1: \documentclass[12pt,preprint]{aastex}
2: %\documentclass[preprint2]{aastex}
3:
4: \shorttitle{Hot subdwarfs in the Galactic Bulge}
5: \shortauthors{G.Busso et al.}
6:
7: \begin{document}
8:
9: \title{Hot subdwarfs in the Galactic Bulge
10: \footnote{Based on observations collected at the European Southern
11: Observatory, Chile (ESO proposal 73.D-0168A) }}
12:
13: \author{G.Busso\altaffilmark{1}, S.Moehler \altaffilmark{1},
14: M.Zoccali\altaffilmark{2}, U.Heber\altaffilmark{3} and S.K.Yi\altaffilmark{4}}
15: \altaffiltext{1}{Institut f\"{u}r Theoretische Physik und Astrophysik der
16: Universitaet Kiel, 24098 Kiel, Germany; busso@astrophysik.uni-kiel.de, moehler@astrophysik.uni-kiel.de}
17:
18: \altaffiltext{2}{Departamento de Astronomia y Astrofisica, Pontificia
19: Universidad Catolica de Chile, Avenida Vucuna Mackenna 4860, 782-0436 Macul,
20: Santiago, Chile; mzoccali@astro.puc.cl}
21:
22: \altaffiltext{3}{Dr.Remeis-Sternwarte, Astronommisches Institut der
23: Universitaet Erlangen-Nurnberg, Sternwartstr. 7, 96049 Bamberg, Germany; heber@sternwarte.uni-erlangen.de}
24:
25: \altaffiltext{4}{Department of Physics, University of Oxford, Keble Road,
26: Oxford OX1 3RH, UK; yi@astro.ox.ac.uk}
27:
28: \begin{abstract}
29: Recent observations and theories suggest that extreme horizontal branch (EHB)
30: stars and their progeny should be the cause of the UV excess seen in the
31: spectra of many elliptical galaxies. Since the Galactic
32: Bulge is the closest representation of an old, metal-rich spheroid in which we
33: are able to study the EHB scenario in detail, we obtained spectra of bulge EHB
34: star candidates and we confirm their status as hot evolved stars. It is the
35: first time that such stars are unambiguously identified in the Galactic Bulge.
36: \end{abstract}
37:
38: \keywords{UV excess; Galactic Bulge; Extreme Horizontal Branch stars: general}
39: \maketitle
40:
41: \section{Introduction}
42: The spectra of elliptical galaxies and bulge regions of spiral galaxies in
43: many cases show a strong and unexpected increase in flux at wavelengths shorter
44: than 2500~\AA. This "UV excess" was one of the most important
45: discoveries of satellite based UV astronomy \citep{code69} but also a puzzle,
46: since it requires the existence of hot stars in these old metal-rich
47: systems. After a long debate most people agree that the observed UV radiation is mainly produced by very hot extreme
48: horizontal branch stars (burning helium in their core) and their progeny, as
49: Post-EarlyAGB and AGB-manqu\'e stars (O'Connell 1999; Greggio \& Renzini 1990,
50: 1999; Dorman et al. 1995; Yi et al. 1998).
51: This view is supported by spectroscopic (Ferguson et al. 1991; Brown et
52: al. 1997, 2002) and photometric \citep{brown02} UV observations of
53: extragalactic systems.
54: Near-UV HST observations of Brown et al.~(2000) in M32 detected for
55: the first time individual EHB star candidates in an elliptical galaxy.
56: The best fit to these observations is achieved with evolutionary tracks for
57: helium- and metal-rich populations, since in this case EHB stars have the
58: longest lifetimes in the temperature range required to reproduce the UV
59: excess.
60:
61: The closest system similar to an elliptical galaxy with respect to age and
62: metallicity, for which it is possible to resolve stars, is the Galactic Bulge.
63: The vast majority of EHB stars known in the Milky Way, however, belongs to the
64: metal-poor globular clusters \citep{sabine01} or to the disk population, where
65: they show up as so-called subdwarf B [sdB] stars \citep{heber86,saffer94,ville95,altmann04}
66:
67: The first sdB candidates in the bulge were found in the two massive
68: globular clusters NGC 6388 and NGC 6441 \citep{rich97,io04}
69: which are, however, not typical for the bulge population. The situation
70: changed recently: the imaging surveys of both Terndrup et al. (2004) and
71: Zoccali et al. (2003) of bulge fields show a sequence of hot stars that are
72: good candidates for EHB stars (see Fig.~1).
