1: % 5/10/06: Barbara e Giacomo
2: % 13/9: Barbara
3: %1-9/9 frf
4: % 29/8/06: Barbara
5: \documentclass[12pt,preprint]{aastex}
6: %\documentclass{emulateapj}
7: %\documentclass[preprint2]{aastex}
8: %\documentstyle[emulateapj,apjfonts,epsf,versioncopy]{article}
9: %\def\versionnumber{9} % <-- here the version number
10: %
11: % modified FRF Jan05
12: % modified AP May05
13: % modified RTR 23 June 05
14: % few more mods rtr 24 June
15: \newcommand{\gras}[1]{\mbox{\boldmath $#1$}}
16: \def\etal{{et al.~}}
17: \def\ltsima{$\; \buildrel < \over \sim \;$}
18: \def\gtsima{$\; \buildrel > \over \sim \;$}
19: \def\lsim{\lower.5ex\hbox{\ltsima}}
20: \def\gsim{\lower.5ex\hbox{\gtsima}}
21: \def\lapp{\ifmmode\stackrel{<}{_{\sim}}\else$\stackrel{<}{_{\sim}}$\fi}
22: \def\gapp{\ifmmode\stackrel{>}{_{\sim}}\else$\stackrel{<}{_{\sim}}$\fi}
23:
24: \newdimen\minuswidth %define @ width of minus sign for tables
25: \setbox0=\hbox{$-$}
26: \minuswidth=\wd0
27: \catcode`@=\active
28: \def@{\kern\minuswidth}
29: \newdimen\digitwidth %define ! a one digit width for tables
30: \setbox0=\hbox{\rm0}
31: \digitwidth=\wd0
32: \catcode`!=\active
33: \def!{\kern\digitwidth}
34:
35: \shorttitle{Binary stars in the AGB of 47 Tucanae?}
36: \shortauthors{Beccari et al.}
37:
38: \begin{document}
39:
40: \title{A population of binaries in the Asymptotic Giant Branch of 47
41: Tucanae?\footnote{Based on observations with the NASA/ESA HST, obtained at
42: the Space Telescope Science Institute, which is operated by AURA, Inc., under
43: NASA contract NAS5-26555. Also based on WFI observations collected at the
44: European Southern Observatory, La Silla, Chile, within the observing programs
45: 62.L-0354 and 64.L-0439.}}
46:
47: \author{Giacomo Beccari\altaffilmark{1,2,3},
48: Francesco R. Ferraro\altaffilmark{4}, Barbara Lanzoni\altaffilmark{1},
49: Michele Bellazzini\altaffilmark{1}}
50: \affil{\altaffilmark{1} INAF--Osservatorio Astronomico
51: di Bologna, via Ranzani 1, I--40127 Bologna, Italy,
52: giacomo.beccari@bo.astro.it}
53: \affil{\altaffilmark{2} Dipartimento di Scienze della Comunicazione,
54: Universit\`a degli Studi di Teramo, Italy}
55: \affil{\altaffilmark{3} INAF--Osservatorio Astronomico
56: di Collurania, Via Mentore Maggini, I--64100 Teramo, Italy}
57: \affil{\altaffilmark{4} Dipartimento di Astronomia, Universit\`a
58: degli Studi di Bologna, via Ranzani 1, I--40127 Bologna, Italy,
59: francesco.ferraro3@unibo.it}
60: \medskip
61:
62:
63: \keywords{Globular clusters: individual (47Tuc); stars: evolution
64: -- binaries: close - blue stragglers}
65:
66: \begin{abstract}
67: We have used a set of archived Hubble Space Telescope/ACS images to probe the
68: evolved populations of the globular cluster 47 Tucanae. We find an excess of
69: Asymptotic Giant Branch (AGB) stars in the cluster core. We interpret this
70: feature as the signature of an extra-population likely made by the progeny of
71: massive stars originated by the evolution of binary systems. Indeed the
72: comparison with theoretical tracks suggests that the AGB population of 47 Tuc
73: can be significantly contaminated by more massive stars currently
74: experiencing the first ascending Red Giant Branch.
75: \end{abstract}
76:
77: \section{INTRODUCTION}
78: \label{intro}
79: In recent years it became clear that stellar evolution and stellar dynamics
80: cannot be studied independently: in fact, physical interactions between
81: single stars and binaries, as well as the formation and evolution of binary
82: systems can significantly alter the properties of the overall stellar
83: populations. This is particularly true for dense stellar systems such as the
84: Globular Clusters (GCs), where dynamical encounters between stars (especially
85: those involving binaries) are most probable. In particular, such a dynamical
86: activity can generate {\it exotic} stellar populations, like Blue Straggler
87: Stars (BSSs), Millisecond Pulsars (MSPs), low-mass X-ray binaries (LMXBs),
88: Cataclysmic Variables (CVs), etc. \citep[see][]{ba95}, that cannot be
89: explained by the standard single-mass stellar evolution theory, and are
90: thought to be originated by the evolution of binary systems. For instance,
91: BSSs consist (or will consist, once the mass transfer or the merge event is
92: completed) in a progeny of mass-enhanced single stars (possibly with a He
93: white dwarf companion), with masses typically ranging between 1 and 2 times
94: the cluster Turn Off (TO) star mass. Then, these objects will evolve along
95: the characteristic paths of a rejuvenated mass-enhanced population, and are
96: therefore expected to "contaminate" the post-main sequence (MS) evolutive
97: sequences of normal cluster stars. However, when considering all possible
98: by-products of the binary evolution, a variety of different exotic objects
99: could be originated (depending on the mass, the evolutionary state of the
100: donor star, the mass ratio and the physical parameters of the binary system),
101: and their identifications in the Color Magnitude Diagram (CMD) may be not
102: obvious. The detection of surface chemical anomalies also seems to be a
103: quite promising tool for identifying by-products of binary evolution, but
104: high-resolution spectroscopic studies of this kind have just started to
105: become feasible \citep[see][and references therein]{fe06}.
106:
107: %Of course in collisional stellar system, binaries play a key role
108: %As pointed out by~\citet[][]{iv05}, there are very few direct
109: %measurements of binary fractions in clusters and anyway, the measured binary fractions
110: %in dense
111: %cluster cores are found to be very small. As an example, the upper limit on the core
112: %binary fraction of NGC 6397 is only 5-7 per cent~\citep[][]{co02}. On the other hand,
113: %in very sparse clusters, such as NGC 288~\citep[][]{be02}, but also in some
114: %other 'core-collapsed' clusters, such as NGC 6752~\citep[][]{ru97}, the upper
115: %limit for the binary fraction can be as high as 30 per cent.
116: %~\citet[][]{al01} searched for binaries
117: %in the central regions of 47 Tuc. They found a binary frequency of $14\% \pm 4\% $ in
118: %the core of the cluster. \\
119:
120: Indeed, 47 Tucanae has been found to harbor a large population of exotic
121: objects in its core \citep[see][]{g01,fe01,k02}, confirming that stars have
122: experienced (and are experiencing) dynamical interactions involving single
123: stars or primordial binaries, leading to mass transfer phenomena and mergers
124: \citep[see][hereafter F04; Mapelli et al. 2004; Ferraro et
125: al. 2006]{fe04}. In particular, the large population of BSSs found in the
126: cluster \citep[][F04]{par91,gu92, fe01} represents the major reservoir of
127: binary by-products currently experiencing the MS stage. Their bimodal radial
128: distribution, discovered by F04, suggests that most of the BSSs strongly
129: segregated in the cluster center have a collisional origin, while those lying
130: in its periphery are mainly originated by the evolution of primordial
131: binaries \citep[see][2006]{ma04}. Also the detection of chemical anomalies
132: via high-resolution spectroscopy has recently provided a number of
133: interesting results in the case of 47 Tuc: {\it (i)} \citet{fe06} have
134: detected a population of carbon-oxygen depleted BSSs and interpreted this
135: feature as the signature of the mass-transfer activity during the BSS
136: formation processes; {\it (ii)} \citet{w06} have detected peculiar {\it s}
137: and {\it r}-process elements enhancement on the surface of 7 Asymptotic Giant
138: Branch (AGB) stars in the external region ($r>10'$) of the cluster. They
139: proposed that at least part of these AGB stars might be formed through mass
140: transfer along the evolutionary path of a $1.4 + 0.5 M_{\odot}$ binary
141: system.
142:
143: %In this scenario plaied a key
144: %role the discover by F04 of the
145: %bimodal radial distribution of the BSS population in the cluster. As confirmed
146: %by~\citet[][]{ma04} through detailed numerical simulations,
147: %this radial distribution require that a sizable fraction of blue stragglers is
148: %generated in the peripheral regions of the cluster inside primordial binaries that
149: %evolve in isolation experiencing mass transfer.\\
150:
151: %
152:
153: In this Letter, we further investigate the issue of identifying possible
154: by-products of binary evolution in 47 Tuc, by reporting on the detection of
155: anomalies (both in terms of star counts and radial distribution) along the
156: canonical evolutionary sequences in the cluster CMD. In particular, we have
157: found a significant enhancement of AGB stars in the core and we discuss the
158: possibility that this is due to a contamination by a progeny of massive stars
159: originated by the evolution of binary systems, and now sunk to the cluster center
160: by mass segregation processes.
161:
162:
163: \section{Observations and data reduction}
164: \label{obs}
165: In this paper we use high-resolution photometric data obtained with the
166: Advanced Camera for Survey (ACS) on board the Hubble Space Telescope
167: (GO-9453). Observations were performed with the {\it ACS-Wide Field Channel},
168: which employs a mosaic of two $4096\times2048$ pixels CCD, providing a
169: plate-scale of $0.05\arcsec/pixel$ and a total field of view (FoV) of
170: $3.4'\times3.4'$. In these images the core of the cluster is fully contained
171: in $chip \# 1$ of the ACS CCD mosaic. Since we were interested in deriving an
172: accurate photometry of the cluster giant population, only images with very
173: short exposure times ($t_{exp}=0.5$ s) obtained through the filters F606W and
174: F814W were retrieved from the ESO/ST-ECF Science Archive. All the images
175: were properly corrected for geometric distortions and effective flux (over
176: the pixel area) following the prescriptions of \citet{si05}. The photometric
177: analysis was performed independently in the two drizzled images by using the
178: aperture photometry code SExtractor \citep[{\it Source-Extractor};] []{be96},
179: and adopting a fixed aperture radius of 2 pixels ($0.1\arcsec$). The two
180: magnitude lists were cross-correlated in order to obtain a combined catalog.
181: %We performed our
182: %aperture photometry using flt images. These images, affected by geometrical
183: %distortion, have been flat-fielded to preserve surface brightness, but not
184: %flux. Therefore, we multiplied each pixel by its projected area to obtain
185: %a proper flux for each star.
186: A sample of bright isolated stars has been used to transform the instrumental
187: magnitudes to a fixed aperture of $0.5\arcsec$, and the extrapolation to
188: infinite has been performed by using the values listed in Table 5 of
189: \citet{si05}.
190: %The final instrumental magnitudes were then reported to a fixed aperture of
191: %$0.5\arcsec$, and extrapolated to infinite by using the values listed in
192: %Table 5 of \citet{si05}.
193: The magnitudes were finally transformed into the VEGAMAG photometric system
194: by adopting the synthetic Zero Points from Table 10 of \citet{si05}.
195:
196: %In order to properly calibrate the magnitude we calculated, for each star, magnitude
197: %with aperture radius of 10 pixels (0.5 arcsec). After determining the aperture
198: %correction $AP_{0.1''}-AP_{0.5''}$ we used values from Table 5 of~\citet[][]{si05}
199: %which provide aperture correction from 0.5'' until ``infinite'' (AC05) for each
200: %filter of ACS camera.
201: %Finally we added the proper Zero Points from Table 10 of~\citet[][]{si05} to
202: %calibrate data in VEGAMAG photometric system. In Table~\ref{ap} we report the values
203: %of aperture corrections together with synthetic Zero Points for both filters.
204:
205: An additional data-set of public short-time exposure ($t_{exp}=20$ s) images
206: in $V$ and $I$ bands, obtained with the {\it Wide Field Imager} (WFI) mounted
207: at the 2.2m telescope at ESO, were also retrieved from the ESO/ST-ECF
208: Archive. The raw WFI images were corrected for bias and flat field, and the
209: over-scan region was trimmed using standard IRAF\footnote{IRAF is distributed
210: by the National Optical Astronomy Observatory, which is operated by the
211: Association of Universities for Research in Astronomy, Inc., under
212: cooperative agreement with the National Science Foundation.} tools. The PSF
213: fitting procedure was performed independently on each $V$ and $I$ image,
214: using DoPhot \citep{dophot}.
215:
216: Each WFI catalog was referred to the absolute astrometric system by adopting
217: the procedure described in F04. The astrometric Guide Star Catalog GSCII was
218: used to identify astrometric standards in the WFI FoV. More than thousand
219: stars were used to find an astrometric solution for each of the eight WFI
220: chips, with an accuracy of the order of $\sim0.2\arcsec$ in both right
221: ascension and declination. Then, we used $\sim1000$ stars in common between
222: WFI and ACS catalogs as {\it secondary astrometric standards} to transform
223: the ACS catalog into the WFI astrometric system. Since the two data sets have
224: the $I$ passband in common (F814W filter corresponds to the $I$ band), we
225: used the stars in common to photometrically homogenize the two samples.
226: %: in particular the WFI $I$ band magnitude was transformed into the ACS
227: %$m_{F814W}$ magnitude.
228: % by applyng a small shift of $\sim 0.01$ mag.
229: %We also used the ($I,V-I$) plane to define a consistent magnitude selection
230: %criterion for the two data sets.
231:
232: In order to avoid spurious effects due to incompleteness of the ground based
233: observations in the most crowded region of the cluster, we restricted the WFI
234: dataset to the outer region ($r>224\arcsec$; hereafter the {\sl WFI
235: sample}). In the inner region only stars observed with ACS have been
236: considered. However, since the ACS FoV was not centered on the cluster
237: center, the most external region ($r\sim 170-224\arcsec$) was only marginally
238: sampled by the high-resolution observations; for this reason we decided to
239: exclude it from the analysis. Hence in the following we consider in the {\sl
240: ACS sample} only stars with $r<170\arcsec$ from the cluster center.
241:
242: %In
243: %Figure~\ref{mappa} we show the position of ACS FoV together with the location of
244: %the set of WFPC2 data (dotted line) already presented in F04. Notice that the core of
245: %47Tuc is completely contained in the ACS FoV.
246:
247: \section{RESULTS}
248: %\subsection{The populations selections and radial distributions}
249: \label{sele}
250: In Figure~\ref{cmd} we show the brightest portion of the CMD obtained from
251: the analysis of the ACS sample. Thanks to the high quality of the data, all
252: the sequences are well populated and clearly separable one from the others.
253: In particular the AGB-clump is clearly visible at $I\sim 12$. This
254: feature, which flags the beginning of the AGB evolutionary phase, has been
255: observed in other clusters (Ferraro et al. 1999) and it is well predicted by
256: theoretical models (e.g., Castellani, Chieffi \& Pulone 1991). In fact the
257: beginning of the AGB phase is characterized by a slowing down of the stellar
258: evolutionary rate, just after the very fast evolution at the end of the
259: Horizontal Branch (HB) phase \citep[see Figure 1 by][]{cast91}. This is
260: nicely confirmed by the star distribution in the CMD shown in
261: Figure~\ref{cmd}, where only a few stars can be found between the HB-clump at
262: $I \sim 13.1$ and the AGB-clump. We therefore selected samples of
263: stars in different evolutionary stages as marked in Figure~\ref{cmd} by
264: different symbols: Red Giant Branch (RGB; {\it crosses}), HB\footnote{We
265: considered ``fiducial HB'' only stars belonging to the HB-clump, while those
266: lying in the HB extension toward brighter magnitudes have been considered as
267: a different population and labeled as {\it bright}-HB (hereafter bHB) stars
268: (they are marked with {\it triangles} in Figure 1), since they can be
269: sensibly contaminated by a progeny of binaries
270: currently experiencing the Helium-burning phase (see the discussion in
271: Section 4).} {\it (open circles)}, and AGB {\it (open squares)},
272: respectively.
273:
274: In {\it panel (a)} of Figure 2 we show the cumulative radial distribution of
275: the selected populations in the innermost region of the cluster
276: ($r<100\arcsec$). Surprisingly AGB stars turn out to be significantly more
277: centrally segregated than HB and RGB stars. In order to check the statistical
278: significance of this result we applied a Kolmorogov-Smirnov
279: %(KS) test, which gives the probability that the selected populations are
280: %extracted from the same parent population. The KS
281: test, that yields a $\sim99\%$ probability that the radial distribution of
282: AGB stars is genuinely different from that of both HB and RGB stars. An
283: analogous test has been performed for the WFI sample, and no significant
284: difference in the radial distribution of the three populations has been
285: detected in the outer regions of the cluster.
286:
287: %detected
288: %differences are due just to statistical fluctuations. The maximum distances in the
289: %normalized distributions ($\sim1.68$ and 1.79 of AGB from HB and RGB respectively)
290: %correspond both to $\sim99\%$ probability that they are difference.
291: %In Figure~\ref{acs_50} we show a detail of the cumulative distribution in the first 50
292: %arcsec. Using WFI sample we extracted the three populations in outer regions ($r>224$
293: %arcsec). In Figure~\ref{wfi_224} we show that, as expected, the cumulative distribution
294: %of the three populations don't shown significant differences i.e. they are ''normally``
295: %segregated.\\
296:
297: In order to further investigate this unexpected result, we followed the
298: approach proposed by F04 for the study of BSSs in this cluster: we divided
299: the surveyed area (ACS and the WFI FoVs) into a set of 9 concentric annuli.
300: Thus, HB and AGB stars has been counted in each annulus (see Table 1) and the
301: population ratio $N_{AGB}/N_{HB}$ have been computed.
302: %In particular we measured the ratio of AGB stars with respect to the HB
303: %population ($N_{AGB}/N_{HB}$)
304: %and RGB stars brighter than $m_{F814W}=12.2$
305: %($N_{AGB}/N_{RGBb}$).
306: Accordingly with the definition by \citet{fe93}, we also computed the {\it
307: double normalized} specific frequency $R_{\rm AGB}$\footnote{Note that this
308: ratio is expected to be equal to unity for any post-MS evolutionary stage,
309: since the observed number of stars is expected to scale linearly with the
310: sampled luminosity following equation 1 of \citet{rf88}.}:
311: \begin{displaymath}
312: R_{\rm AGB} = {{(N_{\rm AGB}/N_{\rm AGB}^{\rm tot})} \over
313: {(L^{\rm sampled}/L_{\rm tot}^{\rm sampled})}}
314: \end{displaymath}
315: where the fraction of AGB stars observed in each annulus ${(N_{\rm
316: AGB}/N_{\rm AGB}^{\rm tot})}$ is normalized to the fraction of luminosity
317: sampled by each annulus ${(L^{\rm sample}/L_{\rm tot}^{\rm sample})}$, as
318: computed from the surface brightness profile.
319: %, taking into account the fraction of area covered by the observations.
320: Figure~\ref{rad_acs} shows the radial distribution of the calculated ratios.
321: As can be seen, AGB stars turn out to be significantly overabundant
322: %with respect to HB stars
323: in the innermost annulus ($r<20\arcsec$, roughly corresponding to 47 Tuc core
324: radius): in fact, while in the outer regions ($r>20\arcsec$) the
325: $N_{AGB}/N_{HB}$\footnote{Note that the same result has been obtained by
326: using the RGB as reference population.} ratio is roughly constant with a mean
327: value ($0.13\pm0.03$) which is in full agreement with what expected on the
328: basis of the evolutive time scales~\citep[see][for a review]{rf88}, the same
329: ratio in the cluster core is $\sim 0.30\pm0.06$, suggesting that AGB stars
330: are twice more abundant there than in the external regions. Also the
331: $R_{AGB}$ ratio, which turns out to be $\sim 1.5$ in the central bin,
332: suggests the presence of $\sim 10$ contaminating stars (mimicking low-mass
333: AGB).
334: %, with respect to what expected on the basis of the sampled light.
335:
336: %For comparison we calculate also the ratio between HB stars and RGB stars
337: %brighter than
338: %Zero Age Horizontal Branch (ZAHB) magnitude
339: %($m_{F814W}^{ZAHB}=13.32$\footnote{We fixed the magnitude of the ZAHB at
340: %the lower envelope of the HB stellar distribution}).
341: %As described in F04, we divided the surveyed area into a set of concentric annuli.
342: %Thus, 9 concentric annuli have been defined. For each annulus we calculated population
343: %ratios. In Table~\ref{rad} we report internal and external radius of each annulus
344: %together with the detailed star counts.
345: %Figure~\ref{rad_acs} show the radial distribution
346: %of calculated ratios. We can notice that the AGB stars in the first $\sim20''$
347: %(roughly corresponding to 47tuc core radius) turns out to
348: %be more segregated than HB and RGB stars. In outer region ($r>20''$) the mean value of
349: %$N_{AGB}/N_{HB}$ is $0.13\pm0.03$ as expected by
350: %observations~\citep[see][for a review]{rf88}. In the first annulus the ratio is
351: %$0.28\pm0.06$. It seems that there is an AGB star excess of $\sim50\%$ than expected.
352: %Figure~\ref{rad_acs_wfi} shows the ratios over the full sampled area.
353:
354: \section{DISCUSSION}
355:
356: We have found evidences of a significant overabundance of AGB stars in the
357: core of 47 Tuc. This finding is in agreement with the preliminary detection
358: of an {\it "enhancement in the frequency of stars about 1 mag brighter than
359: the HB"} already noted by \citet{ba94}. As shown in Figure 3 this
360: enhancement is present only in the cluster core. Such a strong radial
361: segregation is not expected to result from dynamical processes (mass
362: segregation) acting on genuine AGB stars\footnote{AGB stars are expected to
363: be even less massive than the current TO stars because they have experienced
364: significant mass-loss during previous evolutionary stages.}. Hence the
365: combination of these observational facts (overabundance and central
366: segregation) suggests the presence along the AGB of 47 Tuc of an {\it
367: extra-population} of massive (not-genuine AGB) stars, possibly related to the
368: evolution of binary systems, i.e., generated either by mass transfer (or
369: coalescence) between binary companions, or by mergers of single/binary stars
370: due to collisions. Indeed previous multiband observations \citep{cam00, g01,
371: fe01, k02} have demonstrated that the core of 47 Tuc hosts a large population
372: of centrally segregated exotic objects (BSSs, LMXBs, CVs, MSPs, etc,), that
373: can be originated by a variety of interacting binaries. In particular, the
374: large population of BSSs, having masses between 1 and 1.6 $M_{\odot}$, are
375: the natural progenitors of any anomalous high-mass evolved object that can be
376: currently found in the cluster. In this respect, \citet{ba94} first suggested
377: that (at least part of) the excess AGB stars might be evolved BSSs currently
378: experiencing the HB evolutionary stage. However an appropriate comparison
379: with theoretical tracks (see Figure 4) clearly demonstrates that, in a metal
380: rich cluster as 47 Tuc, the Zero Age Horizontal Branch (ZAHB) magnitude level
381: of stars in the mass range $1.1-1.5 M_{\odot}$ is significantly fainter than
382: the observed AGB-clump. Instead, the CMD region populated by the lower mass
383: AGB is fully consistent with the RGB sequence of such massive stars: hence,
384: the observed AGB population can be significantly contaminated by high-mass
385: binary by-products currently ascending for the first time the RGB. Also the
386: evolutionary time-scales suggest that this could be the case: in fact, a $1.3
387: M_{\odot}$ star spends $\sim2.4\times10^7$ yr evolving along the RGB in the
388: magnitude portion covered by the low-mass AGB extension ($12.2<I<10.2$).
389: This time-scale is more than twice the AGB evolutionary time ($\sim 10^7$ yr)
390: of a genuine $0.8 M_{\odot}$ star.
391:
392: From the inspection of Figure 4 we also note that the ZAHB level of a
393: $1.1-1.5 M_{\odot}$ star is fully consistent with the brightest extension of
394: the observed HB-clump (i.e., stars labeled as bHB and plotted as empty
395: triangles in Figure 1). Hence (at least part) of the bHB stars observed
396: between the HB and the AGB clumps could be the progeny of the binary systems
397: currently experiencing the He-burning phase. Interesting enough the radial
398: distribution of these stars turns out to be significantly different (at more
399: than $3\sigma$ level) from that of genuine HB (or RGB) stars, and similar to
400: that of AGB stars (panel $b$ of Figure 2). To further investigate the
401: possibility that the canonical sequences are contaminated by binary
402: populations, we studied the radial distribution of stars at the base of the
403: RGB. In particular, we found that $\sim 30$ faint ($13.7< I <14.7$) blue
404: ($V-I \lsim 0.7$) RGB stars (fRGB) show a radial distribution which is
405: significantly different (at $3\sigma$ level) from that of the other RGB
406: within the same magnitude limits, and nicely similar to that of AGB and bHB
407: stars (panel $b$ of Figure 2). Moreover, the radial distribution of AGB, bHB
408: and fRGB stars in the innermost region of 47 Tuc nicely agrees with that of
409: BSSs \citep[][2004]{fe01}. This seems to indicate a common evolutionary
410: scenario for these four populations\footnote{Note that ratio of the presumed
411: contaminating population in the upper RGB (mimicking the AGB) and in the fRGB
412: is $\sim 3$, in reasonable agreement with the expected evolutionary
413: time-scale ratio ($\sim 4$).}, suggesting that all these objects have been
414: segregated in the cluster core by dynamical processes. As demonstrated by
415: appropriate dynamical simulations computed to model the BSS peculiar radial
416: distribution \citep[][2006]{ma04}, dynamical friction processes can generate
417: the observed central peak of the BSS distribution. Accordingly to these
418: findings, the dynamical friction time-scale here specifically computed for 47
419: Tuc (by using an appropriate model of the cluster) suggests that all the
420: objects with masses of $1.2-1.5 M_{\odot}$ generated within $\sim 10 r_c$
421: (corresponding to $\sim 4'$) from the cluster center are expected to sink
422: into the cluster core in a time comparable to the cluster age ($t=12 Gyr$);
423: hence they could generate the segregated populations detected in the core.
424:
425: The evidences reported in this Letter show the presence of an extra
426: population contaminating the genuine AGB stars in 47 Tuc: these objects are
427: significantly more centrally segregated than RGB and HB stars suggesting they
428: are significantly more massive than the normal stars in the cluster. Hence,
429: they possibly are the progeny of binary system evolution. Future
430: high-resolution spectroscopic observations aimed to search for chemical
431: anomalies (as those observed in 7 AGB stars in the outer region of the
432: cluster and in a sub-sample of BSS by Wylie et al. 2006 and Ferraro et
433: al. 2006, respectively), would identify the binary progeny and possibly
434: confirm the scenario presented here.
435:
436: %The detection of anomalies in the radial distribution and relative abundances
437: %of stars (seemingly) belonging to cluster canonical evolutionary sequences,
438: %as well as the detection peculiar radial distributions and chemical anomalies
439: %in exotic objects like the BSSs (ref...), therefore represent the most
440: %powerful tools to unveil the impact of dynamical processes on stellar
441: %evolution, particularly if coupled with accurate dynamical simulations
442: %(ref...).
443:
444:
445:
446: \acknowledgements{} We acknowledge the anonymous referee for useful comments.
447: This research was supported by contract ASI-INAF I/023/05/0, PRIN/INAF 2006,
448: and the Ministero dell'Istruzione, dell'Universit\`a e della Ricerca.
449:
450:
451: \begin{thebibliography}{}
452:
453: \bibitem[Bailyn (1994)]{ba94} Bailyn, C.~D.\ 1994, \aj, 107, 1073
454:
455: \bibitem[Bailyn \& Pinsonneault (1995)]{ba95} Bailyn, C.~D., \& Pinsonneault,
456: M.~H.\ 1995, \apj, 439, 705
457:
458: \bibitem[Bedin et al.(2005)]{bed05} Bedin, L.~R., Cassisi, S., Castelli, F.,
459: Piotto, G., Anderson, J., Salaris, M., Momany, Y., \& Pietrinferni, A.\ 2005,
460: \mnras, 357, 1038
461:
462: %\bibitem[Bellazzini et al.(2002)]{be02} Bellazzini, M., Fusi
463: %Pecci, F., Messineo, M., Monaco, L., \& Rood, R.~T.\ 2002, \aj, 123, 1509
464:
465: \bibitem[Bertin \& Arnouts (1996)]{be96} Bertin, E., \&
466: Arnouts, S.\ 1996, \aaps, 117, 393
467:
468: \bibitem[Camilo et al. (2000)]{cam00} Camilo, F., Lorimer, D.~R., Freire, P.,
469: Lyne, A.~G., \& Manchester, R.~N.\ 2000, \apj, 535, 975
470:
471: \bibitem[Castellani et al. (1991)]{cast91}Castellani, V.,
472: Chieffi, A., \& Pulone, L.\ 1991, \apjs, 76, 911
473:
474: %\bibitem[Cool \& Bolton (2002)]{co02}Cool A. M., Bolton A. S., 2002, in ASP Conf.
475: %Ser. 263, Stellar Collisions, Mergers and their Consequences. Astron. Soc. Pac.,
476: %San Francisco, p. 163
477:
478: %\bibitem[Ferraro et al.(1992)]{fe92} Ferraro, F.~R., Clementini, G., Fusi
479: %Pecci, F., Sortino, R., Buonanno, R.\ 1992, \mnras 256, 391
480:
481: \bibitem[Ferraro et al.(1993)]{fe93} Ferraro, F.~R., Fusi Pecci, F.,
482: Cacciari, C., Corsi, C., Buonanno, R., Fahlman, G.~G., \& Richer, H.~B.\
483: 1993, \aj, 106, 2324
484:
485: \bibitem[Ferraro et al.(1999)]{fe99} Ferraro F. R., Messineo M., Fusi Pecci F.,
486: De Palo M. A., Straniero O., Chieffi A., Limongi M.\ 1999, \aj, 118,
487: 1738
488:
489: \bibitem[Ferraro et al.(2001)]{fe01} Ferraro, F.~R., D'Amico, N., Possenti,
490: A., Mignani, R.~P., \& Paltrinieri, B.\ 2001, \apj, 561, 337
491:
492: \bibitem[Ferraro et al.(2004)]{fe04} Ferraro, F.~R.,
493: Beccari, G., Rood, R.~T., Bellazzini, M., Sills, A., \& Sabbi, E.\ 2004,
494: \apj, 603, 127 (F04)
495:
496: \bibitem[Ferraro et al.(2006)]{fe06} Ferraro, F.~R., et al.\ 2006, \apjl,
497: 647, 53
498: %Sabbi, E., Gratton, R., Piotto, G., Lanzoni, B., Carretta, E., Rood, R.~T.,
499: %Sills, A., Fusi Pecci, F., Moehler, S., \& 3 coauthors
500:
501: \bibitem[Guhathakurta et al.(1992)]{gu92} Guhathakurta, P., Yanny, B,,
502: Schneider, D.~P., \& Bahcall, J.~N.\ 1992, \aj, 104, 1790
503:
504: \bibitem[Grindlay et al.(2001)]{g01} Grindlay, J.~E.,
505: Heinke, C., Edmonds, P.~D., \& Murray, S.~S.\ 2001, Science, 292, 2290
506:
507: %\bibitem[Ivanova et al.(2005)]{iv05} Ivanova, N.,
508: %Belczynski, K., Fregeau, J.~M., \& Rasio, F.~A.\ 2005, \mnras, 358, 572
509:
510: \bibitem[Knigge et al.(2002)]{k02} Knigge, C., Zurek,
511: D.~R., Shara, M.~M., \& Long, K.~S.\ 2002, \apj, 579, 752
512:
513: \bibitem[Mapelli et al.(2004)]{ma04} Mapelli, M.,
514: Sigurdsson, S., Colpi, M., Ferraro, F.~R., Possenti, A., Rood, R.~T.,
515: Sills, A., \& Beccari, G.\ 2004, \apjl, 605, L29
516:
517: \bibitem[Mapelli et al.(2006)]{ma06} Mapelli, M.,
518: Sigurdsson, S., Ferraro, F.~R., Colpi, M., Possenti, A., \& Lanzoni, B.\
519: 2006, \mnras in press (astro-ph/0609220)
520:
521: \bibitem[Paresce et al.(1991)]{par91} Paresce, F., Meylan, G., Shara, M.,
522: Baxter, D., \& Greenfield, P.\ 1991, Nature, 352, 297
523:
524: \bibitem[Pietrinferni et al.(2006)]{p06} Pietrinferni, A.,
525: Cassisi, S., Salaris, M., \& Castelli, F.\ 2006, \apj, 642, 797
526:
527: %\bibitem[Pulone (1992)]{pu92} Pulone, L., 1992, MmSAI 63, 53
528:
529: \bibitem[Renzini \& Fusi Pecci(1988)]{rf88} Renzini, A., \& Fusi Pecci, F.\
530: 1988, ARA\&A, 26, 199
531:
532: %\bibitem[Rubenstein \& Bailyn(1997)]{ru97} Rubenstein,
533: %E.~P., \& Bailyn, C.~D.\ 1997, \apj, 474, 701
534:
535: \bibitem[Schechter, Mateo \& Saha (1993)]{dophot} Schechter, P. L., Mateo, M.,
536: \& Saha, A.\ 1993, PASP, 105, 1342
537:
538: \bibitem[Sirianni et al.(2005)]{si05} Sirianni, M., et al.\
539: 2005, \pasp, 117, 1049
540:
541: \bibitem[Wylie et al.(2006)]{w06} Wylie, E.~C., Cottrell,
542: P.~L., Sneden, C.~A., \& Lattanzio, J.~C.\ 2006, astro-ph/0605538
543:
544: \end{thebibliography}
545:
546: \newpage
547: \begin{deluxetable}{rrr}
548: \scriptsize \tablewidth{6.5cm}
549: \tablecaption{\label{rad} Number of AGB and HB stars in concentric annuli at
550: different distances from the cluster center.
551: }
552: \startdata \\
553: \hline
554: \hline
555: & & \\
556: $r[arcsec]$ & $N_{AGB}$ & $N_{HB}$ \\
557: % (arcsec) & & \\
558: & & \\
559: \hline
560: & & \\
561: 0--21 & 30 & 101 \\
562: 21--58 & 23 & 195 \\
563: 58--110 & 15 & 104 \\
564: 110--224 & 6 & 45 \\
565: 224--350 & 19 & 145 \\
566: 350--550 & 18 & 152 \\
567: 550--730 & 8 & 61 \\
568: 730--1000 & 5 & 30 \\
569: \hline
570: \enddata
571: %\tablecomments{}
572: \end{deluxetable}
573:
574: \begin{figure}
575: \plotone{f1.ps}
576: \caption{\label{cmd} \footnotesize{CMD for the ACS sample. Stars marked with
577: different symbols belong to different populations }}
578: \end{figure}
579:
580: \begin{figure}
581: \plotone{f2.ps}
582: \caption{\label{acs_50} \footnotesize{ The cumulative radial distribution of
583: AGB stars ({\it heavy solid line}) in the innermost region of the cluster
584: ($r< 100 \arcsec$) as a function of the projected distance r from the cluster
585: center, is compared with those obtained for HB and RGB stars ({\it panel a}),
586: and bHB, faint-blue RGB (fRGB), and BSS stars ({\it panel b}),
587: respectively. AGB stars are significantly more centrally segregated than RGB
588: and HB stars. The radial distribution of AGB, bHB and fRGB stars is nicely in
589: agreement with that of BSS stars. }}
590: \end{figure}
591:
592: \begin{figure}
593: \plotone{f3.ps}
594: \caption{\label{rad_acs} \footnotesize{Relative frequency of AGB-to-HB stars
595: {\it (upper panel)} and double-normalized specific frequency of the AGB stars
596: {\it (lower panel)} as a function of the projected distance from the center,
597: over the entire cluster extention. Horizontal grey regions in the {\it lower
598: panel} show the double-normalized specific frequency of HB stars used as
599: reference population. The dashed area marks the cluster region excluded form
600: the analysis in order to avoid spurious effects due to inclompleteness of the
601: sample. }}
602: \end{figure}
603:
604: \begin{figure}
605: \epsscale{1.1}
606: \plotone{f4.ps}
607: \caption{\label{track} \footnotesize{Theoretical tracks from
608: \citet[][http://www.te.astro.it/BASTI/index.php]{p06} for different masses
609: (from 1.1 to $1.5 M_{\odot}$; solid lines) are overplotted to the ACS CMD of
610: 47 Tuc. The theoretical tracks, with $\alpha-enhancement$ ([$\alpha$/Fe]= 0.4)
611: and [Fe/H]=-0.66 (Z= 0.008), have been transformed in the ACS filters
612: following \citet{bed05}. A distance modulus (m-M$)_0=13.33$ and reddening
613: E(B-V)=0.04, have been adopted \citep{fe99}. The RGB phase for such massive
614: stars turns out to occupy the same region of the CMD where "normal" low-mass
615: AGB stars are observed. The post-RGB evolutionary track for a $1.5
616: M_{\odot}$ star is also shown (dashed line). Note that the ZAHB level for
617: such a massive star lies in the bright extension of the observed HB where the
618: bHB sample has been defined. The evolutionary time needed by a $1.3
619: M_{\odot}$ star to cover the RGB portion between the two horizontal lines is
620: also reported. }}
621: \end{figure}
622:
623: % \begin{figure}
624: % \plotone{f4.ps}
625: % \caption{\label{track} \footnotesize{ Theoretical tracks from \citet{p06} for
626: % different masses (from 1.1 to $1.5 M_{\odot}$; solid lines) are overplotted
627: % to the ACS CMD of 47 Tuc. The RGB phase for such massive stars turns out to
628: % occupy the same region of the CMD where "normal" low-mass AGB stars are
629: % observed. The post-RGB evolutionary track for a $1.5 M_{\odot}$ star is also
630: % shown (dashed line). Note that the ZAHB level for such a massive star lies
631: % in the bright extention of the observed HB where the bHB sample has been
632: % defined. The evolutionary time needed by a $1.3 M_{\odot}$ star to cover the
633: % RGB portion between the two horizontal lines is also reported. }}
634: % \end{figure}
635:
636: \end{document}
637: