolest EHes (e.g. BD+1 ф 4381) seem to
be non-periodic or chaotic with variation timescales of 8
to 25 days (Lawson et al. 1993). Other EHes with known
variability seem to pulsate non-radially, some with multi-
ple periods. For example, Lynas-Gray et al. (1987) report
nding periods of 0.35, 0.71, 1.12, and 1.77 days in the
photometry of HD 160641, an EHe with a temperature
similar to that of LSS 5121 and Je ery et al. (1985) re-
port 3.5 day, 11.2 day, and possibly additional periods
for BD 9 ф 4395. It is possible that chaotic pulsations also
a ect some of the hotter EHes, so that some of the mul-
tiple periods found for the hot non-radial pulsators would
more accurately be referred to as variability timescales or
\quasi-periods," and that di erent periods or no periods

V. M. Woolf, R. Aznar Cuadrado, G. Pandey, and C. S. Je ery: LSS 5121 5
would be found if data from longer, more continuous ob-
servations were available.
Any pulsations of LSS 5121 are either non-periodic or
have periods which our data cannot resolve. Our inability
to nd a period is consistent with LSS 5121 being a non-
radial or a chaotic pulsator, though it does not constitute
proof. Its pulsations are probably similar to those of the
hot extreme helium stars BD 9 ф 4395 and HD 160641,
both of which have been reported to pulsate non-radially
with multiple periods (Je ery & Heber 1992, Lynas-Gray
et al. 1987).
Acknowledgements. We acknowledge nancial support from
the UK PPARC (grant Refs PPA/G/S/1998/00019 and
PPA/G/O/1999/00058). VW thanks Marcel Bergmann for in-
struction on use of the 0.8-m telescope and Nairn Baliber for a
refresher on photometry reduction. GP received support from
grant AST 9618414 from the NSF.
References
Aznar Cuadrado R., Monta~nes Rodrgues P., Je ery C.S., 2000,
In Szabados L., Kurtz D. (eds.) The Impact of Large-Scale
Surveys on Pulsating Star Research, ASP Conf. Ser. 203,
438
Heber U., Jonas G., Drilling J. S., 1986, In Hunger K.,
Schonberner D., Rao N. K. (eds.) Hydrogen De cient Stars
and Related objects, IAU Coll. 87, 69
Je ery C. S., Heber U., 1992, A&A 260, 133
Je ery C. S., Heber U., Hill P. W., et al., 1996, In Je ery C.
S., Heber U. (eds.) Hydrogen De cient Stars, ASP Conf.
Ser. 96, 471
Je ery C. S., Skillen I., Hill P. W., et al., 1985, MNRAS 217,
701
Je ery C. S., Starling R. L. C., Hill P. W., et al., 2001, MNRAS
in press
Kilkenny D., Koen C., Je ery C. S., et al., 1999, MNRAS 310,
1119
Kilkenny D., Lynas-Gray A. E., Roberts G., 1996, MNRAS
283, 1349
Lawson W. A., Kilkenny D., 1998, Observatory 118, 1
Lawson W. A., Kilkenny D., van Wyk F., et al., 1993, MNRAS
265, 351
Lynas-Gray A. E., Schonberner D., Hill P. W. et al., 1984,
MNRAS 209, 387
Lynas-Gray A. E., Kilkenny D., Skillen I., et al., 1987, MNRAS
227, 1073
McCarthy J. K., Sandiford B. A., Boyd D., et al., 1993, PASP
105, 881
Saio H., 1993, MNRAS 260, 465
Saio H., 1995, MNRAS 277, 1393
Saio H., Je ery C. S., 1988, ApJ 328, 714
Saio H., Je ery C. S., 1999, MNRAS 208, 221
Scargle J. D., 1982, ApJ 263, 835
Woolf V. M., Je ery C. S., 2000, A&A 358, 1001