73:
74: \subsection{Disk Stars or Bulge Stars?}
75: As the line-of-sight towards the Galactic bulge passes through the disk it is
76: important to estimate the expected number of disk sdB stars in the
77: observations. To do so we used the values of Villeneuve et al.~(1995)
78: for the space density of local field sdB stars (2--4 $\times$ 10$^{-7}$
79: pc$^{-3}$) to derive the expected number of sdB stars along our
80: line of sight from 4.5 kpc (corresponding to $I\approx$ 18.5) to 11 kpc
81: ($I\approx$ 20.5) within the field of view of the Wide Field Imager (WFI, 30\arcmin
82: $\times$ 30\arcmin). The sdB stars in the field of the Milky Way consist of a
83: mixture of thin and thick disk stars (e.g. Altmann et al.~2004), so we
84: assumed a ratio of 50:50. For the thin disk we used a scale height of
85: 325~pc and a scale length of 3.5 kpc and for the thick disk we used
86: values of 900 pc and 4.7 kpc, respectively \citep{larsen03}.
87: We used a distance to the Galactic center of 8.5~kpc and assumed that
88: the disk ends at a radial distance of 1~kpc from the Galactic Center (Robin et
89: al.~2003).
90: This way we predict a total number of 4 to 9 sdB foreground stars within the
91: full field of WFI. Using the values of Ojha (2001) for the scale lengths
92: (2.8~kpc and 3.5~kpc) instead, we expect 7 to 14 foreground sdB stars.
93: However, we detect many more bulge candidates (about 140) in the WFI
94: photometry of Zoccali et al.~(2003).
95: Since they could be cool foreground stars with low reddening
96: (instead of reddened hot stars), we
97: obtained spectroscopy of 29 candidates in order to derive effective
98: temperatures and surface gravities and then, by means of comparison
99: with HB models, to check their evolutionary status.
100:
101: \section{Observations and Data Reduction}
102:
103: Our spectroscopic targets were selected from the photometric catalogue of
104: bulge stars obtained from Zoccali et al.~2003 (see Fig.~1). The field is
105: located toward the Galactic center, at $l$=$0^{\circ}, b$=$-6^{\circ}$, where
106: the average reddening is E$_{\rm B-V}$=0.45 (Zoccali et al.~2003).
107: We have chosen the stars with $18 < I < 21$ and $0 < V-I <
108: 0.8$ and among them we selected the most isolated ones.
109: After positioning as many slitlets as possible on EHB star
110: candidates, the remaining ones were used to get spectra of cool main
111: sequence and red giant stars that will be helpful for constructing the
112: overall spectrum expected for these bulge regions.
113:
114: We obtained medium-resolution spectra ($R \approx 1200$) of 29 EHB
115: star candidates at the VLT-UT1 (Antu) with FORS2. We used the multi-object
116: spectroscopy (MXU) mode of FORS2 with the standard collimator, a slit width of
117: 0.7" and grism B600, which allows to obtain spectra in the range between 3650
118: and 5200 \AA ~(not all candidates though have full spectral coverage because
119: of the different positions on the CCD).
120:
121: The data reduction was performed as described in Moehler et al.~(2004)
122: except for the following points. Due to the long exposure times (from 2700s to
123: 5400s) the scientific observations contained a large number of cosmic rays and
124: were therefore corrected with the algorithm described in Pych~(2004).
125: Regarding the subtraction of the sky background, we used two different
126: methods depending on whether the target star in the slitlet was isolated or
127: not. If the star was isolated, meaning any other stars in the slitlets were
128: well enough separated from our target to identify regions uncontaminated by
129: any stellar source, we approximated the spatial distribution of the sky
130: background by a constant. If the slitlet showed severe crowding, meaning that
131: the spectra of different stars were overlapping, we fitted each stellar
132: profile with a Lorentzian function so that the whole spatial profile was
133: reproduced by the sum of all the profiles; all profiles but that one of
134: the target were then subtracted (for details see Moehler \& Sweigart, 2006).
135: With the extraction of the spectra, we saw that some (5 of 29) of our targets
136: were actually cool stars. Therefore we did not proceed further with the
137: reduction for these stars.
138: The spectra were flux calibrated using standard star spectra and corrected
139: for any Doppler shifts determined from Balmer lines, as in Moehler et
140: al.~(2004).
141:
142: \section{Analysis}
143:
144: Some examples of the spectra are shown in Fig.~2. The spectra of
145: the hot stars show evidence for high reddening like a strong
146: \ion{Ca}{2} K line and the diffuse interstellar band at 4430~\AA.
147: To fit the spectra (except for one He-rich star) we used ATLAS9 model atmospheres
148: for solar metallicity (Kurucz 1993) to account for effects of radiative
149: levitation (see Moehler et al.~2000 for details), from which we calculated
150: spectra with Lemke's version of the
151: LINFOR program (developed originally by Holweger, Steffen, and Steenbock at
152: Kiel University). The use of NLTE models or of LTE models with higher
153: metallicity does not significantly change the results.
154:
155: To establish the best fit, we used the routines developed by Bergeron
156: et al.~(1992) and Saffer et al.~(1994), as modified by Napiwotzki et
157: al.~(1999), which employ a $\chi^2$ test. The uncertainty necessary for the
158: calculation of $\chi^2$ is estimated from the noise in the continuum
159: regions of the spectra. The fit program normalizes model spectra and
160: observed spectra using the same points for the continuum definition.
161: We used the Balmer lines $H_{\beta}$ to $H_{10}$ (excluding
162: $H_\epsilon$ to avoid the Ca~II H line), the He~I lines at 4026,
163: 4388, 4471, 4921 \AA, and the He~II lines at 4542 and 4686 \AA.
164:
165: We obtained the atmospheric parameters $T_{eff}$, log$g$ and helium
166: abundances and we calculated the absolute V and I magnitudes expected
167: for these values, assuming $M_{star}=0.5M_{\odot}$. We left out one
168: star because the fit was unacceptably poor. Since the formal fit
169: errors are underestimated by a factor 2--4 (Napiwotzki, priv. comm.) a
170: formal error of 0.1 in log$g$ implies an error of 25\%--50\% in the
171: distance. We therefore discuss the bulge membership only for those
172: stars with a formal error in log$g$ of less than 0.1 because for the
173: others (4 of 23) the uncertainty in the distance is too large. Then
174: considering a distance from the Galactic Center of $\approx$ 8.5~kpc
175: and a bulge radius of $\approx$ 1.5 kpc, we find that most of these
176: objects are indeed bulge stars: of 19 hot stars with reasonable errors $\sigma$
177: in the distance, 13 stars are in the bulge within 1 $\sigma$ and 3
178: more are in the bulge within 3 $\sigma$. We thus found 3 probable disk EHB
179: stars in our sample of 29 candidate EHB stars, which corresponds to
180: a contamination of 10\%.
181:
182: The heliocentric radial velocities also suggest a bulge membership of
183: most EHB stars. The field where we are looking is at Galactic
184: coordinates $l$=$0^{\circ}, b$=$-6^{\circ}$, toward the Galactic
185: center, i.e. the expected radial velocities for disk stars are around
186: 0 km~s$^{-1}$. Our velocities are distributed in a range between
187: $-$200 and $+$300~km~s$^{-1}$, in agreement with the values found for
188: K giants in Baade's Window by Terndrup et al. (1995, between $-$240
189: and $+$194~km~s$^{-1}$). We calculated a velocity dispersion of 110
190: $\pm$ 17 km s$^{-1}$ from our bulge EHB stars: the expected value for
191: the disk is 50--70 km s$^{-1}$ (Lewis \& Freedman 1989) while Terndrup
192: et al. (1995) found for Baade's Window, at a distance of 8kpc, 80--110
193: km s$^{-1}$.
194:
195: Finally we compare our results with horizontal branch theoretical tracks: in
196: Fig.~3 we plot the values found for those stars which belong to the bulge in
197: the ($T_{eff}$, $\log g$) diagram. The error bars are the
198: formal errors from the fit procedure, but, as we already mentioned, these
199: errors are underestimated and in addition, they do
200: not include any systematic uncertainties, due to, e.g. sky
201: subtraction, flux calibration, etc.
202: The evolutionary tracks are from Yi et al.~(1997) with metallicity Z=0.004 and
203: helium abundance Y=0.2416.
204: The Zero Age Horizontal Branch (ZAHB), where the star starts to burn helium in
205: its core quietly, and the Terminal Age Horizontal Branch (TAHB), where the star
206: has burned the 99\% of the helium, are shown together with evolutionary tracks
207: for stars with total masses of 0.49, 0.50, and 0.51 $M_{\odot}$ (respectively
208: $M_{env}=$ 0.0075, 0.0127, 0.0226 $M_{\odot}$).
209: Our tracks end at 0.495 $M_{\odot}$, corresponding to $M_{env}=$ 0.0075 $M_{\odot}$; since EHB stars may have $M_{env} <$ 0.005
210: $M_{\odot}$, we extrapolated the ZAHB and TAHB to higher
211: temperatures (dashed curves in Fig.~3) to guide the eye.
212: Proper models for lower envelope masses will be calculated and used in
213: a later paper.
214: The observed points agree quite well with the theoretical tracks, therefore
215: these objects are indeed EHB stars; some objects are above the TAHB
216: meaning that they are in the post-HB phase and then evolving as AGB-manqu\'e
217: stars (Greggio \& Renzini~1990).
218:
219: Finally we want to mention that all stars except one (which is helium-rich)
220: are helium deficient as expected from diffusion.
221:
222: \section{Conclusions}
223:
224: We observed spectra of 29 EHB star candidates in the Galactic Bulge, from which
225: we estimated the atmospherical parameters $T_{eff}$ and log$g$ to verify
226: their evolutionary status. Five objects turned out
227: to be cooler foreground stars with low reddening, and for another one
228: the spectroscopic fit is unacceptably bad. Of the 19 hot
229: stars with reasonable distance errors
230: 16 lie within a radius of 1.5~kpc around the Galactic center at
231: 8.5~kpc. Also the observed radial velocities support a membership of
232: these stars to the bulge.
233: This is the first time that of stars are observed in the bulge and
234: our statistics show that either spectroscopy or multi-colour photometry is
235: required to disentangle hot stars in the bulge from other sources.
236: We will use these spectra and the parameters derived from them
237: to construct the integrated spectrum of the galactic bulge from the UV to the
238: optical, following the method of Santos et al.~(1995).
239: This study will verify the role, so far only predicted, of these stars
240: regarding the UV excess in the elliptical galaxies.
241:
242: \acknowledgments
243: We are grateful to the ESO staff, especially those at the Paranal
244: observatory, for all their help with the observations. We thank an
245: anonymous referee for pertinent comments, which helped us to improve
246: this paper. GB acknowledges support from the Deutsche
247: Forschungsgemeinschaft via grant Mo 602/8.
248:
249: \begin{thebibliography}{}
250: \bibitem [Altmann et al. 2004]{altmann04}
251: Altmann, M., Edelmann, H., de Boer, K. S. 2004, A\&A 414, 181
252: \bibitem[Bergeron et al. 1992]{bergeron92}
253: Bergeron, P., Saffer, R.A., Liebert, J. 1992, \apj, 394, 228
254: \bibitem[Brown et al. 1997]{brown97}
255: Brown, T.M., Ferguson, H.C., Davidsen, A.F., Dorman, B. 1997, \apj, 482, 685
256: \bibitem[Brown et al. 2000]{brown00}
257: Brown, T.M., Bowers, C.W., Kimble, R.A., Sweigart, A.V., Ferguson, H.C. 2000, \apj, 532, 308
258: \bibitem[Brown et al. 2002]{brown02}
259: Brown, T.M., Ferguson, H.C., O'Connell, R.W., Ohl, R.G. 2002, \apjl, 568, 19
260: \bibitem[Busso et al. 2004]{io04} Busso, G., Piotto, G., Cassisi, S. 2004, \memsai, 75, 46
261: \bibitem[Code \& Welch 1969]{code69} Code, A.D., Welch, G.A. 1969, \apj,
262: 228, 95
263: \bibitem[Dorman et al. 1995]{dorman95} Dorman, B., O'Connell, R.W., Rood, R.T., 1995, \apj, 442, 105
264: \bibitem[Ferguson et al. 1991]{fergus91} Ferguson H.C., Davidsen A.F., Kriss,
265: G.A. et al. 1991, \apjl, 382, 69
266: \bibitem[Greggio \& Renzini 1990]{greggio90} Greggio,L.
267: \& Renzini, A. 1990, \apj, 364, 35
268: \bibitem[Greggio \& Renzini 1999]{greggio99} Greggio,L. \&
269: Renzini, A. 1999, \memsai, 70, 691
270: \bibitem[Heber 1986]{heber86}
271: Heber, U. 1986, A\&A, 155, 33
272: \bibitem[Kurucz 1993]{kuru93} Kurucz, R.L. ATLAS9 Stellar
273: Atmospheric Program, \url{http://kurucz.harvard.edu}
274: \bibitem[Larsen \& Humphreys 2003]{larsen03}
275: Larsen, J.A., Humphreys, R.A. 2003, AJ, 125, 1958
276: \bibitem[Moehler et al. 2000]{sabine00} Moehler, S., Sweigart, A.V., Landsman, W.B., Heber, U. 2000, \aap, 360, 120
277: \bibitem[Moehler 2001]{sabine01} Moehler, S. \pasp, 113, 1162
278: \bibitem[Moehler et al. 2004]{sabine04} Moehler, S., Sweigart,
279: A.V., Landsman, W.B., Hammer, N.J., Dreizler, S. 2004, \aap, 415, 313%
280: \bibitem[Moehler \& Sweigart 2006]{sabine06} Moehler, S. \&
281: Sweigart, A.V. 2006, to appear in Proceedings of the 2$^{nd}$ Meeting
282: on Hot Subdwarf Stars and Related Objects, Baltic Astronomy
283: \bibitem[Napiwotzki et al. 1999]{napi99} Napiwotzki, R., Green, P.J., Saffer, R.A. 1999, \apj, 517, 399
284: \bibitem[O'Connell 1999]{oconnell99} O'Connell, R.W 1999, \araa, 37, 603
285: \bibitem[Ojha 2001]{ojha01}
286: Ojha, D.K. 2001, MNRAS, 322, 426
287: \bibitem[Pych 2004]{pych04} Pych, W. 2004, \pasp, 116, 148
288: \bibitem[Rich et al. 1997]{rich97} Rich, R.M., Sosin, C., Djorgovski, S.G. et
289: al. 1997, \apjl, 484, 25
290: \bibitem[Robin et al.\ 2003]{robi03}
291: Robin, A. C., Reyl\'e, C., Derri\`ere, S., Picaud, S. 2003, A\&A, 409, 523
292: \bibitem[Saffer et al. 1994]{saffer94} Saffer, R.A., Bergeron, P., Koester, D., Liebert, J., 1994 \apj, 432, 351
293: \bibitem[Santos et al. 1995]{santos95} Santos, J.F.C., Bica, E., Dottori, H., Ortolani, S., Barbuy, B. 1995, \aap, 303, 753
294: \bibitem[Terndrup et al. 1995]{terndrup95} Terndrup, D.M., Sadler, E.M.,
295: Rich, R.M. 1995, \aj, 110, 1774
296: \bibitem[Terndrup et al. 2004]{terndrup04} Terndrup, D.M., An,
297: D., Hansen, A., Peterson, R.C., Walker, A.R., Sadler, E.M. \apss 291, 247
298: \bibitem[Villeneuve et al. 1995]{ville95} Villeneuve, B.,
299: Wesemael, F., Fontaine, G., Carignan, C. Green, R.F. 1995, \apj, 446, 646
300: \bibitem[Yi et al. 1997]{yi97} Yi, S.K., Demarque, P., Oemler, A.J. 1997,
301: \apj, 486,201
302: \bibitem[Yi et al. 1998]{yi98} Yi, S.K., Demarque, P., Oemler, A.J. 1998,
303: \apj, 492,480
304: \bibitem[Zoccali et al. 2003]{manu03} Zoccali, M., Renzini, A., Ortolani, S. et al. 2003, \aap, 399, 931
305:
306: \end{thebibliography}
307:
308: \clearpage
309:
310: \begin{figure}
311: \epsscale{.90}
312: \plotone{f1.eps}
313: \caption{Colour-magnitude diagram of the Galactic bulge (at $l$=$0^{\circ},
314: b$=$-6^{\circ}$, E$_{\rm B-V}$=0.45) obtained from the
315: Zoccali et al. (2003) observations. Our targets are marked with circles.}
316: \end{figure}
317:
318: \clearpage
319:
320: \begin{figure}[htbp]
321: \epsscale{1.100}
322: \plotone{f2.eps}
323: \caption{Examples of the spectra of the sdB star candidates: the spectrum at
324: the bottom is for a candidate which turned out to be a cool star and shown here for
325: comparison; the others are typical spectra we found, indicating hot stars with strong Balmer lines.}
326: \end{figure}
327:
328: \clearpage
329:
330: \begin{figure*}[htbp]
331: %\epsscale{1.40}
332: %\plotone{fit.eps}
333: \begin{center}
334: \includegraphics[angle=270,scale=.40]{f3.eps}
335: \end{center}
336: \caption{($T_{eff}$, $\log g$) diagram: the squares indicate the
337: isolated stars; the triangles indicate the crowded stars and the
338: pentagon is the He-rich star. The ZAHB and TAHB \citep{yi97} for
339: Z=0.004 and Y=-0.2416 are plotted together with
340: evolutionary tracks for 0.495, 0.50 and 0.51 $M_{\odot}$. The dashed lines are
341: extrapolated from the ZAHB and TAHB tracks.}
342: \end{figure*}
343:
344: \end{document}
345:
346: