\omega Centauri Abundances

Fe and Al Abundances for 180 Red Giants in the Globular Cluster Omega Centauri (NGC 5139)

Christian I. Johnson11affiliation: Department of Astronomy, Indiana University, Swain West 319, 727 East Third Street, Bloomington, IN 47405–7105, USA; cijohnson@astro.indiana.edu; catyp@astro.indiana.edu , Catherine A. Pilachowski11affiliation: Department of Astronomy, Indiana University, Swain West 319, 727 East Third Street, Bloomington, IN 47405–7105, USA; cijohnson@astro.indiana.edu; catyp@astro.indiana.edu , Jennifer Simmerer22affiliation: Lund Observatory, Box 43, SE 221-00 Lund, Sweden; jennifer@astro.lu.se , and Dustin Schwenk33affiliation: Department of Physics, University of Illinois at Urbana–Champaign, 2910 Artesia Crossing, Urbana, IL 61802, USA; schwenk@uiuc.edu
Abstract

We present radial velocities, Fe, and Al abundances for 180 red giant branch (RGB) stars in the Galactic globular cluster Omega Centauri ( Cen). The majority of our data lie in the range 11.0V13.5, which covers the RGB from about 1 mag. above the horizontal branch to the RGB tip. The selection procedures are biased towards preferentially observing the more metal–poor and luminous stars of Cen. Abundances were determined using equivalent width measurements and spectrum synthesis analyses of moderate resolution spectra (R13,000) obtained with the Blanco 4m telescope and Hydra multifiber spectrograph. Our results are in agreement with previous studies as we find at least four different metallicity populations with [Fe/H]=–1.75, –1.45, –1.05, and –0.75, with a full range of –2.20[Fe/H]–0.70. [Al/Fe] ratios exhibit large star–to–star scatter for all populations, with the more than 1.0 dex range of [Al/Fe] decreasing for stars more metal–rich than [Fe/H]–1.4. The minimum [Al/Fe] abundance observed for all metallicity populations is [Al/Fe]+0.15. The maximum abundance of log (Al) is reached for stars with [Fe/H]–1.4 and does not increase further with stellar metallicity. We interpret these results as evidence for type II SNe providing the minimum [Al/Fe] ratio and a mass spectrum of intermediate mass asymptotic giant branch stars causing the majority of the [Al/Fe] scatter. These results seem to fit in the adopted scheme that star formation occurred in Cen over 1 Gyr.

stars: abundances, globular clusters: general, globular clusters: individual ( Centauri, NGC 5139). stars: Population II

1 Introduction

The Galactic globular cluster Omega Centauri ( Cen) presents a unique opportunity to study the chemical evolution of both a small stellar system and stars with common formation histories covering a metallicity range of more than a factor of 10, a defining characteristic of Cen that has been known since the initial discovery of its unusually broad red giant branch (RGB) by Woolley (1966). Although Cen is the most massive Galactic globular cluster, with an estimated mass of 2–710 M (Richer et al. 1991; Meylan et al. 1995; van de Ven et al. 2006), it does not appear to have an exceptionally deep gravitational potential well (Gnedin et al. 2002). This seems to negate a simple explanation that Cen evolved as a typical globular cluster that was more easily able to retain supernova (SN) and asymptotic giant branch (AGB) ejecta for self–enrichment. This fact coupled with the cluster’s retrograde orbit and disk crossing time of 1–210 years (e.g., Dinescu et al. 1999), which could severely inhibit star formation, are some of the strongest arguments against Cen having a Galactic origin. Instead, it has been proposed (e.g., Dinescu et al. 1999; Smith et al. 2000; Gnedin et al. 2002; Bekki & Norris 2006) that Cen may be the remaining nucleus of a dwarf spheroidal galaxy that evolved in isolation and was later accreted by the Milky Way, suggesting the progenitor system was perhaps a factor of 100–1000 times more massive than what is presently observed.

Recent spectroscopic and photometric studies (Norris & Da Costa 1995; Norris et al. 1996; Suntzeff & Kraft 1996; Lee et al. 1999; Hilker & Richtler 2000; Hughes & Wallerstein 2000; Pancino et al. 2000; Smith et al. 2000; van Leeuwen et al. 2000; Rey et al. 2004; Stanford et al. 2004; Piotto et al. 2005; Sollima et al. 2005a; Sollima et al. 2005b; Kayser et al. 2006; Sollima et al. 2006; Stanford et al. 2006; Stanford et al. 2007; van Loon et al. 2007; Villanova et al. 2007) have confirmed the existence of up to five separate stellar populations ranging in metallicity from [Fe/H]–2.2 to –0.5, with a peak in the metallicity distribution near [Fe/H]–1.7 and a long tail extending to higher metallicities. In addition to the metal–poor and intermediate metallicity populations initially seen in the Woolley (1966) photometric study, Lee et al. (1999) and Pancino et al. (2000) discovered the existence of the most metal–rich RGB at [Fe/H]–0.5, commonly referred to as the anomalous RGB (RGB–a). The RGB–a is primarily observed in the central region of the cluster and contains approximately 5 of the total stellar population (Pancino et al. 2000), in contrast to the dominant metal–poor population that contains roughly 75 of cluster stars. Additionally, there is some evidence (Norris et al. 1997) that the metal–rich population exhibits smaller radial velocity dispersion and rotation than the metal–poor population. Sollima et al. (2005b) confirmed the Norris et al. (1997) results but also showed that the most metal–rich stars ([Fe/H]–1) exhibit an increasing velocity dispersion as a function of increasing metallicity, which could be evidence for accretion events occurring within Cen’s progenitor system (Ferraro et al. 2002; Pancino et al. 2003); however, this result is not yet confirmed (Platais et al. 2003, but see also Hughes et al. 2004). It should be noted that Pancino et al. (2007), using radial velocity measurements of 650 members with measurement uncertainties of order 0.5 km s, have found no evidence for rotational differences among the different metallicity groups.

The distribution of main–sequence turnoff (MSTO) and subgiant branch (SGB) stars matches that observed on the RGB, such that one can trace the evolutionary sequence of each population from at least the MSTO to the RGB using high precision photometry (e.g., Villanova et al. 2007). The main–sequence (MS) has proved equally as complex as the SGB and RGB, with the discovery by Anderson (1997) of a red and blue MS (BMS). Interestingly, Piotto et al. (2005) discovered that the BMS was more metal–rich than the red MS, suggesting the BMS could be explained assuming a higher He content, perhaps as high as Y0.38 (Bedin et al. 2004; Norris 2004; Lee et al. 2005; Piotto et al. 2005).

While it is clear that multiple populations are present in this cluster, there has been some debate regarding the age of each population. There is general agreement that the age range is between about 0 and 6 Gyrs (Norris & Da Costa 1995; Hilker & Richtler 2000; Hughes & Wallerstein 2000; Pancino et al. 2002; Origlia et al. 2003; Ferraro et al. 2004; Hilker et al. 2004; Rey et al. 2004; Sollima et al. 2005a; Sollima et al. 2005b; Villanova et al. 2007), though the recent work by Stanford et al. (2006) suggests the most likely age range is 2–4 Gyrs, with the metal–rich stars being younger. For the case of monotonic chemical enrichment in a single system, one would expect the more metal–rich stars to be younger than the more metal–poor; however, this assumption has been questioned by Villanova et al. (2007) who suggested the metal–rich stars and 33 of the metal–poor stars are the oldest with the remaining 2/3 of the metal–poor population being 3–4 Gyrs younger. The picture of Cen’s formation is further compounded by observations of RR Lyrae horizontal branch (HB) stars that reveal a bimodal metallicity distribution without a trend in He enhancement as a function of [Fe/H] (Sollima et al. 2006). The important point here is that a group of RR Lyrae stars exists with the same metallicity as the BMS but without the presumed He enhancement. A He–rich secondary population would not produce a significant RR Lyrae population unless a 4 Gyr age difference was present with respect to the dominant metal–poor population (Sollima et al. 2006). The required age difference is therefore inconsistent with most age spread estimates that put 4 Gyrs.

Cen’s chemical evolution history has so far proved difficult to interpret from measured abundances of light (Z27), , Fe–peak, s–process, and r–process elements. In “normal” Galactic globular clusters, C, N, O, F, Na, Mg (sometimes), and Al often exhibit large star–to–star variations, in some cases exceeding more than a factor of 10 (e.g., see recent review by Gratton et al. 2004). In contrast, the heavier –elements (e.g., Ca and Ti) show little to no variation and are enhanced relative to Fe at [/Fe]+0.30, with a decreasing ratio for clusters with [Fe/H]–1. Likewise, Fe and all other Fe–peak, s–process, and r–process elements show star–to–star variations of 0.10–0.30 dex. Additionally, nearly all globular clusters are enriched in r–process relative to s–process elements by about 0.20 dex. In Cen, [Fe/H] covers a range of more than 1.5 dex and, as previously stated, it has a potential well comparable to that of other globular clusters, suggesting it had to be different in the past to undergo self–enrichment. The scenario of two or more globular clusters merging seems unlikely now given the results of Pancino et al. (2007) and the typically large orbital velocities coupled with the small velocity dispersions of clusters (Ikuta & Arimoto 2000). While Cen exhibits large abundance variations for several of the light elements at various metallicities (e.g., Norris & Da Costa 1995; Smith et al. 2000), the mean heavy –element enhancement is surprisingly uniform at [/Fe]+0.30 to +0.50 (Norris & Da Costa 1995; Smith et al. 2000; Villanova et al. 2007), with perhaps a trend of decreasing [/Fe] at [Fe/H]–1 (Pancino et al. 2002). The s–process elements show a clear increase in abundance relative to Fe with a plateau occurring at [Fe/H]–1.40 to –1.20 (Norris & Da Costa 1995; Smith et al. 2000). However, unlike in globular clusters, s–process elements are overabundant with respect to r–process elements, where [Ba/Eu] typically reaches between 0.5 and 1.0 (Smith et al. 2000), indicating a strong presence of AGB ejecta.

Many globular cluster giants show clear C–N, O–Na, O–Al, Mg–Al, and in the case of M4 (Smith et al. 2005), F–Na anticorrelations alongside a Na–Al correlation (e.g., Gratton et al. 2004). In addition to these anomalies being present in the atmospheres of RGB stars, similar relations have been observed in some globular cluster MS and MSTO stars (e.g., Cannon et al. 1998; Gratton et al. 2001; Cohen et al. 2002; Briley et al. 2004a; 2004b; Boesgaard et al. 2005). According to standard evolutionary theory, first dredgeup brings the products of MS core hydrogen burning to the surface and homogenizes approximately 70–80 of the star, resulting in C depletion, N enhancement, and a lowering of the C/C ratio from about 90 to 25 (e.g., Salaris et al. 2002). The decline in [C/Fe] and C/C has been verified via observations in both globular cluster (Bell et al. 1979; Carbon et al. 1982; Langer et al. 1986; Bellman et al. 2001) and field stars (Charbonnel & do Nascimento 1998; Gratton et al. 2000; Keller et al. 2001) as strong evidence for in situ mixing occurring along the RGB. However, as the advancing hydrogen–burning shell (HBS) crosses the molecular weight discontinuity left by the convective envelope’s deepest point of penetration, extra mixing not predicted by canonical theory occurs in both field and cluster stars, driving down [C/Fe] further and allowing C/C to reach the CN–cycle equilibrium value of 4. The mechanism responsible for this extra mixing is not known, though thermohaline mixing (Charbonnel & Zahn 2007) may ameliorate the problem. While halo field and cluster giants share these same trends, differences arise when considering O, Na, and Al abundances. Field stars do not exhibit most of the familiar correlations/anticorrelations and large star–to–star variations seen in globular cluster stars and instead remain mostly constant from the MS to the RGB tip (e.g., Ryan et al. 1996; Fulbright 2000; Gratton et al. 2000).

The reason for the observed differences between cluster and field giants is not known, but obviously the higher density cluster environment is a key factor. Coupled O depletions and Na/Al enhancements are clear signs of high temperature (T4010 K) H–burning via the ON, NeNa, and MgAl proton–capture cycles, but this does not necessarily mean those cycles are operating in the RGB stars we presently observe and instead may be from the ejecta of intermediate mass (IM) AGB stars (3–8 M) that underwent hot bottom burning (HBB) and polluted the gas from which the current stars formed. One of the strongest arguments against in situ mixing is the observed abundance relations on the MS and MSTO matching those on the RGB because these stars are both too cool for the ON, NeNa, and MgAl cycles to operate and their shallow envelope convection zones do not reach deep enough to bring up even CN–cycled material. Additionally, Shetrone (1996) showed that at least in M13 giants, Mg is anticorrelated with Al instead of Mg and/or Mg, which means temperatures not achievable in low mass RGB stars (at least 7010 K) are needed to activate the full MgAl chain (Langer et al. 1997); however, these temperatures are reached in HBB conditions. Current models of low mass RGB stars (e.g., Denissenkov & Weiss 2001) indicate Al is only produced deep in the stellar interior by burning Mg and convective mixing reaching these depths would cause a second increase in the surface abundance of both Na and He. It should be noted that if it is instead Al (110 yrs) causing the abundance anomalies on the upper RGB, then the O–Na and Na–Al relations can be explained in a self–consistent manner via in situ mixing (Denissenkov & Weiss 2001). Also, there is some evidence that O depletions and Na/Al enhancements become stronger in the upper 0.7 mag before the RGB tip in M13 (e.g., Sneden et al. 2004; Johnson et al. 2005), indicating the possible operation of additional deep mixing episodes in some stars. Although it is more difficult to believe in situ mixing is responsible for the Mg–Al anticorrelation, the same may not be true for O and Na. In or just above the HBS of a metal–poor low mass RGB star, the O–Na anticorrelation can be naturally explained because the ON and NeNa cycles can operate at T4010 K (Denisenkov & Denisenkova 1990; Langer et al. 1993). Of course, this cannot be the case for any O–Na anticorrelation observed in MSTO and SGB stars and does require convective mixing in RGB stars to penetrate past the radiative zone separating the bottom of the convective envelope and the top of the HBS.

While pollution from a previous generation of more massive AGB stars seems an attractive explanation, there are a few important issues. Predicted IM–AGB stellar yields are sensitive to the adopted treatment of convection because it affects other important parameters such as luminosity, number of thermal pulses, third dredgeup efficiency, envelope temperature structure, and mass loss (Ventura & D’Antona 2005a). The two most common methods employed are mixing length theory (MLT) (e.g., Fenner et al. 2004) and the full spectrum of turbulence (FST) model (e.g., Ventura & D’Antona 2005b), with the latter providing more efficient convection. In Cen and all other globular clusters observed, the [C+N+O/Fe] sum is constant (Pilachowski et al. 1988; Dickens et al. 1991; Norris & Da Costa 1995; Smith et al. 1996; Ivans et al. 1999), but models based on MLT indicate stars forming from different generations of AGB ejecta should show a large increase in the CNO sum (e.g., Lattanzio et al. 2004). In contrast, FST models keep [C+N+O/Fe] constant to within about a factor of 2 due to enhanced mass loss and fewer third dredgeup episodes (Ventura & D’Antona 2005b). Although Na and Al production could be due to HBB, it is difficult to produce the observed O depletion of 1.0 to 1.5 dex along with the required Na enhancement (e.g., Denissenkov & Herwig 2003; but see also Ventura & D’Antona 2005b). Self–consistent models of globular cluster enrichment from AGB ejecta fail to reproduce the MgAl anticorrelation seen in several globular clusters, including Cen, where Mg increases relative to Al instead of decreases (Fenner et al. 2004). Without an evolutionary scenario, O deficient, Na/Al enhanced stars must have preferentially formed out of enriched gas relative to “O–normal” stars (i.e., [O/Fe]+0.30) and Yong et al. (2003) point out that even with no O present in the enriched gas, these stars would require a composition of 90 enriched, 10 “normal” material to obtain the observed O deficiency. Lastly, AGB stellar envelopes contain roughly 36 He by mass (Lattanzio et al. 2004), but O–poor, Na/Al–rich stars do not appear to be particularly He–rich; however, this does not rule out AGB stars as the source of the He–rich BMS observed in Cen. Given the evidence for and against evolutionary and primordial processes, a hybrid scenario probably needs to be invoked to explain all abundance anomalies.

Given the inherently large spread in metallicity of stars in Cen and that Al is the heaviest element sensitive to proton–capture nucleosynthesis at temperatures achieved in the interiors of low mass metal–poor RGB stars, we present radial velocities, Fe, and Al abundances for 180 RGB stars covering –2.20[Fe/H]–0.70. With additional data from the literature covering from the MS to the RGB tip, we address the issues of star formation and possible pollution sources driving the chemical evolution of Cen as a function of metallicity.

2 Observations and Reductions

The observations of all 180 giants in Cen were obtained with the Blanco 4m telescope using the Hydra multifiber positioner and bench spectrograph at the Cerro Tololo Inter–American Observatory. All observations were obtained using the “large” 300 (2) fibers. The full spectral coverage ranged from 6450–6750 Å, centered on 6600 Å; however, wavelengths blueward of 6500 Å lie on the shoulder of the filter response curve, making continuum placement difficult. Therefore, we truncated the spectra to include only the region from 6500–6750 Å. The 316 line mm echelle grating and Blue Air Schmidt Camera provided a resolving power of R(/)13,000 (0.5 Å FWHM) at 6600 Å. A list of our observation dates and exposure times is provided in Table 1.

Target stars, coordinates, photometry, and membership probability were taken from the proper motion study by van Leeuwen et al. (2000). Stars were given priority in the Hydra assignment program based on V magnitude, with a focus on stars in the range 11.0V14.0, which includes all giants in the cluster brighter than the HB up to the RGB tip. Only stars with membership probabilities 80 were included for possible study. All observations took place between 2003 July 17 and 2003 July 19. Three different Hydra setups were used with exposure times ranging from 1800 to 3600 seconds. Each setup allowed approximately 100 fibers to be placed on targets, yielding a total initial sample size of nearly 300 stars. At V13.5, reaching a signal–to–noise (S/N) ratio of 100 requires 3 hours of total integration time. Unfortunately, weather and time constraints led to one of the setups receiving less than 2 hours of integration time with an average S/N of less than 50. Many of these stars had to be excluded from analysis due to poor S/N; however, the final sample size still includes nearly 200 stars. These are shown in Figure 1 along with the complete sample given in van Leeuwen et al. (2000) for 11.0V14.0.

Due to Cen’s broad RGB, selection effects must be taken into account when interpreting abundance results. Figure 2 shows our observed completion fraction of RGB stars both as a function of V magnitude and B–V color compared to the deeper photometric study by Rey et al. (2004). Since our observing program is biased towards selecting brighter stars, our sample includes more metal–poor than metal–rich stars because metal–rich stars have lower V magnitudes due to H opacity increasing with increasing metallicity. While we observed 75 of all RGB tip stars available, the fraction of stars observed decreases to 15–50 in the range 11.5V13.0. Likewise, in considering completeness in B–V color, our sample includes stars of higher luminosity for a given B–V, biasing our results towards the more metal–poor regime.

Figure 3 shows the location of our observed stars in right ascension and declination relative to the cluster center, defined by van Leeuwen et al. (2000) as 132645.9, –472837.0 (J2000) and marked with a cross in the figure. Since some evidence exists for a correlation between metallicity and distance from the cluster center (Norris et al. 1996; Suntzeff & Kraft 1996; Norris et al. 1997; Hilker & Richtler 2000; Pancino et al. 2000; Rey et al. 2004), we have observed stars as uniformly as possible at radii extending out to 20. Near the cluster center, crowding and the physical size of the fibers limited the number of observations inside about 2 core radii, where the core radius is approximately 1.40 (Harris 1996; rev. 2003 February). We illustrate this effect with the ellipses in Figure 3 that correspond to 1, 5, and 10 core radii.

Basic data reductions were accomplished using the IRAF111IRAF is distributed by the National Optical Astronomy Observatories, which are operated by the Association of Universities for Research in Astronomy, Inc., under cooperative agreement with the National Science Foundation. package ccdproc to trim the bias overscan region and apply bias level corrections. The IRAF task dohydra was employed to correct for scattered light, extract the one–dimensional spectra, remove cosmic rays, apply a flat–field correction, linearize the wavelength scale, and subtract the sky spectrum. Wavelength calibrations were carried out using a high S/N solar sky spectrum because the ThAr lamp was unavailable. Standard IRAF tasks were used to co–add and normalize the spectra. Typical S/N ratios for individual exposures ranged from 25–50, with co–added spectra having S/N between 75 and 150.

3 Radial Velocity Determinations

Cen’s location in the thick disk (Dinescu et al. 1999) makes field star contamination a more serious problem than for typical halo globular clusters. While we initially only chose targets with high membership probabilities from van Leeuwen et al. (2000), direct measurements of target radial velocities assist with membership confirmation. Radial velocities were determined using the IRAF tasks rvcor, to correct for heliocentric motion, and fxcor, to determine the heliocentric radial velocity. For the comparison spectrum, we used the same high S/N daylight sky spectrum that was used for wavelength calibration. A summary of our determined radial velocities along with membership probabilities from van Leeuwen et al. (2000) are given in Table 2.

The largest radial velocity study of Cen stars to date is by Reijns et al. (2006), who determined radial velocities for 2,000 RGB stars. Their study finds a strongly peaked distribution near 232 km s, with a median uncertainty of less than 2 km s and a velocity dispersion of 6 km s for the inner 25 of the cluster. Similarly, Mayor et al. (1997) find V=232.8 0.7 km s (17.5 km s) for 471 stars and Suntzeff & Kraft (1996) find V=234.7 1.3 km s (=11.3 km s) for their “bright” sample of 199 stars. Recently, Pancino et al. (2007) determined radial velocities for 650 RGB stars and found V=233.4 0.5 km s (=13.2 km s). We find in agreement with these studies: V=231.8 km s 1.6 km s (=11.6 km s). Our observations do not provide an absolute velocity calibration, but comparison with the other observations of the average velocity of cluster stars suggests that the systematic error of our radial velocities is less than about 2 km s. Since all of our stars listed in Table 2 are less than 3 away from the cluster averaged velocity and Cen’s velocity is high relative to the general field population, it is unlikely any of our targets are field stars.

4 Analysis

We have derived Fe and Al abundances using lines available in the spectral range 6500–6750 Å with either equivalent width or synthetic spectrum analyses. Spectrum synthesis was used to determine Al abundances in metal–rich and/or CN–strong stars. When multiple lines were available, the stated abundances represent the average of the individual lines. Effective temperatures (T) and gravities (log g) were estimated using published (V–K) photometry. T and microturbulence (V) were further refined via spectroscopic analyses. The analysis follows the methods described in Johnson et al. (2005) and Johnson & Pilachowski (2006).

4.1 Model Stellar Atmospheres

Using V photometry from van Leeuwen et al. (2000) and K photometry from 2MASS, we estimated T with the color–temperature relation described in Alonso et al. (1999; 2001), which is based on the infrared flux method (Blackwell & Shallis 1977). However, the Alonso et al. (1999) method requires the photometry to be on the Carlos Sánchez Telescope (TCS) photometric system. We transformed the V and K magnitudes onto the TCS system using the transformations provided in Alonso et al. (1994; 1998) and Carpenter (2001), as summarized in Johnson et al. (2005). To correct for interstellar reddening and extinction, we applied the correction recommended by Harris (1996; rev. 2003 February) of E(B–V)=0.12 and used E(V–K)/E(B–V)=2.7 (Johnson 1965). While Calamida et al. (2005) claim differential reddening, perhaps differing by as much as a factor of two near the core, could be a problem, the well defined evolutionary sequences seen in Villanova et al. (2007) seem to indicate it is not too severe. Van Loon et al. (2007) find some evidence for interstellar absorption by gas in the cluster, but this is concentrated near the core where our observations are sparse. Therefore, we have only applied a uniform reddening correction. Bolometric corrections were applied using the empirical relations given in Alonso et al. (1999) assuming a distance modulus of (m–M)=13.7 (van de Ven et al. 2006).

Applying the proper color–temperature relation requires knowledge of a star’s metallicity. Therefore, we took the empirical relation given in van Leeuwen et al. (2000; their eq. 15), which gives [Ca/H] as a function of V and B–V, and assumed [Ca/Fe]+0.30 for [Fe/H]–1.0 (e.g., Norris & Da Costa 1995), with a linear decrease towards [Ca/Fe]=0.0 at [Fe/H]=0.0. This gave a rough estimate of [Fe/H] for each star and allowed us to choose the proper equation in Alonso et al. (1999).

Since only one Fe II line was available for analysis (6516 Å), we determined surface gravity using the standard relation,

(1)

instead of the ionization equilibrium of Fe. We assumed M=0.80 M for all stars, regardless of metallicity. Though there may be an intrinsic age spread of a few Gyr on the RGB (see 5 for further discussion on this issue), this will lead to a mass difference only of order 0.05 M, which is negligible for surface gravity determinations.

In addition to T, log g, and [Fe/H] estimates, we also needed a starting point with V. Initial estimates were based on the empirical relation derived in Pilachowski et al. (1996), which gives V as a function of T for metal–poor field giants and subgiants. Typical V values ranged from about 1.3–2.3 km s in the temperature range 5000–3800 K, respectively.

We generated the model stellar atmospheres by interpolating in the ATLAS9222The model atmosphere grids can be downloaded from http://cfaku5.cfa.harvard.edu/grids.html. (Castelli et al. 1997) grid of models without convective overshoot. Initial models were created using the T, log g, [Fe/H], and V estimates as described above. T was further refined by removing trends in Fe abundance as a function of excitation potential. Likewise, V was improved by removing trends in Fe abundance as a function of reduced width (log(EW/)). A comparison between photometric and spectroscopically determined temperatures is given in the top panel of Figure 4. Typical photometric and spectroscopic temperature estimates agree to within approximately 100 K. The bottom panel of Figure 4 shows our spectroscopically determined V as a function of T for different metallicity bins with a linear least squares fit given by,

(2)

which is independent of metallicity. This fit agrees to within 0.10–0.15 km s to that given in Pilachowski et al. (1996). Figure 5 shows our derived [Fe II/H] given as a function of [Fe/H]. As stated above, we only had one Fe II line available for analysis, but the fact that both Fe estimates agree to within 0.16 dex on average (=0.12 dex) leads us to believe our surface gravity estimates are not in serious error. A complete list of our adopted model atmosphere parameters is provided in Table 3.

4.2 Derivation of Abundances

Abundances were determined using equivalent width analyses for all Fe lines and most Al lines, with the exception of cases where evidence for considerable CN contamination near the 6696, 6698 Å Al doublet (i.e., metal–rich and/or CN–strong stars) existed and spectrum synthesis was used instead. We measured equivalent widths using a FORTRAN program developed for this project that interactively fits a Gaussian curve to each absorption line by implementing a Levenberg-Marquardt algorithm (Press et al. 1992) to find the least–squares fit given a continuum level and limits of integration. A high resolution, high S/N Arcturus spectrum333The Arcturus Atlas can be downloaded from the NOAO Digital Library at http://www.noao.edu/dpp/library.html. was simultaneously overplotted for each spectrum to aide in continuum placement and line identification. The program also has the ability to fit up to five Gaussians simultaneously for deblending purposes; however, all equivalent widths were verified independently using IRAF’s splot package.

4.2.1 Equivalent Width Analysis

Final abundances were calculated using the abfind driver in the 2002 version of the local thermodynamic equilibrium line analysis code MOOG (Sneden 1973). Adopted log gf values are the same as those employed in Johnson et al. (2006), which were adapted from line lists provided in Thévenin (1990), Sneden et al. (2004; modified from Ivans et al. 2001), and Cohen & Meléndez (2005). A summary of our line list is given in Table 4 and the measured equivalent widths are provided in Table 5.

While we had identified 20 Fe I lines for analysis, in most cases only 10–15 lines could be used due to severe line blending, bad ccd pixels, or line strength. In this sense, only lines lying on the linear part of the curve of growth were used, which meant neglecting almost all lines with a reduced width larger than about –4.5 (roughly 200 mÅ at 6600 Å). This unfortunately meant that many lines in metal–rich stars are too strong to give accurate abundances using our techniques. For the cases where Al abundances were determined using equivalent width measurements, weak line blends were taken into account using deblending methods. As stated above, stars with strong line blending or molecular line blanketing in the region near the Al doublet were analyzed with spectrum synthesis.

Typical uncertainties are quite small for [Fe/H] determinations with internal line–to–line spreads of 0.10–0.15 dex and / 0.05 dex on average. Sample spectra for stars of approximately the same T but different metallicities are shown in Figure 6. Here we illustrate that our [Fe/H] determinations are at the very least consistent in a relative sense as one notices the increasing Fe line strengths and CN–band strengths with increasing metallicity. The uncertainty in Al abundances is larger given that only two lines are available, but the two lines give a remarkably consistent abundance, with an average /=0.08 dex. It should be noted that in several of our spectra only one Al line could be confidently measured due mostly to bad pixels. In Figure 6, the reader can see the stark contrast in line strength between a star such as 51021, which has [Al/Fe]=+0.15 at [Fe/H]=–1.44, and star 61085, which has [Al/Fe]=+0.97 at [Fe/H]=–1.15. A summary of all derived abundances and associated / values is given in Table 6.

4.2.2 Spectrum Synthesis Analysis

As mentioned above, we determined Al abundances for metal–rich and/or CN–strong stars using the synth driver in MOOG. Candidates for spectrum synthesis were chosen based on visual inspection of the 6680–6700 Å region, where the majority of lines surrounding the Al doublet are CN lines. Stars where CN contamination was seen between the Al lines were designated for synthetic spectrum analysis (e.g., see Figure 6, lower two spectra).

The atomic line list (with the exception of the two Al lines) was taken from the Kurucz atomic line database444The Kurucz line list database can be accessed via http://kurucz.harvard.edu/linelists.html.. We adjusted the oscillator strengths from this line list so the line strengths matched those in the solar spectrum. For the CN molecular line list, we used a combination of one available from Kurucz and one provided by Bertrand Plez (2007, private communication; for a description on how the line list was prepared, see Hill et al. 2002).

Since most of the program stars do not have known C, N, or C/C abundances, we started with [C/Fe]=–0.5, [N/Fe]=+1.5, and C/C=5, values roughly consistent with previous work (e.g., Norris & Da Costa 1995; Smith et al. 2002). We then treated the nitrogen abundance as a free parameter and adjusted it until a satisfactory fit was achieved. Typical best fit [N/Fe] values were +1.0 to +1.5. To test the effect of different C/C ratios, we generated two sets of spectra with C/C=5 and C/C=1000. The fits to the CN lines were indistinguishable between the two cases, meaning C is the dominant isotope in this spectral region and thus synthesized CN lines are insensitive to the C abundance.

With the CN lines fit, we were then able to adjust the Al abundance until the synthetic spectrum matched the observed. Sample synthesis fits are given in Figure 7 for a metal–poor and metal–rich case. Aside from the CN lines, the Fe I line near the 6696 Å feature is the only other contaminating line in the region, but this line has an excitation potential of nearly 5 eV, making its contribution mostly negligible in these cool stars. Generally, the abundances given by the 6696 and 6698 Å lines agreed to within about 0.10 dex. Since a significant percentage of our Al abundances were determined using synthesis analyses, we tested for systematic offsets between synthesis and equivalent width methods. For sample stars that were both metal–poor and did not show signs of CN contamination, the difference in [Al/Fe] determined via both methods was less than 0.05 dex. However, for higher metallicity stars and those with possible CN contamination, the difference was 0.10–0.20 dex, with equivalent width analyses always overestimating the abundance. The quoted values for Al abundances derived via spectrum synthesis are given as the average from those two lines. A summary of our derived abundances is given in Table 6. Stars with Al determinations via synthesis are designated by “Syn” in the 6696 and 6698 Å columns of Table 5.

4.2.3 Abundance Sensitivity to Model Atmosphere Parameters

We tested the effects on derived abundances from changes in model atmosphere parameters by altering T 100 K, log g 0.25 cm s, and V 0.25 km s for models of [Fe/H]=–2.0, –1.5, and –1.0. As can be seen in Table 7, T uncertainties are the primary source of error for Fe I and Al I, and surface gravity is the primary source for Fe II abundances. This seems logical given that Fe I and Al I reside in a subordinate ionization state, and Fe II exists in the primary ionization state.

Following Table 7, an uncertainty of order 100 K in T leads to an error of 0.10–0.20 dex in Fe I, though the effect is somewhat reduced at higher metallicity. The opposite is true for Fe estimates based solely on the Fe II line, where the error range is 0.05–0.10 dex and the uncertainty becomes larger with increasing metallicity. Though the variation in Al I abundance as a function of T uncertainty is smaller than for Fe I, it is still of order 0.10 dex with a weak dependence on metallicity.

The effects of surface gravity uncertainty are of order 0.10 dex for the Fe II line, but are negligible for the neutral Fe and Al lines. For this reason, enforcing ionization equilibrium between different species is often used for constraining surface gravity estimates. As mentioned in 4.2.1, having only one Fe II line means the Fe abundance derived from Fe II is probably no more accurate than the typical line–to–line scatter present in Fe I (0.10–0.15 dex). Combined with the sensitivity of Fe II to surface gravity estimates of order 0.25 cm s, the fact that agreement between Fe I and Fe II is better than about 0.10 dex (see Figure 5) suggests estimates based on evolutionary arguments provide a decent approximation to the surface gravity; however, Table 7 shows this has little effect on our derived Fe I and Al I abundances. From this, we can safely assume that contamination from AGB stars, which have M0.60 M and thus a lower surface gravity, will not significantly alter our results.

The ad hoc microturbulence parameter, adjusted to remove abundance trends as a function of reduced width, has the strongest effect for lines lying on the flat part of the curve of growth. As is seen in Table 7, the effect on the Fe I abundance due to uncertainty in V increases with increasing metallicity because the lines become progressively stronger. However, Fe II and Al I are mostly unaffected due to their relatively small equivalent widths and the effect on Fe I is still 0.10 dex even at [Fe/H]=–1.0.

In addition to variations in model stellar atmosphere parameters we tested the sensitivity of Al abundance to CN strength via spectrum synthesis by varying [N/Fe]0.30 dex. Changing the nitrogen abundance by this amount worsens the fit to the CN lines in the spectrum, but alters the derived [Al/Fe] abundance less than 0.10 dex at all metallicities. Note that since [O/Fe] is unknown for most of our program stars and [O/Fe] can have values ranging from about +0.30 to less than –0.50, it is not possible to constrain the molecular equilibrium equations to derive true [C/Fe] and [N/Fe]. We present the [Al/Fe] results for each metallicity bin in Table 7.

4.3 Comparison with the Literature

While Cen has been the subject of multiple abundance studies (see 1 for a brief review), most of these are low resolution studies that do not involve elements other than Fe and/or Ca. Therefore, we are only comparing results in the literature for which moderate to high resolution Al data are available and with which we have three or more stars in common. This limits the comparison to Brown & Wallerstein (1993; 3 stars), Norris & Da Costa (1995; 24 stars), Zucker et al. (1996; 4 stars), and Smith et al. (2000; 3 stars).

In Figure 8, we present the values of T, log g, [Fe/H], and V given in the literature versus those obtained in this study. As can be seen from the figure, agreement is quite good for the temperature and surface gravity estimates, with the scatter increasing slightly for the metallicity and microturbulence estimates. For T, the average offset between our study and the literature is –7 K (50 K), and the average difference for log g is –0.02 cm s (0.10 cm s). This indicates that any disagreement between literature Fe and Al abundances and ours is not due to choices of T and log g. Similarly, [Fe/H] measurements agree to within 0.02 dex on average (0.20 dex). The reason for the larger dispersion in microturbulence estimates is not entirely clear, but it could be due to factors such as the number of lines available, data quality, continuum placement, and type of lines used (i.e., high and/or low excitation potential). However, on average the agreement is within 0.10 km s (0.25 km s).

Comparison between our derived [Al/Fe] abundances versus those in the literature are provided in Figure 9. Given the various data qualities, choices of model atmospheres and parameters, and adopted atomic line data, agreement is again quite good. The average offset between our derived abundances and those available in the literature is 0.06 dex (0.30 dex). Given that typical uncertainties in [Al/Fe] are of order 0.10–0.20 dex, agreement is comparable to that range.

5 Results and Discussion

5.1 Fe Abundances

As discussed in 1, it has been known for many years and shown by several authors that Cen has a considerable spread in metallicity that ranges from slightly less than [Fe/H]=–2.0 to more than [Fe/H]=–0.7. While several lower resolution spectroscopic (Norris et al. 1996; Suntzeff & Kraft 1996; Sollima et al. 2005b; Kayser et al. 2006; Stanford et al. 2006; Stanford et al. 2007; van Loon et al. 2007555The referee noted discrepancies between the [Fe/H] values derived by Norris & Da Costa (1995) and van Loon et al. (2007). We note that our results agree with Norris & Da Costa and a detailed resolution of this problem is beyond the scope of this paper.; Villanova et al. 2007) and photometric (Lee et al. 1999; Hilker & Richtler 2000; Hughes & Wallerstein 2000; Pancino et al. 2000; van Leeuwen et al. 2000; Rey et al. 2004; Stanford et al. 2004; Sollima et al. 2005a; Stanford et al. 2006) studies have obtained metallicity estimates for a large number of stars (N500 in some cases), there have only been a few high resolution spectroscopic studies with a significant number (N10) of stars observed (Norris & Da Costa 1995; Smith et al. 2000; Piotto et al. 2005; Sollima et al. 2006). However, aside from the present study, Norris & Da Costa (1995) still represents the largest (N=40) single high resolution analysis of Cen RGB stars. The general results from the metallicity studies can be summarized as: (1) few stars exist at [Fe/H]–2.0, (2) a primary peak in the metallicity distribution is observed at [Fe/H]–1.8 to –1.6, (3) there is a long tail of increasing metallicity up to [Fe/H]–0.5, and (4) there appear to be multiple peaks in the distribution at various [Fe/H] values.

In Figure 10, we present a histogram of our derived metallicity distribution function for all 180 stars. We find in agreement with previous studies that there are at least four distinct populations with the most metal–poor having [Fe/H]–1.75, the two intermediate metallicity populations have [Fe/H]–1.45 and –1.05, and the most metal–rich population has [Fe/H]–0.75. While our observations are skewed towards observing more metal–poor stars (see Figure 2), there are intrinsically more metal–poor than metal–rich stars, as can be seen in Figure 1. This means our derived metallicity distribution is affected by both the actual distribution and observational selection effects. Given that we only observed one star on the most metal–rich branch, it is possible that stars with metallicities higher than [Fe/H]=–0.75 exist. However, since our observed completion fraction is significantly higher for the most metal–poor stars, it is likely that our observed distribution function is accurate in a relative sense such that the cluster was rapidly enriched from the primordial metallicity of [Fe/H]–2.15 to the first major epoch of star formation at [Fe/H]–1.75. The absence of stars more metal–poor than [Fe/H]–2.2 means the proto– Cen environment was already pre–enriched, perhaps from processes such as cloud–cloud collisions (Tsujimoto et al. 2003), when the primary metal–poor population formed. In contrast, field stars in the Galactic halo exhibit a wide range of metallicities from [Fe/H]0.0 to [Fe/H]–4.0 (e.g., Gratton et al. 2004), indicating that the two do not share a common chemical enrichment history.

The distribution shown in Figure 10 suggests that if Cen evolved as a single entity (i.e., without significant contributions from mergers), then there were four to five significant star formation episodes that occurred. This seems to fit the high resolution photometric data from Sollima et al. (2005a) and Villanova et al. (2007) that show the multiple giant branches appear in discrete groups instead of as a continuous distribution. This trend is similarly reproduced in Figure 11, where our derived metallicities are superimposed on the photometric data from van Leeuwen et al. (2000). Here, even when binning by the approximate 3 value of each peak in the distribution from Figure 10 (0.3 dex), the different metallicity groups can be separated. The metallicity distribution from Figure 10 is very well produced in the hydrodynamical chemical enrichment simulations of Marcolini et al. (2007), where they assumed Cen is the core remnant of a dwarf spheroidal galaxy that was captured and tidally stripped 10 Gyr ago with star formation occurring over roughly 1.5 Gyr. The simulated metallicity peaks from Marcolini et al. (2007) lie at [Fe/H]–1.6, –1.35, –1.0, and –0.70, which are very similar to ours at [Fe/H]=–1.75, –1.45, –1.05, and –0.75.

There is some evidence that different metallicity populations may be spatially and kinematically unique (Norris et al. 1996; 1997; Suntzeff & Kraft 1996; Hilker & Richtler 2000; Pancino et al. 2000; 2003). In Figure 12, we present Fe and Al abundances as a function of distance from the cluster center. Keeping in mind our observational bias, we find a marginal tendency for the more metal–rich stars to be located in the inner regions of the cluster while the more metal–poor stars are rather evenly distributed at all radii sampled here. However, given our small sample size in the metal–rich regime, we are unable to make any definitive arguments for or against a metallicity–radius relationship. It should be noted though that Ikuta & Arimoto (2000) and Rey et al. (2004) do not find any strong evidence for the metal–poor and metal–rich populations having a spatially different structure. Even though the relaxation time for Cen is thought to exceed 5 Gyr (Djorgovski 1993; Merritt et al. 1997), any correlation between projected spatial position and metallicity is apparently subtle. However, it has been pointed out in deep photometric surveys (e.g., Rey et al. 2004) that the most metal–rich RGB–a is predominately seen in CMDs of the inner region of the cluster.

The main result indicating that at least the most metal–rich population may have a different formation history is that those stars appear to have a lower velocity dispersion (i.e. are kinematically cooler) than the other populations and do not show signs of rotation (Norris et al. 1997). In Figure 13 we show our derived radial velocities plotted both as a function of log (Fe)666log (X)=log(N/N)+12 and log (Al), where the error bars indicate the velocity dispersion in the data. To within one standard deviation, we do not find significant evidence for any of the stellar populations having a different bulk radial velocity or velocity dispersion. It seems unlikely that a larger sample size would provide significantly different results because Reijns et al. (2006) determined radial velocities for nearly 2000 Cen members and concluded the RGB–a stars had radial velocity and dispersion values consistent with the entire cluster. Pancino et al. (2007) have shown the rotational velocities for all populations are comparable to one another, but interestingly they find an underlying sinusoidal pattern in their measured velocities as a function of position angle. However, the metal–poor, intermediate metallicity, and anomalous giant branches all show the same sinusoidal pattern. Whether any true kinematic anomaly exists for this cluster or not remains to be seen.

5.2 Al Abundances

The bulk of aluminum production in galaxies and globular clusters is thought to arise from quiescent carbon and neon burning in massive stars (M8 M) and HBB occurring in the envelopes of IM–AGB stars via the MgAl cycle (e.g., Arnett & Truran 1969; Arnett 1971). In most Galactic globular clusters, there is a very small (0.10 dex) spread in the abundance of heavy and Fe–peak elements, with a somewhat larger spread (0.3–0.6 dex) in s– and r–process elements (e.g., Sneden et al. 2000). However, the lighter elements carbon through aluminum are typically not uniform and in some cases show star–to–star variations of more than a factor of 10. While Cen does not share all of the same chemical characteristics as globular clusters, the primary production locations of each element should be similar to globular clusters and/or the Galactic halo. The lesson learned from the monometallicity of “normal” globular clusters is that however Al manifests itself onto the surface of stars, the process must not alter Fe–peak, s–process, or r–process abundance ratios. This means that the often large star–to–star variation of [Al/Fe] seen in globular clusters (but not in halo field stars) are not due to supernova yields or the s–process, leaving either in situ deep mixing or HBB as the possible sites for [Al/Fe] variation. With these two scenarios in mind, we explore Al abundances with the goal of helping to constrain the source of Al variation and chemical evolution in Cen.

While the literature on Fe abundances for both evolved and main sequence stars is quite extensive, the spectroscopic surveys by Norris & Da Costa (1995) and Smith et al. (2000) represent the only studies to consider light element abundances that include Al for a large (N10) number of RGB stars in Cen. The results of those two studies indicate that the full range of [Al/Fe] is larger than 1.0 dex, Al and Na are correlated, Al and O are anticorrelated, and there is a hint of a decrease in [Al/Fe] with increasing [Fe/H]. We present the results of our larger sample plotting [Al/Fe] as a function of [Fe/H] in Figure 14. Even for the lowest metallicity stars, a large range in [Al/Fe] of 0.70 dex is already present. Near the first metallicity peak at [Fe/H]=–1.75, where it is assumed the first episode of star formation after the initial enrichment period occurred, the full range in [Al/Fe] reaches a maximum value of 1.3 dex. This star–to–star variation remains mostly constant until about [Fe/H]=–1.4, where the variation begins to decrease smoothly with increasing [Fe/H]. Interestingly, the “floor” Al abundance remains mostly constant at [Al/Fe]+0.15, regardless of the star’s metallicity; a characteristic shared with many globular clusters of various metallicity and in agreement with [Al/Fe] values typical of Galactic halo stars in Cen’s metallicity regime.

In Figure 15, we overlay a boxplot on top of the underlying distribution from Figure 14. The median [Al/Fe] ratio typically resides between about 0.45 and 0.80 dex for all well–sampled metallicities, with a relatively constant interquartile range. This implies that the average amount of Al in the cluster must increase with increasing Fe abundance, at least up to [Fe/H]–1.4. This result is confirmed in Figure 16, where log (Al) is plotted against log (Fe). It appears that for metallicities higher than about log (Fe)=6.0 ([Fe/H]–1.50), log (Al) no longer increases beyond log (Al)6.40 and the star–to–star scatter decreases. This result is likely robust against our observational bias because all stars observed in the metal–rich regime are located at or near the RGB tip (see Figure 1), where it is believed any Al enhancements due to deep mixing should be the most apparent. However, no obvious trend is seen between Al abundance and evolutionary state.

As discussed previously, there is some evidence for a correlation between Fe abundance and distance from the cluster center and we show the results from this study in the bottom panel of Figure 12. In the top panel of Figure 12, we present the same data but for Al instead of Fe. While there may be a tendency for the most metal–rich stars to be located inwards of about 10–15, there is no evidence of a trend for Al. Instead, stars of varying Al abundance are uniformly spread throughout the entire region sampled, at least out to 20. Likewise, the top panel of Figure 13 shows average radial velocities for Al abundances in 0.10 dex bins. To within uncertainties, there appears to be no trend in either radial velocity or velocity dispersion with log (Al). The fact that we do not find any preference of Al abundance or star–to–star dispersion with distance from the cluster center or radial velocity suggests star formation occurred on timescales shorter than those required to uniformly mix the gas.

5.3 Possible Implications on Chemical Evolution

From our available spectroscopic data for 180 RGB stars, we have confirmed the existence of at least four stellar populations ranging in metallicity from –2.2[Fe/H]–0.70, in agreement with previous photometric, low resolution spectroscopic, and smaller sample high resolution spectroscopic studies. Additionally, we have determined [Al/Fe] abundances for about 165 giants, most of which for the first time, with a sample larger by more than a factor of four than what was previously available in the literature. We find a constant Al abundance floor of [Al/Fe]+0.15 present at all metallicities, but with a largely varying and metallicity dependent spread above the floor. The star–to–star variation reaches a maximum extent in the intermediate metallicity regime, which is consistent with the second peak in the metallicity distribution, and begins to decline at higher metallicities. The floor itself is consistent with observations of field stars and is predicted by Galactic chemical evolution models, but the large [Al/Fe] variations are not predicted. Observations of some Galactic globular cluster stars, especially more metal–poor than [Fe/H]–1.5, show similar large star–to–star variations in [Al/Fe]. Combining our determined Fe and Al abundances with those available in the literature for these and other elements now allows us to examine each metallicity regime in turn.

5.3.1 The Metal–Poor Population

A prominent feature of the metal–poor stars ([Fe/H]–1.6) in Cen is the rapidly increasing abundances of Na, Al, and light and heavy s–process elements relative to Fe as the metallicity increases from [Fe/H]=–2.2 to the first metallicity peak at [Fe/H]=–1.75 (e.g., Norris & Da Costa 1995; Smith et al. 2000). These increases are accompanied by nearly constant heavy [/Fe]+0.30, low Cu abundances ([Cu/Fe]–0.60), and low r–process abundances ([Eu/Fe]–0.50). These results seem to indicate that massive stars exploding as type II SNe are the primary contributors for Fe–peak and heavy –element enhancement in the cluster, but the low Eu abundances, which should be synthesized in the same stars, are puzzling. Additionally, the growing s–process component appears to be best fit by models of 1.5–3 M AGB ejecta (Smith et al. 2000). The lack of clear evidence for type Ia SNe having contributed to the chemical composition of metal–poor stars in Cen (e.g., Smith et al. 2000; Cunha et al. 2002; Pancino et al. 2002; Platais et al. 2003) is consistent with the 1 Gyr timescales needed for type Ia SNe to evolve and the fact that they might not efficiently form in metal–poor environments (Kobayashi et al. 1998).

As mentioned above, the majority of Al present in the atmospheres of these RGB stars was likely produced in type II SNe explosions that polluted the pristine gas from which these stars formed. While the heavy element data do not support high mass (8M) stars being the source for the more than 1.0 dex [Al/Fe] variations, that may be explained from HBB occurring in IM–AGB stars, in situ deep mixing, or a hybrid scenario. In Figures 1416, we have shown that [Al/Fe]0 for all metal–poor stars sampled, but a constant Al abundance floor is setup at [Al/Fe]+0.15 with a rapidly increasing star–to–star dispersion that reaches about 1.3 dex in extent by [Fe/H]=–1.75. For the neutron capture elements, which are the only other group exhibiting a variations with metallicity, Smith et al. (2000) showed stars with [Fe/H]–2 are dominated by an r–process component with a shift to a primarily s–process component by [Fe/H]–1.8.

In the pure pollution scenario, which does not invoke deep mixing affecting elements heavier than N, type II SNe, low and IM–AGB stars, and perhaps winds from less evolved very massive stars (e.g., Maeder & Meynet 2006) are responsible for all abundance anomalies. Adding our large Al data set to the sample of stars previously observed may help constrain enrichment timescales and polluting AGB masses. Conventional theory suggests light and s–process elements do not share the same origin and Cen’s s–process component is best fit with lower mass AGB stars, but masses lower than 3–4 M undergo third dredgeup without significant HBB (e.g., Karakas & Lattanzio 2007) and thus should not appreciably alter their envelope Al abundances. Additionally, Ventura & D’Antona (2007) suggest globular cluster light element anomalies can only be explained with ejecta from AGB stars in the mass range of 5–6.5 M. While our sample only includes two stars with [Fe/H]–2 (36036 & 51091), the elevated [Al/Fe] ratios of +0.40 and +1.13 suggest IM–AGB stars, with lifetimes of about 50–15010 yrs (Schaller et al. 1992), have already polluted the Cen system. In this case, the low metallicity environment would favor high [Al/Fe] yields from HBB processes occurring in IM–AGB stars. The rapidly rising average value of log (Al) shown in Figure 16 in the metallicity regime –2.0[Fe/H]–1.6 implies a continued contribution from IM–AGB stars, presumably forming from the same star formation event that creates the first peak in the metallicity distribution. The top two panels of Figure 17 show binned [Al/Fe] for this metallicity regime and we note approximately four sub–populations with [Al/Fe]+0.15, +0.45, +0.85, and +1.05. Predicted yields from type II SNe (e.g., Woosley & Weaver 1995) and measurements of field stars (e.g., Fulbright 2000) suggest type II SNe should enrich the ISM with [Al/Fe]+0.10 to +0.30 while 5–6.5 M AGB stars should produce [Al/Fe]+0.50 to +1.10 (e.g., D’Antona & Ventura 2007), which could explain our observed distribution. Given the rather short lifetimes of stars believed to produce Al and the fact that evidence for 1.5–3.0 M pollution does not appear until [Fe/H]–1.8, it would seem that Cen was probably enriched from [Fe/H]=–2.2 to –1.75 in 0.5–1.0 Gyr.

5.3.2 The Intermediate Metallicity Populations

For the two intermediate metallicity populations ([Fe/H]=–1.45 and [Fe/H]=–1.05), the heavy [/Fe] ratio remains constant and the s–process abundances level off with very little star–to–star dispersion (Norris & Da Costa 1995; Smith et al. 2000). As in the most metal–poor stars, r–process and Cu ratios relative to Fe remain low and mostly unchanged. However, the star–to–star scatter in O, Na, and Al is still quite large. It is interesting to point out that log (Al) reaches its maximum value at about the same metallicity at which the s–process elements reach a constant ratio relative to Fe. The [Al/Fe] abundance floor is constant throughout this metallicity regime at [Al/Fe]+0.15, which means the scatter, still considerably larger than for [Ba/Fe], decreases as a function of increasing metallicity. This trend should presumably be present for Na and in the opposite sense for O assuming the Na–Al correlation and O–Al anticorrelation exist at all metallicities.

Had the scatter in Al abundances been comparable to that of other heavier elements in this metallicity range (0.10–0.30 dex) with a nearly constant [Al/Fe] ratio, as is seen in field stars, we might be inclined to believe Al enhancement in the cluster was due solely to production in massive stars and that typical type II SNe ejecta have [Al/Fe]+0.15. It is interesting to note that the [Al/Fe] floor tracks closely (with a slight offset of 0.2-0.3 dex) to the Galactic chemical evolution model presented in Timmes et al. (1995; their Figure 19), assuming the amount of Fe ejected is decreased by a factor of two, and Samland (1998; their Figure 10), with an increase in secondary (i.e., metal–dependent) Al production by a factor of five. If the well–known light element correlations/anticorrelations seen in previously observed Cen stars (e.g., Norris & Da Costa 1995) holds at all metallicities and for all stars, those with [Al/Fe]+0.15 should also have [O/Fe]+0.30, heavy [/Fe]+0.30, and [Na/Fe]–0.20, which are consistent with predicted yields from type II SNe (e.g., Woosley & Weaver 1995). It could be that these stars formed preferentially out of SNe ejecta without significant IM–AGB contamination.

While the maximum observed log (Al) increases with metallicity for the most metal–poor Cen giants, this trend halts at [Fe/H]–1.4, which coincides with the second peak in the metallicity distribution (i.e., the next round of star formation). We know the heavy [/Fe], [Ba/Fe], and floor [Al/Fe] ratios remain constant at higher metallicities, indicating an increase in log (Ba), log (), and the minimum log (Al) that track with Fe. The question now posed by the Al data is why does the process producing the high Al values shut off or become less efficient at [Fe/H]–1.45? Increases in metallicity lead to lower temperatures at the bottom of the convective envelope and require higher masses for HBB to occur. It may be that we are observing the result of lower convective efficiency at higher metallicity and/or that fewer IM stars form in higher metallicity environment. IM–AGB models in the metallicity range of –1.5[Fe/H]–0.7 (e.g., Fenner et al. 2004; Ventura & D’Antona 2007; 2008) predict [Al/Fe] yields of +0.5 to +1.0, with lower [Al/Fe] yields at higher [Fe/H]. This may explain the bimodal distribution in the bottom panels of Figure 17, with the abundances in between possibly being due to varying degrees of ejecta dilution. The fact that the metallicity at which the heavy elements cease to increase in abundance more quickly than Fe and the metallicity where the maximum [Al/Fe] begins to decrease coincide suggests an important parameter changed in Cen at this point in its evolution. It may even be the case that this is when the progenitor dwarf galaxy began to change structurally via encounters with the Galactic disk. It appears that at metallicities higher than [Fe/H]=–1.45, the cluster slowly approaches a constant [Al/Fe], which is consistent with values observed in the halo.

While type Ia ejecta have been mostly ruled out by previous studies as contributors to the most metal–poor population, the metallicity at which they become important contributors is unclear. Marcolini et al. (2007) claim that their intermediate metallicity peak at [Fe/H]–1.4 is due primarily to inhomogeneous pollution by type Ia SNe. It is interesting to note that in this same metallicity bin we find a median [Al/Fe] value about 0.40 dex lower than the two surrounding bins as well as the only star with [Al/Fe]+0.15. It is uncertain whether this is a real effect or simply due to an anomalous selection of stars. Inhomogeneous pollution by type Ia SNe may also explain the bimodal distribution seen in the bottom panels of Figure 17 where stars polluted by both type Ia ejecta and IM–AGB stars exhibit lower [Al/Fe] ratios and “normal” stars polluted by type II SNe and IM–AGB stars have higher [Al/Fe] values. While the same trend is not particularly apparent for s–process elements (e.g., Smith et al. 2000), this may be due to a smaller sample size, especially if inhomogeneous pollution only affected a small percentage of intermediate metallicity stars; however, this could explain the few observations in the literature of stars with [Fe/H]–1.4 and [Ba/Fe]0 (e.g., Smith et al. 1995).

5.3.3 The Metal–Rich Population

For stars more metal–rich than [Fe/H]–1, there is some evidence of a decrease in [/Fe] and an increase in [Cu/Fe] (Pancino et al. 2002; but see also Cunha et al. 2002), which, if true, likely indicates an increased contribution from type Ia SNe. Similarly, there appears to be a decrease in [Eu/Fe] with perhaps a similar decrease in the abundance of s–process elements relative to Fe (Norris & Da Costa; Smith et al. 2000). Although the Al data are rather incomplete in this metallicity regime, the general trends seen in slightly more metal–poor stars appear to continue.

While the scope of an age spread amongst the various metallicity populations is still unknown, the Al data presented here seem to indicate that the age difference between the intermediate and metal–rich populations is not especially large. In particular, stars with the largest values of log (Al) appear with [Fe/H] ranging from –1.5 to –0.7, perhaps indicating that they formed from gas polluted by the same generation of IM–AGB ejecta. In this scenario, the lower [Al/Fe] ratios at high metallicity might be due to those stars forming in regions where [Fe/H] increased due to inhomogeneous pollution by type Ia SNe, as mentioned in Marcolini et al. (2007). In their scenario, this effect should be more important for the inner regions of the cluster. This may be corroborated by our finding that there is no apparent relationship between log (Al) and distance from the cluster center, but a trend might be present for Fe such that stars with [Fe/H]–1 are preferentially located closer to the cluster center. In any case, additional data are required in this metallicity regime to determine whether the decreasing [Al/Fe] ratios are a real effect or the result of incomplete statistics. It will be interesting to see if O and Na display similar behavior to Al as a function of [Fe/H].

6 Summary

We have determined radial velocities, Fe, and Al abundances for 180 RGB stars in the Galactic globular cluster Cen using moderate resolution (R13,000) spectroscopy. The bulk of our sample includes stars with V14.0, but an observational bias is present such that we preferentially observed more luminous and more metal–poor stars. The spectra ranged from 6500–6750 Å and Fe abundances were based on an average of approximately 10–20 Fe I lines. Al abundances were determined using either spectrum synthesis or equivalent width analyses of the 6696, 6698 Å Al I doublet, with synthesis being reserved for CN–strong and/or metal–rich stars.

With respect to our determined Fe abundances, we find in agreement with previous studies that at least four or more different metallicity populations are present in the cluster. Peaks in the metallicity distribution function appear at [Fe/H]=–1.75, –1.45, –1.05, and –0.75, indicating the presence of multiple star formation episodes. We do not find evidence suggesting any of the different metallicity populations are kinematically or spatially unique, but it should be noted that our observed completion fraction is low for stars more metal–rich than [Fe/H]–1.0 and we only observed stars between about 2 and 20 from the cluster center.

Our Al data corroborate the Fe results such that there does not appear to be any correlation between Al abundance and distance from the cluster center or radial velocity. This suggests that the cluster gas was not significantly mixed while star formation was still occurring. In a plot of [Al/Fe] versus [Fe/H], the data reveal a star–to–star variation of nearly 1.3 dex that stays mostly constant until [Fe/H]–1.45, in which case the spread in [Al/Fe] declines monotonically with increasing [Fe/H]. Additionally, the [Al/Fe] floor remains nearly constant across all metallicities sampled here at [Al/Fe]+0.15. This result is similar to what is predicted based on type II SNe yields and closely mimics the trend seen in Galactic halo field stars. The anomalously low median [Al/Fe] ratio at [Fe/H]=–1.45 may be evidence for inhomogeneous pollution from type Ia SNe and could explain the bimodal [Al/Fe] distribution seen in intermediate metallicity stars, but more observations are required to confirm whether this is real or the result of an incomplete sample.

The source of the [Al/Fe] spread that has also been observed in other light elements remains an open problem, but the results obtained here pose some interesting questions. A plot of log (Al) versus log (Fe) shows that log (Al) no longer increases beyond about 6.40 at metallicities higher than [Fe/H]–1.45, which is coincident with the second peak in the metallicity distribution function. Apparently, whatever process is responsible for manifesting very high Al abundances shuts down or becomes less efficient at intermediate and high metallicities. In “normal” metal–poor globular clusters, the large star–to–star variations seen in the light elements are not shared by Fe–peak and neutron capture elements, and it has been suggested that HBB occurring in IM–AGB stars or in situ deep mixing are responsible for the light element abundance anomalies. Without a comparable sample of O and Na data to supplement the Al abundances here, it is difficult to determine the role either source plays. However, AGB yields of stars undergoing HBB indicate stars forming from material polluted by AGB ejecta can only reach [Al/Fe] ratios between about +0.5 and +1.0, with perhaps slightly lower and higher values being reached in higher and lower metallicity environments, respectively.

It may be possible to explain the Al data such that core–collapse SNe drive the [Al/Fe] floor and an AGB mass spectrum with varying HBB efficiencies and mixing depths are responsible for much of the additional scatter present. The decrease in the maximum [Al/Fe] with increasing [Fe/H] might then be attributed to requiring higher mass stars for HBB to occur at temperatures adequate to activate the full Mg to Al cycle, which means the burning material is exposed for a shorter amount of time and thus leads to less [Al/Fe] enhancement. Whether this can be made to work quantitatively in light of the problems associated with AGB pollution scenarios (see 1) remains to be seen.

We would like to thank the anonymous referee for a detailed and helpful report which improved the manuscript and for pointing out the possible significance of type Ia SN pollution at intermediate metallicities. We would also like to thank Bob Kraft and Chris Sneden for helpful discussions regarding this paper and Bertrand Plez for providing an electronic copy of his CN linelist. This publication makes use of data products from the Two Micron All Sky Survey, which is a joint project of the University of Massachusetts and the Infrared Processing and Analysis Center/California Institute of Technology, funded by the National Aeronautics and Space Administration and the National Science Foundation. This research has made use of NASA’s Astrophysics Data System Bibliographic Services. This research has made use of the SIMBAD database, operated at CDS, Strasbourg, France. Support for DS was provided by grant AST-0139617 from the NSF for a summer REU program. Support of the College of Arts and Sciences and the Daniel Kirkwood fund at Indiana University Bloomington for CIJ is gratefully acknowledged. Facilities: CTIO

References

  • Alonso et al. (1994) Alonso, A., Arribas, S., & Martinez-Roger, C. 1994, A&A, 282, 684
  • Alonso et al. (1998) Alonso, A., Arribas, S., & Martinez-Roger, C. 1998, A&AS, 131, 209
  • Alonso et al. (1999) Alonso, A., Arribas, S., & Martínez-Roger, C. 1999, A&AS, 140, 261
  • Alonso et al. (2001) Alonso, A., Arribas, S., & Martínez-Roger, C. 2001, A&A, 376, 1039
  • Anders & Grevesse (1989) Anders, E., & Grevesse, N. 1989, Geochim. Cosmochim. Acta, 53, 197
  • Anderson (1997) Anderson, J. 1997, Ph.D. thesis, Univ. California at Berkeley
  • Arnett (1971) Arnett, W. D. 1971, ApJ, 166, 153
  • Arnett & Truran (1969) Arnett, W. D., & Truran, J. W. 1969, ApJ, 157, 339
  • Bedin et al. (2004) Bedin, L. R., Piotto, G., Anderson, J., Cassisi, S., King, I. R., Momany, Y., & Carraro, G. 2004, ApJ, 605, L125
  • Bekki & Norris (2006) Bekki, K., & Norris, J. E. 2006, ApJ, 637, L109
  • Bell et al. (1979) Bell, R. A., Dickens, R. J., & Gustafsson, B. 1979, ApJ, 229, 604
  • Bellman et al. (2001) Bellman, S., Briley, M. M., Smith, G. H., & Claver, C. F. 2001, PASP, 113, 326
  • Blackwell & Shallis (1977) Blackwell, D. E., & Shallis, M. J. 1977, MNRAS, 180, 177
  • Boesgaard et al. (2005) Boesgaard, A. M., King, J. R., Cody, A. M., Stephens, A., & Deliyannis, C. P. 2005, ApJ, 629, 832
  • Briley et al. (2004a) Briley, M. M., Cohen, J. G., & Stetson, P. B. 2004a, AJ, 127, 1579
  • Briley et al. (2004b) Briley, M. M., Harbeck, D., Smith, G. H., & Grebel, E. K. 2004b, AJ, 127, 1588
  • Brown & Wallerstein (1993) Brown, J. A., & Wallerstein, G. 1993, AJ, 106, 133
  • Calamida et al. (2005) Calamida, A., et al. 2005, ApJ, 634, L69
  • Cannon et al. (1998) Cannon, R. D., Croke, B. F. W., Bell, R. A., Hesser, J. E., & Stathakis, R. A. 1998, MNRAS, 298, 601
  • Carbon et al. (1982) Carbon, D. F., Romanishin, W., Langer, G. E., Butler, D., Kemper, E., Trefzger, C. F., Kraft, R. P., & Suntzeff, N. B. 1982, ApJS, 49, 207
  • Carpenter (2001) Carpenter, J. M. 2001, AJ, 121, 2851
  • Castelli et al. (1997) Castelli, F., Gratton, R. G., & Kurucz, R. L. 1997, A&A, 318, 841
  • Charbonnel & Do Nascimento (1998) Charbonnel, C., & Do Nascimento, J. D., Jr. 1998, A&A, 336, 915
  • Charbonnel & Zahn (2007) Charbonnel, C., & Zahn, J.-P. 2007, A&A, 467, L15
  • Cohen et al. (2002) Cohen, J. G., Briley, M. M., & Stetson, P. B. 2002, AJ, 123, 2525
  • Cohen & Meléndez (2005) Cohen, J. G., & Meléndez, J. 2005, AJ, 129, 303
  • Cunha et al. (2002) Cunha, K., Smith, V. V., Suntzeff, N. B., Norris, J. E., Da Costa, G. S., & Plez, B. 2002, AJ, 124, 379
  • D’Antona & Ventura (2007) D’Antona, F., & Ventura, P. 2007, MNRAS, 379, 1431
  • Denisenkov & Denisenkova (1990) Denisenkov, P. A., & Denisenkova, S. N. 1990, Soviet Astronomy Letters, 16, 275
  • Denissenkov & Weiss (2001) Denissenkov, P. A., & Weiss, A. 2001, ApJ, 559, L115
  • Denissenkov & Herwig (2003) Denissenkov, P. A., & Herwig, F. 2003, ApJ, 590, L99
  • Dickens et al. (1991) Dickens, R. J., Croke, B. F. W., Cannon, R. D., & Bell, R. A. 1991, Nature, 351, 212
  • Dinescu et al. (1999) Dinescu, D. I., Girard, T. M., & van Altena, W. F. 1999, AJ, 117, 1792
  • Djorgovski et al. (1997) Djorgovski S., 1993, ASP Conf. Ser. Vol. 50, Structure and Dynamics of Globular Clusters. Astron. Soc. Pac., San Francisco, p. 373
  • Fenner et al. (2004) Fenner, Y., Campbell, S., Karakas, A. I., Lattanzio, J. C., & Gibson, B. K. 2004, MNRAS, 353, 789
  • Ferraro et al. (2004) Ferraro, F. R., Sollima, A., Pancino, E., Bellazzini, M., Straniero, O., Origlia, L., & Cool, A. M. 2004, ApJ, 603, L81
  • Fulbright (2002) Fulbright, J. P. 2002, AJ, 123, 404
  • Gnedin et al. (2002) Gnedin, O. Y., Zhao, H., Pringle, J. E., Fall, S. M., Livio, M., & Meylan, G. 2002, ApJ, 568, L23
  • Gratton et al. (2000) Gratton, R. G., Sneden, C., Carretta, E., & Bragaglia, A. 2000, A&A, 354, 169
  • Gratton et al. (2001) Gratton, R. G., et al. 2001, A&A, 369, 87
  • Gratton et al. (2004) Gratton, R., Sneden, C., & Carretta, E. 2004, ARA&A, 42, 385
  • Harris (1996) Harris, W. E. 1996, AJ, 112, 1487
  • Hilker & Richtler (2000) Hilker, M., & Richtler, T. 2000, A&A, 362, 895
  • Hilker et al. (2004) Hilker, M., Kayser, A., Richtler, T., & Willemsen, P. 2004, A&A, 422, L9
  • Hill et al. (2002) Hill, V., et al. 2002, A&A, 387, 560
  • Hughes & Wallerstein (2000) Hughes, J., & Wallerstein, G. 2000, AJ, 119, 1225
  • Hughes et al. (2004) Hughes, J., Wallerstein, G., van Leeuwen, F., & Hilker, M. 2004, AJ, 127, 980
  • Ikuta & Arimoto (2000) Ikuta, C., & Arimoto, N. 2000, A&A, 358, 535
  • Ivans et al. (1999) Ivans, I. I., Sneden, C., Kraft, R. P., Suntzeff, N. B., Smith, V. V., Langer, G. E., & Fulbright, J. P. 1999, AJ, 118, 1273
  • Ivans et al. (2001) Ivans, I. I., Kraft, R. P., Sneden, C., Smith, G. H., Rich, R. M., & Shetrone, M. 2001, AJ, 122, 1438
  • Johnson (1965) Johnson, H. L. 1965, ApJ, 141, 923
  • Johnson et al. (2005) Johnson, C. I., Kraft, R. P., Pilachowski, C. A., Sneden, C., Ivans, I. I., & Benman, G. 2005, PASP, 117, 1308
  • Johnson & Pilachowski (2006) Johnson, C. I., & Pilachowski, C. A. 2006, AJ, 132, 2346
  • Karakas & Lattanzio (2007) Karakas, A., & Lattanzio, J. C. 2007, Publications of the Astronomical Society of Australia, 24, 103
  • Kayser et al. (2006) Kayser, A., Hilker, M., Richtler, T., & Willemsen, P. G. 2006, A&A, 458, 777
  • Keller et al. (2001) Keller, L. D., Pilachowski, C. A., & Sneden, C. 2001, AJ, 122, 2554
  • Kobayashi et al. (1998) Kobayashi, C., Tsujimoto, T., Nomoto, K., Hachisu, I., & Kato, M. 1998, ApJ, 503, L155
  • Langer et al. (1986) Langer, G. E., Kraft, R. P., Carbon, D. F., Friel, E., & Oke, J. B. 1986, PASP, 98, 473
  • Langer et al. (1997) Langer, G. E., Hoffman, R. E., & Zaidins, C. S. 1997, PASP, 109, 244
  • Lattanzio et al. (2004) Lattanzio, J., Karakas, A., Campbell, S., Elliott, L., & Chieffi, A. 2004, Memorie della Societa Astronomica Italiana, 75, 322
  • Lee et al. (1999) Lee, Y.-W., Joo, J.-M., Sohn, Y.-J., Rey, S.-C., Lee, H.-C., & Walker, A. R. 1999, Nature, 402, 55
  • Lee et al. (2005) Lee, Y.-W., et al. 2005, ApJ, 621, L57
  • Maeder & Meynet (2006) Maeder, A., & Meynet, G. 2006, A&A, 448, L37
  • Marcolini et al. (2007) Marcolini, A., Sollima, A., D’Ercole, A., Gibson, B. K., & Ferraro, F. R. 2007, MNRAS, 382, 443
  • Mayor et al. (1997) Mayor, M., et al. 1997, AJ, 114, 1087
  • Merritt et al. (1997) Merritt, D., Meylan, G., & Mayor, M. 1997, AJ, 114, 1074
  • Meylan et al. (1995) Meylan, G., Mayor, M., Duquennoy, A., & Dubath, P. 1995, A&A, 303, 761
  • Norris & Da Costa (1995) Norris, J. E., & Da Costa, G. S. 1995, ApJ, 447, 680
  • Norris et al. (1996) Norris, J. E., Freeman, K. C., & Mighell, K. J. 1996, ApJ, 462, 241
  • Norris et al. (1997) Norris, J. E., Freeman, K. C., Mayor, M., & Seitzer, P. 1997, ApJ, 487, L187
  • Norris (2004) Norris, J. E. 2004, ApJ, 612, L25
  • Origlia et al. (2003) Origlia, L., Ferraro, F. R., Bellazzini, M., & Pancino, E. 2003, ApJ, 591, 916
  • Pancino et al. (2000) Pancino, E., Ferraro, F. R., Bellazzini, M., Piotto, G., & Zoccali, M. 2000, ApJ, 534, L83
  • Pancino et al. (2002) Pancino, E., Pasquini, L., Hill, V., Ferraro, F. R., & Bellazzini, M. 2002, ApJ, 568, L101
  • Pancino et al. (2003) Pancino, E., Seleznev, A., Ferraro, F. R., Bellazzini, M., & Piotto, G. 2003, MNRAS, 345, 683
  • Pancino et al. (2007) Pancino, E., Galfo, A., Ferraro, F. R., & Bellazzini, M. 2007, ApJ, 661, L155
  • Pilachowski (1988) Pilachowski, C. A. 1988, ApJ, 326, L57
  • Pilachowski et al. (1996) Pilachowski, C. A., Sneden, C., & Kraft, R. P. 1996, AJ, 111, 1689
  • Piotto et al. (2005) Piotto, G., et al. 2005, ApJ, 621, 777
  • Platais et al. (2003) Platais, I., Wyse, R. F. G., Hebb, L., Lee, Y.-W., & Rey, S.-C. 2003, ApJ, 591, L127
  • Press et al. (1992) Press, W. H., Teukolsky, S. A., Vetterling, W. T., & Flannery, B. R. 1992, Numerical Recipes in FORTRAN: The Art of Scientific Computing (2nd ed.; Cambridge: Cambridge Univ. Press)
  • Reijns et al. (2006) Reijns, R. A., Seitzer, P., Arnold, R., Freeman, K. C., Ingerson, T., van den Bosch, R. C. E., van de Ven, G., & de Zeeuw, P. T. 2006, A&A, 445, 503
  • Rey et al. (2004) Rey, S.-C., Lee, Y.-W., Ree, C. H., Joo, J.-M., Sohn, Y.-J., & Walker, A. R. 2004, AJ, 127, 958
  • Richer et al. (1991) Richer, H. B., Fahlman, G. G., Buonanno, R., Fusi Pecci, F., Searle, L., & Thompson, I. B. 1991, ApJ, 381, 147
  • Ryan et al. (1996) Ryan, S. G., Norris, J. E., & Beers, T. C. 1996, ApJ, 471, 254
  • Salaris et al. (2002) Salaris, M., Cassisi, S., & Weiss, A. 2002, PASP, 114, 375
  • Samland (1998) Samland, M. 1998, ApJ, 496, 155
  • Schaller et al. (1992) Schaller, G., Schaerer, D., Meynet, G., & Maeder, A. 1992, A&AS, 96, 269
  • Shetrone (1996) Shetrone, M. D. 1996, AJ, 112, 2639
  • Smith et al. (1995) Smith, V. V., Cunha, K., & Lambert, D. L. 1995, AJ, 110, 2827
  • Smith et al. (1996) Smith, G. H., Shetrone, M. D., Bell, R. A., Churchill, C. W., & Briley, M. M. 1996, AJ, 112, 1511
  • Smith et al. (2000) Smith, V. V., Suntzeff, N. B., Cunha, K., Gallino, R., Busso, M., Lambert, D. L., & Straniero, O. 2000, AJ, 119, 1239
  • Smith et al. (2002) Smith, V. V., Terndrup, D. M., & Suntzeff, N. B. 2002, ApJ, 579, 832
  • Smith et al. (2005) Smith, V. V., Cunha, K., Ivans, I. I., Lattanzio, J. C., Campbell, S., & Hinkle, K. H. 2005, ApJ, 633, 392
  • Smith (2006) Smith, G. H. 2006, PASP, 118, 1225
  • Sneden (1973) Sneden, C. 1973, ApJ, 184, 839
  • Sneden et al. (1991) Sneden, C., Kraft, R. P., Prosser, C. F., & Langer, G. E. 1991, AJ, 102, 2001
  • Sneden et al. (2000) Sneden, C., Pilachowski, C. A., & Kraft, R. P. 2000, AJ, 120, 1351
  • Sneden et al. (2004) Sneden, C., Kraft, R. P., Guhathakurta, P., Peterson, R. C., & Fulbright, J. P. 2004, AJ, 127, 2162
  • Sollima et al. (2005a) Sollima, A., Ferraro, F. R., Pancino, E., & Bellazzini, M. 2005a, MNRAS, 357, 265
  • Sollima et al. (2005b) Sollima, A., Pancino, E., Ferraro, F. R., Bellazzini, M., Straniero, O., & Pasquini, L. 2005b, ApJ, 634, 332
  • Sollima et al. (2006) Sollima, A., Borissova, J., Catelan, M., Smith, H. A., Minniti, D., Cacciari, C., & Ferraro, F. R. 2006, ApJ, 640, L43
  • Stanford et al. (2006) Stanford, L. M., Da Costa, G. S., Norris, J. E., & Cannon, R. D. 2006, ApJ, 647, 1075
  • Stanford et al. (2007) Stanford, L. M., Da Costa, G. S., Norris, J. E., & Cannon, R. D. 2007, ApJ, 667, 911
  • Suntzeff & Kraft (1996) Suntzeff, N. B., & Kraft, R. P. 1996, AJ, 111, 1913
  • Thevenin (1990) Thévenin, F. 1990, A&AS, 82, 179
  • Timmes et al. (1995) Timmes, F. X., Woosley, S. E., & Weaver, T. A. 1995, ApJS, 98, 617
  • Tsujimoto & Shigeyama (2003) Tsujimoto, T., & Shigeyama, T. 2003, ApJ, 590, 803
  • van de Ven et al. (2006) van de Ven, G., van den Bosch, R. C. E., Verolme, E. K., & de Zeeuw, P. T. 2006, A&A, 445, 513
  • van Leeuwen et al. (2000) van Leeuwen, F., Le Poole, R. S., Reijns, R. A., Freeman, K. C., & de Zeeuw, P. T. 2000, A&A, 360, 472
  • van Loon et al. (2007) van Loon, J. T., van Leeuwen, F., Smalley, B., Smith, A. W., Lyons, N. A., McDonald, I., & Boyer, M. L. 2007, MNRAS, 382, 1353
  • Ventura & D’Antona (2005a) Ventura, P., & D’Antona, F. 2005a, A&A, 431, 279
  • Ventura & D’Antona (2005b) Ventura, P., & D’Antona, F. 2005b, ApJ, 635, L149
  • Ventura & D’Antona (2008) Ventura, P., & D’Antona, F. 2008, A&A, 479, 805
  • Ventura & D’Antona (2008) Ventura, P., & D’Antona, F. 2008, MNRAS, 255
  • Villanova et al. (2007) Villanova, S., et al. 2007, ApJ, 663, 296
  • Woolley (1966) Woolley, R. R. 1966, Royal Observatory Annals, 2, 1
  • Woosley & Weaver (1995) Woosley, S. E., & Weaver, T. A. 1995, ApJS, 101, 181
  • Yong et al. (2003) Yong, D., Grundahl, F., Lambert, D. L., Nissen, P. E., & Shetrone, M. D. 2003, A&A, 402, 985
  • Zucker et al. (1996) Zucker, D., Wallerstein, G., & Brown, J. A. 1996, PASP, 108, 911
Hydra Wavelength UT Date Exposure
Setup [Å] [s]
1 6600 2003 July 17 1 x 1800
2 6600 2003 July 18 1 x 1800
2 6600 2003 July 18 4 x 2700
3 6600 2003 July 19 2 x 2700
3 6600 2003 July 19 1 x 3600
Table 1: Hydra Observations of Cen Giants
StaraaIdentifier from van Leeuwen et al. (2000). Alt. IDbbIdentifier from Woolley (1966). V Error from Mean Mem. Prob.ccMembership probability from van Leeuwen et al. (2000).
[km s] [km s]
9 370 211.9 1.8 1.7 99
5009 548 230.7 1.6 0.1 100
6017 240 248.4 1.6 1.4 98
10012 43 236.2 1.6 0.4 98
11019 537 238.3 1.7 0.6 99
11024 91 221.7 1.3 0.9 99
12013 394 230.4 1.4 0.1 98
14010 435 245.5 3.3 1.2 98
15022 180 227.5 1.5 0.4 100
16009 252 222.4 1.9 0.8 99
16015 213 225.5 1.6 0.5 100
17015 325 223.6 1.7 0.7 100
17032 605 238.9 1.5 0.6 100
18047 408 238.8 1.6 0.6 100
19062 464 225.2 1.8 0.6 98
20049 6355 231.5 2.7 0.0 100
21032 172 221.8 1.3 0.9 100
21042 348 232.1 1.7 0.0 99
23061 296 229.2 1.2 0.2 100
24013 56 221.3 1.6 0.9 98
24027 5969 220.2 1.7 1.0 100
24040 5993 258.8 2.2 2.3 100
24046 74 215.1 1.3 1.4 100
24056 364 235.1 1.5 0.3 100
24062 352 236.4 1.4 0.4 100
25018 5964 221.2 2.4 0.9 100
25043 89 215.0 1.4 1.4 100
25062 46 227.6 1.5 0.4 100
25065 227.0 1.9 0.4 100
25068 58 235.1 1.4 0.3 100
26025 61 242.3 1.5 0.9 100
26088 161 247.0 1.4 1.3 100
27048 313 240.0 1.6 0.7 100
27055 5837 234.5 2.4 0.2 100
27095 139 244.4 1.6 1.1 100
28016 5585 234.5 1.5 0.2 99
28044 246 208.3 1.3 2.0 100
28092 380 234.2 1.5 0.2 100
29029 545 237.0 2.2 0.4 100
29059 458 225.1 1.5 0.6 100
29072 385 231.7 1.9 0.0 100
30022 496 216.2 1.7 1.3 99
31041 361 228.9 1.5 0.2 100
31079 200 223.5 1.4 0.7 100
31094 292 224.6 1.5 0.6 100
31110 195 250.3 1.4 1.6 100
31119 327 219.0 1.6 1.1 100
31141 261 235.1 1.3 0.3 100
31152 5522 232.2 2.2 0.0 100
32014 474 259.2 2.2 2.3 100
32026 544 211.6 1.4 1.7 100
32171 251 232.9 1.6 0.1 100
33011 159 227.8 1.5 0.3 100
33051 202.5 1.3 2.5 100
33099 175 237.6 1.9 0.5 100
34175 119 245.5 1.6 1.2 100
35029 4676 226.2 1.6 0.5 99
35046 257 224.4 1.5 0.6 100
35066 67 218.8 1.4 1.1 100
35074 326 228.0 1.4 0.3 100
35172 237 252.6 2.1 1.8 100
35235 125 229.1 1.4 0.2 100
36036 65 237.1 1.4 0.5 100
36182 215 256.4 1.7 2.1 100
37247 238 251.7 1.3 1.7 100
37329 351 246.9 1.6 1.3 100
38011 253 226.9 1.6 0.4 100
38303 293 229.1 1.5 0.2 100
39013 484 229.1 1.3 0.2 99
39026 287 219.9 1.3 1.0 100
39034 334 225.4 1.5 0.5 100
39037 94 218.2 1.4 1.2 100
39044 258 239.8 1.4 0.7 100
39067 86 224.1 1.5 0.7 99
39088 304 211.1 1.5 1.8 100
39352 97 238.7 1.3 0.6 100
39401 345 221.2 1.3 0.9 99
40135 78 233.2 1.5 0.1 100
40371 8091 237.4 1.6 0.5 100
40479 4369 234.3 1.9 0.2 99
41025 4159 236.3 1.8 0.4 100
41035 233 227.0 1.5 0.4 100
41435 202 229.8 1.2 0.2 100
41455 235.4 1.6 0.3 100
41476 179 234.5 1.9 0.2 100
41494 4339 237.4 1.6 0.5 100
42023 170 226.3 1.3 0.5 100
42084 259 225.4 1.5 0.5 100
42250 8006 258.9 1.6 2.3 100
42501 305 247.6 1.6 1.4 100
43010 591 240.1 2.4 0.7 98
43024 3911 226.4 1.6 0.5 100
43061 357 232.4 2.0 0.1 100
43095 116 222.2 1.2 0.8 100
43108 236.5 2.1 0.4 100
43111 210.6 1.5 1.8 100
43134 221.5 2.2 0.9 99
43412 88 245.0 1.5 1.1 100
43485 265 238.1 1.4 0.5 100
44038 3929 217.9 1.5 1.2 100
44065 350 207.0 1.5 2.1 100
44115 64 241.8 1.5 0.9 100
44148 9173 240.2 2.0 0.7 100
44449 100 215.5 1.9 1.4 100
45082 318 216.3 1.5 1.3 100
45454 42 224.5 1.5 0.6 100
46024 40 213.3 1.4 1.6 100
46062 62 231.0 1.6 0.1 100
46381 329 256.1 1.5 2.1 100
47012 155 232.5 1.3 0.1 99
47420 530 232.2 1.3 0.0 100
48028 193 236.3 1.4 0.4 100
48049 76 222.7 1.4 0.8 100
48060 52 217.1 1.3 1.3 100
48083 191 232.8 1.8 0.1 100
48099 300 234.9 2.2 0.3 100
48392 120 257.9 1.4 2.2 100
49013 312 218.3 1.2 1.1 99
49022 430 224.5 1.8 0.6 96
49148 217.7 1.5 1.2 100
49238 254.8 1.5 2.0 100
49333 3292 251.8 1.8 1.7 100
50046 588 224.4 1.6 0.6 95
50253 79 227.5 1.3 0.4 100
51021 171 213.4 1.5 1.6 100
51074 372 226.1 1.8 0.5 100
51091 198 240.1 2.4 0.7 100
51254 212.3 1.5 1.7 100
51257 602 245.3 2.8 1.2 100
51259 423 233.7 1.6 0.2 100
52017 66 223.0 1.5 0.8 100
52035 215.9 1.5 1.4 100
52180 441 227.4 1.4 0.4 100
52222 245.2 1.6 1.1 100
53012 483 233.4 1.9 0.1 100
53054 599 224.1 1.7 0.7 100
53067 163 242.5 1.4 0.9 100
53114 138 223.2 1.5 0.7 100
53185 124 249.9 1.7 1.6 100
54018 2588 242.5 1.8 0.9 100
54063 555 231.2 1.6 0.1 100
55029 339 224.4 1.5 0.6 100
55063 177 224.3 1.8 0.6 100
55071 248 230.2 1.9 0.1 100
55114 132 217.2 1.8 1.2 100
55121 135 232.0 2.1 0.0 100
55149 505 236.1 1.7 0.4 99
56024 378 221.2 1.4 0.9 100
56087 81 241.0 1.4 0.8 100
57010 207 225.3 1.4 0.5 99
57054 110 225.7 1.5 0.5 100
57073 368 232.0 2.4 0.0 100
58043 531 220.8 1.9 0.9 100
58087 133 245.7 1.6 1.2 100
59024 232.2 1.9 0.0 98
59036 289 260.6 1.2 2.5 100
59047 192 245.4 1.3 1.2 100
59085 183 237.4 1.6 0.5 100
60065 288 213.6 1.4 1.6 100
60101 446 248.3 1.7 1.4 100
61015 53 237.5 1.6 0.5 99
61026 2042 238.4 1.9 0.6 100
61085 158 248.1 1.9 1.4 100
62058 407 234.4 1.7 0.2 100
63027 1898 233.5 2.1 0.1 100
63052 461 240.1 1.4 0.7 100
64049 181 221.6 1.2 0.9 100
64067 269 225.8 1.5 0.5 99
65057 1802 242.3 2.0 0.9 100
66047 472 245.1 1.3 1.1 100
67063 199 241.1 1.3 0.8 99
69012 109 250.0 1.6 1.6 99
70035 595 231.1 1.6 0.1 100
70049 389 234.8 1.5 0.3 98
73025 150 233.1 1.7 0.1 99
75024 1308 249.6 1.9 1.5 100
76027 297 223.8 1.5 0.7 100
77025 194 216.1 1.4 1.3 99
82024 1092 233.4 1.5 0.1 99
85027 264 236.8 1.3 0.4 99
Cluster Mean Values
231.8 1.6 0.8
Median 232.0 1.5 0.6
11.7 0.3 0.6
Table 2: Radial Velocity and Membership Information
StaraaIdentifier from van Leeuwen et al. (2000). Alt. IDbbIdentifier from Woolley (1966). V B–V V–K M T log g [Fe/H] V
TCS [K] [cm s] Spec. [km s]
9 370 12.529 1.250 2.870 1.543 4460 1.20 1.26 1.95
5009 548 12.912 1.080 2.841 1.160 4525 1.40 1.90 1.60
6017 240 12.233 1.420 3.387 1.839 4110 0.85 1.36 1.85
10012 43 11.529 1.618 3.782 2.543 3900 0.40 1.49 2.10
11019 537 12.841 1.223 2.985 1.231 4450 1.30 1.57 2.00
11024 91 11.738 1.333 3.291 2.334 4200 0.70 1.76 1.90
12013 394 12.579 1.319 3.142 1.493 4275 1.10 1.50 2.05
14010 435 12.807 0.993 2.647 1.265 4635 1.45 1.74 1.40
15022 180 11.982 1.243 2.997 2.090 4400 0.95 1.79 1.95
16009 252 12.232 1.201 3.081 1.840 4375 1.00 1.88 2.10
16015 213 12.127 1.122 2.885 1.945 4475 1.05 1.93 1.90
17015 325 12.430 1.156 2.901 1.642 4475 1.15 1.77 1.35
17032 605 12.989 1.150 2.852 1.083 4475 1.40 1.74 2.05
18047 408 12.570 1.104 2.841 1.502 4525 1.25 1.53 1.30
19062 464 12.803 1.144 2.899 1.269 4500 1.30 1.74 1.50
20049 6355 13.273 1.058 2.778 0.799 4650 1.55 1.70 1.60
21032 172 11.947 1.394 3.290 2.125 4100 0.75 1.63 1.95
21042 348 12.494 1.179 2.887 1.578 4525 1.20 1.52 1.70
23061 296 12.337 1.188 2.915 1.735 4650 1.10 1.34 1.85
24013 56 11.596 1.589 3.753 2.476 3915 0.40 1.75 2.70
24027 5969 13.013 1.099 2.757 1.059 4575 1.45 1.44 1.50
24040 5993 13.129 0.952 2.497 0.943 4850 1.65 1.48 1.55
24046 74 11.657 1.367 3.215 2.415 4200 0.70 1.88 2.20
24056 364 12.474 1.145 2.858 1.598 4460 1.20 1.74 1.50
24062 352 12.628 1.307 3.124 1.444 4350 1.15 1.40 1.50
25018 5964 13.904 0.916 2.463 0.168 4850 2.00 1.47 1.20
25043 89 11.734 1.500 3.438 2.338 4100 0.60 1.49 2.05
25062 46 11.583 1.545 3.704 2.489 3950 0.45 1.83 2.45
25065 12.101 1.689 3.883 1.971 3875 0.55 1.07 2.25
25068 58 11.542 1.434 3.329 2.530 4350 0.60 1.51 2.05
26025 61 11.411 1.591 3.615 2.661 3975 0.40 1.68 2.20
26088 161 11.895 1.379 3.243 2.177 4185 0.80 1.64 1.85
27048 313 12.442 1.241 2.941 1.630 4400 1.15 1.66 2.00
27055 5837 13.824 1.060 2.621 0.248 4700 1.85 0.98 1.50
27095 139 11.817 1.452 3.310 2.255 4145 0.70 1.39 2.05
28016 5585 13.177 1.038 2.776 0.895 4535 1.50 1.65 1.60
28044 246 12.323 1.169 3.004 1.749 4450 1.05 1.50 1.55
28092 380 12.521 1.207 2.966 1.551 4475 1.15 1.41 1.40
29029 545 12.911 1.139 2.812 1.161 4510 1.40 1.51 1.50
29059 458 12.820 1.140 2.919 1.252 4415 1.30 1.64 1.60
29072 385 12.665 1.119 2.880 1.407 4600 1.25 1.45 1.30
30022 496 12.793 0.998 2.668 1.279 4575 1.40 1.71 1.55
31041 361 12.596 1.091 2.927 1.476 4600 1.20 1.32 1.40
31079 200 12.151 1.202 2.908 1.921 4415 1.05 1.72 1.60
31094 292 12.405 1.130 2.804 1.667 4500 1.20 1.78 1.70
31110 195 12.242 1.354 3.359 1.830 4300 0.85 1.08 1.70
31119 327 12.586 1.234 3.242 1.486 4300 1.05 1.36 2.15
31141 261 12.368 1.159 3.051 1.704 4350 1.05 1.60 1.60
31152 5522 13.195 1.046 2.723 0.877 4550 1.55 1.71 1.50
32014 474 12.809 1.042 2.731 1.263 4560 1.40 1.69 1.40
32026 544 12.978 1.083 2.821 1.094 4500 1.40 1.40 1.70
32171 251 12.189 1.383 3.103 1.883 4285 0.95 1.32 1.80
33011 159 11.879 1.337 3.130 2.193 4300 0.80 1.65 2.00
33051 11.979 1.213 2.953 2.093 4375 0.95 1.66 1.55
33099 175 12.100 1.483 3.371 1.972 4200 0.80 0.97 2.15
34175 119 11.994 1.430 3.272 2.078 4200 0.80 1.55 2.10
35029 4676 13.264 1.015 2.720 0.808 4800 1.60 1.23 1.30
35046 257 12.398 1.091 3.017 1.674 4450 1.10 1.69 1.70
35066 67 11.444 1.486 3.407 2.628 4080 0.50 1.78 1.90
35074 326 12.627 1.120 2.869 1.445 4550 1.25 1.65 1.65
35172 237 12.414 1.399 3.250 1.658 4285 1.00 0.91 1.95
35235 125 11.693 1.393 3.290 2.379 4300 0.65 1.57 1.95
36036 65 11.425 1.498 3.453 2.647 4050 0.50 2.05 2.45
36182 215 12.352 1.302 3.218 1.720 4350 0.95 1.51 1.75
37247 238 12.430 1.163 3.033 1.642 4500 1.10 1.63 1.75
37329 351 12.458 1.188 2.921 1.614 4410 1.15 1.54 1.75
38011 253 12.217 1.365 3.090 1.855 4275 1.00 1.25 1.90
38303 293 12.476 1.238 3.077 1.596 4350 1.10 1.37 1.65
39013 484 12.755 1.195 2.789 1.317 4650 1.35 1.41 1.65
39026 287 12.333 1.373 3.238 1.739 4400 0.95 1.19 1.90
39034 334 12.513 1.087 2.946 1.559 4450 1.15 1.61 1.45
39037 94 11.629 1.393 3.273 2.443 4175 0.65 1.92 2.00
39044 258 12.263 1.157 3.045 1.809 4310 1.00 1.91 1.80
39067 86 11.545 1.480 3.126 2.527 4250 0.70 1.38 2.00
39088 304 12.324 1.214 2.945 1.748 4600 1.10 1.46 1.85
39352 97 11.740 1.380 3.202 2.332 4400 0.75 1.56 2.00
39401 345 12.625 1.186 3.048 1.447 4400 1.15 1.63 2.20
40135 78 11.773 1.353 3.315 2.299 4135 0.70 1.90 1.80
40371 8091 12.324 1.324 3.465 1.748 4060 0.85 1.50 1.85
40479 4369 13.063 1.105 2.917 1.009 4600 1.40 1.43 1.60
41025 4159 13.059 0.996 2.672 1.013 4750 1.55 1.49 1.70
41035 233 12.141 1.219 3.052 1.931 4225 0.95 1.85 1.60
41435 202 12.331 1.240 3.294 1.741 4200 0.95 1.44 1.75
41455 11.566 1.558 3.627 2.506 3975 0.45 1.29 2.50
41476 179 12.031 1.651 4.027 2.041 3885 0.30 1.00 2.40
41494 4339 13.328 1.080 2.844 0.744 4485 1.55 1.27 1.50
42023 170 11.949 1.275 3.114 2.123 4265 0.85 1.87 1.80
42084 259 12.236 1.297 3.037 1.836 4325 1.00 1.73 2.05
42250 8006 12.226 1.190 3.184 1.846 4325 0.95 1.78 1.65
42501 305 12.512 1.242 3.051 1.560 4425 1.10 1.64 1.85
43010 591 13.009 1.042 2.770 1.063 4535 1.45 1.90 1.75
43024 3911 13.133 1.016 2.745 0.939 4550 1.50 1.81 1.65
43061 357 12.602 1.431 3.781 1.470 4000 0.85 0.72 2.00
43095 116 11.997 1.232 3.168 2.075 4325 0.85 1.79 1.80
43108 13.033 1.024 2.737 1.039 4475 1.50 1.61 1.20
43111 12.918 1.065 2.742 1.154 4475 1.45 1.69 1.70
43134 12.755 1.106 2.796 1.317 4510 1.35 1.87 0.90
43412 88 11.740 1.436 3.297 2.332 4175 0.70 1.87 1.95
43485 265 12.520 1.183 3.005 1.552 4350 1.15 1.81 1.75
44038 3929 13.239 0.975 2.649 0.833 4400 1.60 1.77 1.10
44065 350 12.434 1.132 2.897 1.638 4475 1.15 1.72 1.70
44115 64 11.632 1.464 3.366 2.440 4200 0.60 1.66 2.00
44148 9173 13.048 1.107 3.025 1.024 4600 1.35 1.00 1.35
44449 100 11.789 1.584 3.483 2.283 4140 0.65 1.02 2.10
45082 318 12.606 1.082 2.904 1.466 4450 1.20 1.81 1.60
45454 42 11.644 1.495 3.559 2.428 4100 0.55 1.77 2.10
46024 40 11.291 1.479 3.301 2.781 4140 0.50 1.69 1.80
46062 62 11.494 1.595 3.757 2.578 3900 0.40 1.88 2.40
46381 329 12.607 1.192 3.171 1.465 4235 1.10 1.62 1.80
47012 155 11.890 1.415 3.266 2.182 4225 0.75 1.71 2.10
47420 530 12.969 1.094 2.873 1.103 4455 1.40 1.56 1.45
48028 193 12.062 1.244 3.074 2.010 4375 0.90 1.68 1.60
48049 76 11.525 1.511 3.574 2.547 4000 0.45 1.76 1.80
48060 52 11.316 1.622 3.515 2.756 4000 0.40 1.97 2.50
48083 191 12.044 1.376 3.116 2.028 4375 0.90 1.36 1.80
48099 300 12.443 1.684 3.955 1.629 3915 0.70 1.04 2.10
48392 120 11.802 1.391 3.283 2.270 4200 0.70 1.66 1.55
49013 312 12.325 1.299 3.060 1.747 4450 1.05 1.44 2.05
49022 430 12.726 1.076 2.874 1.346 4525 1.30 1.71 1.40
49148 14.252 0.428 1.242 0.180 4100 0.75 1.61 1.95
49238 12.421 1.194 3.053 1.651 4450 1.10 1.59 1.45
49333 3292 13.044 1.049 2.784 1.028 4650 1.45 1.32 1.60
50046 588 13.195 1.032 2.720 0.877 4500 1.55 1.93 1.80
50253 79 11.658 1.375 3.247 2.414 4100 0.70 1.71 1.70
51021 171 11.984 1.470 3.521 2.088 4030 0.70 1.44 2.00
51074 372 12.706 1.300 3.331 1.366 4275 1.10 0.75 2.00
51091 198 12.320 1.143 2.935 1.752 4400 1.10 2.17 1.50
51254 12.402 1.316 3.079 1.670 4300 1.05 1.41 2.00
51257 602 12.955 1.022 2.694 1.117 4600 1.45 1.58 1.45
51259 423 12.597 1.135 2.810 1.475 4500 1.25 1.48 1.55
52017 66 11.435 1.616 3.666 2.637 3975 0.40 1.86 2.35
52035 11.498 1.574 3.333 2.574 4250 0.55 1.71 2.50
52180 441 12.733 1.169 2.865 1.339 4575 1.30 1.37 1.55
52222 12.447 1.097 2.805 1.625 4525 1.20 1.45 1.60
53012 483 12.742 1.063 2.735 1.330 4600 1.35 1.27 0.90
53054 599 12.981 1.076 2.735 1.091 4650 1.45 1.63 1.50
53067 163 11.941 1.303 3.163 2.131 4400 0.85 1.53 1.70
53114 138 12.037 1.390 3.504 2.035 4035 0.70 1.70 2.20
53185 124 11.776 1.380 3.344 2.296 4275 0.70 1.69 2.05
54018 2588 13.475 1.042 2.798 0.597 4450 1.60 1.80 1.80
54063 555 12.989 1.110 2.836 1.083 4485 1.40 1.47 1.50
55029 339 12.387 1.356 3.108 1.685 4315 1.05 1.39 2.00
55063 177 11.955 1.414 3.195 2.117 4175 0.80 1.32 1.90
55071 248 12.150 1.695 3.922 1.922 3825 0.55 0.91 2.40
55114 132 11.654 1.705 3.876 2.418 3875 0.40 1.64 2.50
55121 135 11.957 1.673 3.630 2.115 4060 0.65 0.90 2.30
55149 505 12.894 1.321 3.151 1.178 4265 1.20 1.00 2.20
56024 378 12.716 1.158 3.143 1.356 4300 1.15 1.37 1.45
56087 81 11.404 1.543 3.473 2.668 4050 0.45 1.92 2.30
57010 207 12.154 1.412 3.254 1.918 4185 0.90 1.36 1.80
57054 110 11.589 1.595 3.560 2.483 4150 0.50 1.49 2.40
57073 368 12.470 1.151 2.831 1.602 4480 1.20 1.82 1.55
58043 531 12.923 1.079 2.728 1.149 4650 1.45 1.76 1.40
58087 133 11.760 1.337 3.117 2.312 4350 0.80 1.62 1.75
59024 11.855 1.629 3.672 2.217 3900 0.50 0.72 2.40
59036 289 12.396 1.182 2.931 1.676 4500 1.15 1.56 1.80
59047 192 11.975 1.413 3.146 2.097 4250 0.85 1.54 1.95
59085 183 11.918 1.328 3.134 2.154 4275 0.85 1.79 1.95
60065 288 12.335 1.226 2.910 1.737 4450 1.10 1.79 2.05
60101 446 12.680 1.092 2.819 1.392 4490 1.30 1.48 1.70
61015 53 11.503 1.644 3.768 2.569 4000 0.40 1.66 2.45
61026 2042 12.994 1.070 2.724 1.078 4750 1.50 1.30 1.30
61085 158 11.846 1.699 3.724 2.226 4050 0.55 1.15 2.20
62058 407 12.564 1.280 2.906 1.508 4475 1.20 1.15 1.85
63027 1898 13.024 1.058 2.685 1.048 4550 1.50 1.82 1.95
63052 461 12.709 1.235 2.840 1.363 4485 1.30 1.40 1.60
64049 181 12.015 1.346 3.029 2.057 4330 0.95 1.74 1.85
64067 269 12.259 1.223 2.981 1.813 4550 1.05 1.21 1.55
65057 1802 13.476 1.042 2.770 0.596 4700 1.65 1.25 1.40
66047 472 12.704 1.351 3.042 1.368 4375 1.20 1.24 2.10
67063 199 12.084 1.348 3.175 1.988 4400 0.90 1.36 1.75
69012 109 11.666 1.390 3.216 2.406 4275 0.70 1.85 2.15
70035 595 12.969 1.229 2.901 1.103 4400 1.40 1.35 1.65
70049 389 12.621 1.137 2.938 1.451 4400 1.20 1.38 1.85
73025 150 11.864 1.685 3.657 2.208 3970 0.55 1.27 2.30
75024 1308 13.659 1.122 2.786 0.413 4550 1.70 1.08 1.65
76027 297 12.366 1.280 3.083 1.706 4400 1.05 1.47 1.70
77025 194 12.197 1.339 3.154 1.875 4250 0.95 1.71 1.90
82024 1092 13.648 1.084 2.666 0.424 4650 1.75 1.18 1.70
85027 264 12.370 1.260 3.119 1.702 4270 1.05 1.65 1.80
Table 3: Photometry and Model Atmosphere Parameters
Element Exc. Pot. log gf
[Å] [eV]
Fe II 6516.08 2.89 3.45
Fe I 6533.93 4.56 1.36
Fe I 6546.24 2.76 1.54
Fe I 6551.68 0.99 5.77
Fe I 6574.25 0.99 5.02
Fe I 6592.92 2.73 1.47
Fe I 6593.88 2.43 2.42
Fe I 6597.57 4.79 0.95
Fe I 6608.04 2.28 3.96
Fe I 6609.12 2.56 2.69
Fe I 6625.02 1.01 5.37
Fe I 6627.54 4.55 1.58
Fe I 6633.75 4.79 0.80
Fe I 6646.96 2.61 3.96
Fe I 6648.12 1.01 5.92
Fe I 6677.99 2.69 1.35
Al I 6696.03 3.14 1.57
Al I 6698.66 3.14 1.89
Fe I 6703.57 2.76 3.01
Fe I 6710.32 1.48 4.83
Fe I 6726.67 4.61 1.07
Fe I 6733.15 4.64 1.48
Fe I 6739.52 1.56 4.79
Table 4: Line list
ccWavelengths are given in units of Å. 6516 6533 6546 6551 6574 6592 6593 6597 6608 6609 6625 6627 6633 6646 6648 6677 6696 6698 6703 6710 6726 6733 6739
9ddDesignation is from van Leeuwen et al. (2000). 63 20 151 159 130 34 40 100 51 14 15 36 170 23 28 49
5009 9 92 6 25 115 62 11 31 14 8 105 9 18 10 6 7
6017 41 30 140 108 179 134 25 52 102 105 16 39 82 202 Syn Syn 91 81 31 25 49
10012 54 21 179 62 133 186 157 27 77 145 116 20 73 210 Syn Syn 83 95 35
11019 13 139 18 68 141 118 17 17 82 24 8 15 141 22 18 22 24 33
11024 62 11 143 24 82 146 120 18 43 94 49 7 20 27 158 11 38 42 18 9 35
12013 18 149 31 89 165 145 22 37 74 17 161 Syn Syn 61 64 9
14010 80 24 86 67 14 42 13 109 19 10
15022 48 5 119 53 121 101 19 26 12 142 13 19 16 14
16009 130 55 132 99 26 5 10 133 10 17 13 15
16015 46 107 7 24 110 80 12 56 24 3 112 4 9 11 16
17015 33 96 14 30 107 73 14 7 116 11 19 12 5 15
17032 38 109 10 47 132 108 15 66 7 142 51 36 29 15 9 8 23
18047 44 13 90 21 36 110 91 22 27 66 22 9 19 8 12 122 36 18 25 14 19
19062 7 97 8 36 107 79 19 69 34 7 116 17 8 16 15 9 9 13
20049 45 90 26 96 70 11 39 15 6 122 Syn Syn 13
21032 41 13 155 44 114 156 130 20 50 114 77 22 45 176 15 18 68 55 14 54
21042 46 119 14 50 125 98 24 62 21 13 140 8 22 40 19
23061 56 19 122 12 50 131 97 17 35 73 26 7 33 130 22 9 16
24013 39 187 51 154 210 171 18 62 124 11 26 59 216 Syn Syn 77 88 68
24027 53 104 45 122 89 15 17 76 30 8 15 128 Syn Syn 27 18 22 27
24040 89 18 92 80 8 96 5 13 15 7
24046 45 136 17 82 145 121 12 35 87 52 6 13 28 169 Syn Syn 46 30 10 27
24056 40 107 14 36 112 76 25 68 19 6 7 120 15 12 23 18 5 16
24062 39 17 124 33 126 111 26 35 31 24 149 22 12 30 37
25018 36 85 14 82 53 13 18 102 13 14
25043 46 23 158 45 121 175 147 20 47 118 98 16 37 60 195 Syn Syn 74 84 29 87
25062 46 12 170 45 122 181 154 61 130 97 21 50 196 13 6 60 78 31 58
25065 32 43 207 90 173 216 181 38 97 159 34 53 105 234 Syn Syn 116 134 72 24 110
25068 161 22 87 142 121 26 50 101 61 9 24 162 Syn Syn 46 37 28 34
26025 35 171 48 130 188 151 24 66 130 97 20 64 198 39 26 62 73 24 14 54
26088 50 13 145 101 155 122 18 61 96 61 12 42 158 18 16 44 41 22 10
27048 126 16 67 136 109 10 22 76 40 8 10 151 Syn Syn 45 39 8 22
27055 138 95 81 25 157 Syn Syn 35 50
27095 36 30 161 108 188 142 26 62 124 81 18 210 Syn Syn 87 40 13
28016 14 104 11 42 103 84 9 56 139 18 28 20 15
28044 36 109 15 56 127 97 15 30 92 36 15 13 22 137 40 30 18 23
28092 51 21 110 23 56 129 107 20 22 35 21 140 18 26 26 31
29029 32 111 15 45 120 95 14 36 68 15 16 14 17 125 24 14 22 36 21 7 12
29059 46 13 119 69 122 88 18 34 7 138 39 16 27 28
29072 42 13 100 11 37 105 80 20 65 30 23 9 110 12 12 16
30022 43 7 100 13 27 115 72 8 12 7 106 15 16 17
31041 40 24 94 25 47 116 107 22 26 59 34 12 143 32 26 22 25 26
31079 46 18 114 17 43 113 86 9 24 80 12 123 13 7 30 22 8 7 13
31094 43 6 110 27 116 87 16 52 14 7 10 117 23 13 14 5
31110 48 34 166 105 178 144 67 126 85 25 32 51 179 Syn Syn 81 64 53 25 67
31119 159 31 101 167 146 28 131 72 16 30 197 62 59 21 70
31141 51 14 135 56 131 109 25 79 38 10 12 28 140 51 29 37 37 15 22
31152 38 100 106 79 53 110 11 16
32014 100 28 111 73 11 14 39 14 14 109 21 20 10 6
32026 40 15 123 22 62 130 109 17 25 90 40 20 29 148 18 11 31 29 23
32171 57 149 107 157 132 46 114 55 23 52 181 Syn Syn 51 67 38 21
33011 52 14 139 17 90 149 111 13 99 50 11 28 171 15 4 46 38 22 31
33051 13 115 59 120 101 12 37 9 15 132 56 31 41 40 15 22
33099 83 161 220 166 61 135 133 27 49 242 Syn Syn 137 75 39 75
34175 54 13 157 37 101 178 139 28 114 64 17 43 185 12 17 55 64 26
35029 46 20 103 25 118 70 22 57 30 15 102 18 15
35046 120 11 41 112 90 27 73 8 12 134 6 26 25 12 8 19
35066 34 16 137 40 102 153 126 19 33 104 57 10 25 41 167 11 53 70 13 37
35074 10 103 9 43 108 84 7 19 64 26 6 127 25 23 20 14 5 13
35172 63 169 122 186 141 51 132 116 31 50 94 236 Syn Syn 111 116 52 30
35235 48 21 139 33 73 157 123 45 106 47 11 31 166 37 26 45 22 7 32
36036 42 9 155 24 87 159 131 10 39 103 60 9 22 170 11 5 41 46 12 31
36182 35 12 151 17 89 131 104 25 82 64 16 11 166 Syn Syn 56 48 29 8 31
37247 52 6 125 11 49 115 84 9 30 66 20 11 26 19 131 34 17 26 10 8 13
37329 127 17 67 135 99 16 36 87 38 22 150 43 31
38011 55 35 159 112 164 134 28 111 90 18 19 191 Syn Syn 86 83 21
38303 26 67 131 119 13 44 91 63 15 43 21 44 162 Syn Syn 56 53 15 26
39013 45 14 114 34 127 91 19 24 9 12 132 15 6 27 23 13 16
39026 57 154 38 106 163 131 24 51 119 74 21 38 32 187 Syn Syn 58 57 37 24
39034 17 101 23 58 123 95 21 123 23 23 27 22 28
39037 48 137 25 77 141 110 26 90 46 6 17 158 9 6 36 32 6 35
39044 44 12 116 59 126 83 18 69 30 5 12 9 130 24 23 13 3 14
39067 66 22 156 34 112 165 135 21 51 122 82 19 29 57 197 Syn Syn 65 69 21 68
39088 46 10 114 12 39 139 93 81 22 12 9 133 49 28 26 22 23 9 14
39352 53 142 17 76 135 114 17 42 99 41 11 23 158 40 21 37 33 17 18
39401 55 138 72 140 125 102 35 10 157 32 43 37 23
40135 45 11 131 27 77 136 113 35 83 45 6 10 22 149 13 5 42 39
40371 19 144 35 111 174 139 18 61 109 97 15 27 64 176 Syn Syn 63 85 22 61
40479 113 14 41 119 101 27 65 26 11 14 134 14 9 25 11 7 19
41025 62 95 32 112 80 50 117 13 11 18 15 14 7 10
41035 30 9 118 72 125 107 14 21 77 37 13 21 134 7 28 34 12
41435 46 17 148 95 165 121 17 52 105 64 23 37 20 45 174 Syn Syn 60 68 18 61
41455 38 216 68 171 220 193 42 78 151 150 23 43 252 Syn Syn 113 126 86
41476 53 35 200 100 159 186 196 45 105 145 158 14 27 82 203 Syn Syn 92 90 53 30 88
41494 141 35 145 95 28 39 82 134 20 20 37 23
42023 34 7 117 21 66 124 101 14 33 82 34 18 142 7 30 23 8 7
42084 54 10 132 57 139 105 15 23 88 49 165 Syn Syn 42 47 6 24
42250 126 11 45 125 88 14 38 71 24 11 17 126 28 17 38 32
42501 60 9 126 18 56 142 98 87 37 131 14 10 32 28 16 10 26
43010 34 5 92 32 112 7 107 8 12 10 10 11
43024 87 24 106 74 8 8 57 18 113 39 19 19
43061 63 193 84 183 224 166 52 102 162 53 69 117 229 Syn Syn 115 151 85 40 97
43095 59 10 121 15 59 132 98 78 37 146 49 13 27 29 14 6
43108 95 41 109 83 12 56 6 9 14 117 45 25 26 16
43111 40 9 106 14 41 116 93 71 31 11 15 139 21 19 11 12
43134 29 67 23 81 51 8 57 10 85 23 23 5 16
43412 38 9 136 23 73 144 116 36 84 53 6 17 150 20 9 40 42 18 27
43485 49 9 114 50 116 106 12 33 15 143 12 34 22 17
44038 26 88 15 91 74 17 56 33 113 23 16 29 12 5
44065 39 13 110 10 50 112 78 12 19 63 14 5 11 132 6 7 24 14
44115 43 20 157 36 91 156 123 18 42 107 61 9 21 37 173 20 9 47 51 20 34
44148 122 30 68 135 106 81 62 27 18 161 Syn Syn 41 18 32
44449 78 74 180 131 214 163 34 55 143 130 45 112 248 Syn Syn 113 119 38 89
45082 11 111 44 102 80 62 21 7 120 12 4 13 4 20
45454 42 154 96 162 124 35 105 76 16 40 179 60 42 66 56 40
46024 44 15 142 35 96 157 120 13 50 61 11 20 39 8 10 48 53 24
46062 48 11 167 40 124 176 147 12 53 131 93 26 52 195 53 32 52 60 10 48
46381 44 18 148 73 145 108 100 58 15 37 160 70 51 45 54 17 38
47012 49 143 29 81 160 130 11 35 105 60 13 29 175 19 43 22 30
47420 36 14 113 19 57 129 84 26 79 29 10 11 18 123 56 19 31 21 7 18
48028 41 13 115 23 57 120 98 21 83 31 7 12 12 137 31 19 29 32 15 24
48049 38 18 146 44 103 150 127 13 45 113 86 24 46 161 24 15 60 61 22 52
48060 55 8 161 34 113 186 130 35 108 76 10 15 41 198 34 24 47 57 15 5 41
48083 63 29 140 32 83 156 109 22 26 104 66 19 175 Syn Syn 52 66 37 14 37
48099 39 59 218 93 175 194 163 94 138 22 68 88 Syn Syn 108 114 54 34
48392 28 81 135 108 35 83 52 22 38 167 57 32 45 39 21
49013 39 20 140 21 73 147 119 21 39 106 33 13 31 157 25 14 40 36 27 7 33
49022 36 7 93 29 107 80 14 61 17 13 108 16 13 20
49148 44 23 153 36 114 160 124 17 42 121 84 30 49 181 20 25 54 70 27 11 52
49238 11 115 19 55 127 90 16 53 26 8 16 124 21 28 23
49333 126 44 135 94 22 11 13 11 123 12 12 17 27
50046 88 28 111 84 7 56 21 101 17 21 4 9
50253 30 17 131 90 154 113 33 98 68 21 40 156 73 60 60 55 17
51021 57 26 169 56 121 169 142 21 68 139 112 17 29 63 203 19 19 81 77 32 19 72
51074 71 198 135 205 164 77 146 138 25 60 236 Syn Syn 98 107 79 42
51091 44 82 42 74 58 37 9 2 95 13 19 12 9 11
51254 59 152 92 164 124 21 33 120 74 14 23 187 Syn Syn 59 70 14 51
51257 111 109 67 39 8 6 114 14 24
51259 17 121 22 122 98 23 68 29 9 19 143 37 36 26 26
52017 42 13 157 46 123 179 148 19 50 115 85 19 47 199 10 56 65 49
52035 49 9 158 18 82 169 153 110 49 18 35 191 72 48 34 52 12
52180 28 126 54 129 85 28 83 26 12 19 127 30 25 23 8 23
52222 18 136 20 122 86 29 77 126 27 14 16 20
53012 13 100 99 67 31 60 22 17 15 114 Syn Syn 32 39 22
53054 10 92 6 26 98 69 11 8 58 21 112 12 21 12
53067 48 130 65 135 102 17 95 47 12 14 145 23 37 36 18 12 25
53114 45 11 161 36 110 168 136 20 54 120 90 10 20 61 188 25 19 61 70 30 18 42
53185 38 12 152 26 90 151 117 36 94 53 167 34 36 10
54018 40 98 16 44 112 82 10 75 25 143 60 24 29 18 11 14
54063 34 116 46 124 103 27 65 15 146 Syn Syn 30
55029 48 30 147 25 100 164 126 18 53 112 76 13 18 42 186 Syn Syn 52 51 34 21 47
55063 41 160 51 117 161 153 32 58 117 85 17 28 Syn Syn 78 92 46 80
55071 15 63 222 189 243 194 101 160 32 64 126 260 Syn Syn 127 132 31 143
55114 47 193 68 155 207 169 27 81 139 11 29 72 208 38 21 81 90 33 11 80
55121 198 181 239 188 73 66 170 165 40 69 143 251 Syn Syn 117 140 38 93
55149 55 52 185 140 208 161 30 61 146 102 37 46 220 Syn Syn 100 48 29 77
56024 55 125 78 140 110 28 35 95 65 18 41 168 Syn Syn 50 48 15
56087 37 158 38 99 170 133 46 99 67 6 12 37 177 13 8 48 49 8 39
57010 39 25 159 40 106 164 141 25 50 121 96 20 29 54 173 43 31 66 70 39 55
57054 32 174 43 130 170 171 33 57 145 123 17 19 50 216 56 24 64 74 15 54
57073 42 11 96 29 97 78 5 3 11 128 7 21 16 15 4 13
58043 80 19 84 11 46 10 3 102 8 10 12
58087 62 11 132 60 139 112 12 31 81 40 13 13 20 147 9 8 31 35
59024 228 217 264 210 102 188 35 74 152 280 Syn Syn 156 163 57
59036 41 127 54 126 93 20 67 32 6 10 147 48 31 31 10 19
59047 33 147 35 87 156 119 37 109 73 18 183 Syn Syn 74 61 13 38
59085 46 17 138 18 65 134 107 11 36 93 43 8 18 156 7 36 29 14 7 26
60065 34 124 11 48 122 99 19 58 24 136 13 21 18 16 4 18
60101 29 13 132 13 136 105 20 10 14 128 16 31 21
61015 50 179 53 140 195 167 24 68 143 104 12 25 53 207 Syn Syn 70 79 29 16 59
61026 41 17 99 40 104 76 18 25 50 118 35 12 15
61085 59 41 151 210 73 150 38 96 230 Syn Syn 100 106 57 30
62058 37 135 28 104 143 144 41 101 54 16 18 190 Syn Syn 23
63027 109 27 112 82 17 40 125 12 13 21 8
63052 46 123 65 131 101 21 35 16 30 152 34 30 7
64049 121 22 62 140 106 18 82 32 12 147 52 24 35 33
64067 128 31 68 135 106 34 92 33 22 24 153 36 29 16
65057 105 43 116 91 22 19 65 10 10 133 73 56 37 25
66047 60 25 169 27 98 170 144 37 63 113 75 22 30 41 186 32 12 77 55 29 18 42
67063 61 20 152 67 143 124 27 27 73 58 17 19 167 61 38 65 60
69012 33 146 17 74 137 105 32 86 43 9 12 13 151 14 8 35 35 12 4 19
70035 53 21 137 73 137 113 27 90 73 24 29 159 31 16 49 23 30
70049 50 140 33 71 147 123 37 96 10 41 41 171 Syn Syn 41 48 23 12 39
73025 50 190 70 161 214 173 39 72 128 22 45 100 238 Syn Syn 100 113 24 90
75024 37 30 132 28 81 146 38 21 172 27 45 14
76027 20 122 60 134 122 19 52 11 9 25 154 39 45 29 37
77025 27 136 74 134 117 38 93 64 9 14 153 16 16 48 45 32 5 28
82024 24 139 56 128 96 18 78 45 13 12 152 14 11 39 41
85027 51 13 142 67 136 121 14 26 80 58 20 38 162 24 40 43 24 29
Table 5: Equivalent Widthsa,ba,bfootnotemark:
StaraaThe designation “Syn” indicates a synthetic spectrum comparison method was used. log (Fe) [Fe/H]bbEquivalent widths are given in units of mÅ. Num. Lines / log (Al) [Al/Fe]ccAssumed the solar log (Al)=6.47 (Anders & Grevesse 1989). Num. Lines /
9 6.26 1.26 14 0.03 5.78 0.57 1
5009 5.62 1.90 15 0.05 5.39 0.82 1
6017 6.16 1.36 18 0.05 6.28 1.17 2 0.07
10012 6.03 1.49 16 0.03 5.35 0.37 2 0.02
11019 5.95 1.57 16 0.03 5.53 0.63 2 0.11
11024 5.76 1.76 19 0.03 4.94 0.23 1
12013 6.02 1.50 14 0.03 5.13 0.16 2 0.07
14010 5.78 1.74 10 0.03
15022 5.73 1.79 12 0.04 5.14 0.46 1
16009 5.64 1.88 11 0.03 5.04 0.45 1
16015 5.59 1.93 13 0.05 4.69 0.15 1
17015 5.75 1.77 12 0.03 5.09 0.39 1
17032 5.78 1.74 14 0.03 6.02 1.29 2 0.04
18047 5.99 1.53 19 0.04 5.81 0.87 1
19062 5.78 1.74 16 0.05 5.34 0.61 2 0.03
20049 5.82 1.70 10 0.04 5.15 0.38 2 0.07
21032 5.89 1.63 17 0.02 5.18 0.34 2 0.19
21042 6.00 1.52 13 0.04 5.33 0.38 1
23061 6.18 1.34 15 0.04 5.89 0.76 1
24013 5.77 1.75 15 0.03 5.60 0.88 2 0.07
24027 6.08 1.44 15 0.03 6.06 1.03 2 0.07
24040 6.04 1.48 9 0.06 4.99 0.00 1
24046 5.64 1.88 17 0.03 5.32 0.73 2 0.07
24056 5.78 1.74 15 0.03 5.36 0.63 2 0.08
24062 6.12 1.40 12 0.04 5.37 0.30 2 0.01
25018 6.05 1.47 8 0.06 5.44 0.44 1
25043 6.03 1.49 18 0.03 6.14 1.16 2 0.07
25062 5.69 1.83 18 0.03 4.79 0.15 2 0.02
25065 6.45 1.07 18 0.03 6.08 0.68 2 0.07
25068 6.01 1.51 16 0.04 5.51 0.55 2 0.07
26025 5.84 1.68 17 0.03 5.44 0.65 2 0.05
26088 5.88 1.64 16 0.04 5.25 0.42 2 0.13
27048 5.86 1.66 16 0.02 6.11 1.30 2 0.07
27055 6.54 0.98 7 0.08 6.28 0.79 2 0.07
27095 6.13 1.39 14 0.03 5.94 0.86 2 0.07
28016 5.87 1.65 12 0.04 5.70 0.88 1
28044 6.02 1.50 16 0.04 5.82 0.85 1
28092 6.11 1.41 14 0.03 5.34 0.28 1
29029 6.01 1.51 18 0.04 5.53 0.57 2 0.01
29059 5.88 1.64 11 0.03 5.68 0.85 1
29072 6.07 1.45 13 0.02 5.40 0.38 2 0.13
30022 5.81 1.71 13 0.04
31041 6.20 1.32 15 0.06 5.89 0.74 2 0.10
31079 5.80 1.72 16 0.05 5.16 0.41 2 0.02
31094 5.74 1.78 15 0.03 2 0.11
31110 6.44 1.08 17 0.02 5.51 0.12 2 0.07
31119 6.16 1.36 15 0.03
31141 5.92 1.60 16 0.04 5.91 1.04 2 0.04
31152 5.81 1.71 7 0.02
32014 5.83 1.69 13 0.04 5.50 0.72 1
32026 6.12 1.40 16 0.04 5.40 0.33 2 0.04
32171 6.20 1.32 14 0.04 6.06 0.91 2 0.07
33011 5.87 1.65 16 0.03 4.99 0.17 2 0.13
33051 5.86 1.66 14 0.02 6.00 1.19 2 0.05
33099 6.55 0.97 13 0.06 6.32 0.82 2 0.07
34175 5.97 1.55 15 0.03 5.18 0.26 2 0.27
35029 6.29 1.23 12 0.07
35046 5.83 1.69 15 0.03 5.15 0.37 1
35066 5.74 1.78 18 0.03 5.13 0.44 1
35074 5.87 1.65 17 0.02 5.91 1.09 1
35172 6.61 0.91 16 0.06 6.27 0.71 2 0.07
35235 5.95 1.57 16 0.03 5.71 0.81 2 0.06
36036 5.47 2.05 17 0.02 4.82 0.40 2 0.00
36182 6.01 1.51 17 0.04 5.20 0.24 2 0.07
37247 5.89 1.63 18 0.05 5.70 0.86 2 0.04
37329 5.98 1.54 13 0.03
38011 6.27 1.25 14 0.03 6.08 0.86 2 0.07
38303 6.15 1.37 17 0.05 5.99 0.89 2 0.07
39013 6.11 1.41 14 0.03 5.29 0.23 2 0.11
39026 6.33 1.19 17 0.03 5.46 0.18 2 0.07
39034 5.91 1.61 11 0.05 5.66 0.80 2 0.16
39037 5.60 1.92 15 0.02 4.85 0.30 2 0.05
39044 5.61 1.91 17 0.03
39067 6.14 1.38 18 0.03 6.17 1.08 2 0.07
39088 6.06 1.46 16 0.04 6.04 1.03 2 0.04
39352 5.96 1.56 16 0.04 5.75 0.84 2 0.04
39401 5.89 1.63 11 0.04 5.94 1.10 1
40135 5.62 1.90 15 0.02 4.87 0.30 2 0.11
40371 6.02 1.50 18 0.04 5.53 0.56 2 0.07
40479 6.09 1.43 15 0.04 5.38 0.34 2 0.06
41025 6.03 1.49 11 0.03 5.48 0.50 2 0.14
41035 5.67 1.85 15 0.03 4.74 0.12 1
41435 6.08 1.44 18 0.03 5.43 0.40 2 0.07
41455 6.23 1.29 16 0.04 5.86 0.68 2 0.07
41476 6.52 1.00 19 0.05 5.42 0.35 2 0.07
41494 6.25 1.27 10 0.05 5.55 0.35 2 0.18
42023 5.65 1.87 16 0.04 4.78 0.18 1
42084 5.79 1.73 14 0.03 5.89 1.15 2 0.07
42250 5.74 1.78 14 0.05 5.54 0.85 2 0.01
42501 5.88 1.64 14 0.03 5.27 0.44 2 0.05
43010 5.62 1.90 10 0.04 5.34 0.77 1
43024 5.71 1.81 11 0.04 5.88 1.22 1
43061 6.80 0.72 18 0.05 6.32 0.57 2 0.07
43095 5.73 1.79 13 0.02 5.65 0.97 2 0.24
43108 5.91 1.61 12 0.02 5.90 1.04 2 0.05
43111 5.83 1.69 15 0.04
43134 5.65 1.87 12 0.06
43412 5.65 1.87 16 0.03 5.15 0.55 2 0.05
43485 5.71 1.81 12 0.03 5.37 0.71 1
44038 5.75 1.77 11 0.02 5.50 0.80 2 0.07
44065 5.80 1.72 15 0.04 5.04 0.29 2 0.20
44115 5.86 1.66 18 0.03 5.20 0.39 2 0.04
44148 6.52 1.00 13 0.04 6.37 0.90 2 0.07
44449 6.50 1.02 16 0.07 6.17 0.72 2 0.07
45082 5.71 1.81 12 0.04 5.07 0.41 2 0.10
45454 5.75 1.77 13 0.02 5.86 1.16 2 0.03
46024 5.83 1.69 15 0.03 4.97 0.19 2 0.21
46062 5.64 1.88 17 0.03 5.55 0.96 2 0.01
46381 5.90 1.62 14 0.03 6.10 1.25 2 0.02
47012 5.81 1.71 15 0.03 5.19 0.43 1
47420 5.96 1.56 17 0.03 5.89 0.98 2 0.22
48028 5.84 1.68 17 0.02 5.64 0.85 2 0.01
48049 5.76 1.76 17 0.03 5.19 0.48 2 0.05
48060 5.55 1.97 18 0.02 5.43 0.93 2 0.06
48083 6.16 1.36 17 0.03 5.95 0.84 2 0.07
48099 6.48 1.04 15 0.04 6.05 0.62 2 0.07
48392 5.86 1.66 13 0.04 5.87 1.06 2 0.04
49013 6.08 1.44 18 0.04 5.51 0.48 2 0.01
49022 5.81 1.71 13 0.04
49148 5.91 1.61 18 0.03 5.34 0.48 2 0.21
49238 5.93 1.59 14 0.04 5.75 0.87 1
49333 6.20 1.32 11 0.05 5.41 0.26 2 0.14
50046 5.59 1.93 11 0.05 5.36 0.82 1
50253 5.81 1.71 14 0.03 6.12 1.36 2 0.06
51021 6.08 1.44 19 0.03 5.18 0.15 2 0.18
51074 6.77 0.75 15 0.04 6.22 0.46 2 0.07
51091 5.35 2.17 11 0.05 5.43 1.13 2 0.26
51254 6.11 1.41 15 0.03 6.38 1.32 2 0.07
51257 5.94 1.58 8 0.07 5.62 0.73 1
51259 6.04 1.48 13 0.03 5.95 0.96 2 0.15
52017 5.66 1.86 16 0.02 5.04 0.43 1
52035 5.81 1.71 14 0.05 6.07 1.31 2 0.01
52180 6.15 1.37 15 0.04
52222 6.07 1.45 12 0.06
53012 6.25 1.27 13 0.05 5.60 0.40 2 0.07
53054 5.89 1.63 13 0.03 5.25 0.41 1
53067 5.99 1.53 15 0.03 5.79 0.85 1
53114 5.82 1.70 19 0.03 5.26 0.49 2 0.08
53185 5.83 1.69 13 0.03
54018 5.72 1.80 13 0.04 5.96 1.29 2 0.15
54063 6.05 1.47 9 0.05 6.30 1.30 2 0.07
55029 6.13 1.39 19 0.03 6.25 1.17 2 0.07
55063 6.20 1.32 15 0.03 5.96 0.81 2 0.07
55071 6.61 0.91 15 0.03 6.08 0.52 2 0.07
55114 5.88 1.64 17 0.02 5.28 0.45 2 0.00
55121 6.62 0.90 16 0.06 6.65 1.08 2 0.07
55149 6.52 1.00 16 0.04 6.27 0.80 2 0.07
56024 6.15 1.37 14 0.03 6.01 0.91 2 0.07
56087 5.60 1.92 16 0.02 4.94 0.39 2 0.01
57010 6.16 1.36 18 0.03 5.68 0.57 2 0.06
57054 6.03 1.49 18 0.04 5.69 0.71 2 0.09
57073 5.70 1.82 14 0.05 4.93 0.28 1
58043 5.76 1.76 10 0.03 5.10 0.39 1
58087 5.90 1.62 15 0.03 5.06 0.21 2 0.10
59024 6.80 0.72 13 0.05 6.05 0.30 2 0.07
59036 5.96 1.56 13 0.03 5.99 1.08 2 0.01
59047 5.98 1.54 14 0.03 6.27 1.34 2 0.07
59085 5.73 1.79 18 0.03 5.10 0.42 1
60065 5.73 1.79 14 0.02 5.20 0.52 1
60101 6.04 1.48 10 0.04 5.64 0.65 2 0.33
61015 5.86 1.66 18 0.03 5.06 0.25 2 0.07
61026 6.22 1.30 10 0.04 5.77 0.60 2 0.16
61085 6.37 1.15 12 0.03 6.29 0.97 2 0.07
62058 6.37 1.15 13 0.05 5.97 0.65 2 0.07
63027 5.70 1.82 10 0.05 5.50 0.85 1
63052 6.12 1.40 12 0.05
64049 5.78 1.74 12 0.02 5.84 1.11 2 0.08
64067 6.31 1.21 14 0.04
65057 6.27 1.25 12 0.02 6.34 1.12 2 0.07
66047 6.28 1.24 19 0.03 5.46 0.23 2 0.12
67063 6.16 1.36 14 0.04 6.08 0.97 2 0.03
69012 5.67 1.85 17 0.03 5.14 0.52 2 0.04
70035 6.17 1.35 14 0.04 5.56 0.44 2 0.01
70049 6.14 1.38 16 0.04 6.16 1.07 2 0.07
73025 6.25 1.27 16 0.03 6.18 0.98 2 0.07
75024 6.44 1.08 10 0.04 5.58 0.19 1
76027 6.05 1.47 15 0.04
77025 5.81 1.71 15 0.04 5.28 0.52 2 0.18
82024 6.34 1.18 13 0.04 5.39 0.10 2 0.10
85027 5.87 1.65 16 0.04 5.34 0.52 1
aafootnotetext: Identifier from van Leeuwen et al. (2000).bbfootnotetext: Assumed the solar log (Fe)=7.52 (Sneden et al. 1991).
Table 6: Derived Abundances
Element T100 log g0.25 V0.25 N0.30
[K] [cm s] [km s] [dex]
[Fe/H]–2.0
Fe I 0.17 0.02 0.04
Fe II 0.05 0.11 0.03
Al I 0.07 0.02 0.00 0.02
[Fe/H]–1.5
Fe I 0.16 0.00 0.06
Fe II 0.06 0.12 0.04
Al I 0.09 0.01 0.01 0.05
[Fe/H]–1.0
Fe I 0.10 0.01 0.08
Fe II 0.08 0.11 0.04
Al I 0.08 0.00 0.01 0.08
Table 7: Abundance Sensitivity to Model Parameters
Figure 1: A color–magnitude diagram of the upper RGB for Cen. The large filled circles indicate program stars and the small filled circles are those available from the van Leeuwen et al. (2000) proper motion study.
Figure 2: Histogram showing the observed completion fraction of this study. The data are compared to the deeper photometric study of Rey et al. (2004). The top panel shows the completion fraction binned by apparent V magnitude with bin sizes of 0.5 mag. and the bottom panel shows the completion fraction binned by B–V color in 0.1 mag. intervals.
Figure 3: Program stars are shown in terms of position in the field. The cross indicates the field center at 201.691, –47.4769 (J2000) (132645.9, –472837.0). The ellipses indicate 1, 5, and 10 times the core radius of 1.40.
Figure 4: The top panel shows the relation between the effective temperature estimated via V–K photometry versus the spectroscopically determined temperature. The straight line indicates perfect agreement. The bottom panel illustrates microturbulent velocity versus effective temperature. Different symbols indicate stars in different metallicity bins as indicated above. A linear least–squares fit is provided along with the equation relating microturbulence to effective temperature.
Figure 5: Derived [Fe II/H] abundances are plotted versus [Fe I/H]. The line indicates perfect agreement.
Figure 6: Several sample spectra are shown for various [Fe/H]. The spectra have been offset for display purposes. For reference the vertical dashed lines indicate the location of the Al I lines and two additional Fe I lines. From top to bottom, the [Al/Fe] values for the stars shown are +0.40, +0.45, +0.15, +0.97, and +0.57, respectively.
Figure 7: Sample spectrum syntheses of the Al region are shown. The dashed line indicates log (Al)=–5.0, the solid line shows the best–fit Al abundance, and the dotted lines indicate abundance 0.30 dex from the best–fit Al value.
Figure 8: The four panels show our adopted model atmosphere parameters versus those available in the literature. A straight line indicates perfect agreement in all panels.
Figure 9: Al abundances available in the literature are plotted versus those derived here. The straight line indicates perfect agreement. The error bars are those given from each study and this one. If no error is provided, a base value of 0.10 dex is assumed.
Figure 10: A histogram of derived [Fe/H] values with bin sizes of 0.10 dex.
Figure 11: Color–magnitude diagram of program stars displayed in various metallicity bins as shown above.
Figure 12: Al and Fe are plotted as a function of radial distance from the cluster center.
Figure 13: The top panel shows average radial velocity versus log (Al) and the bottom panel is for log (Fe). The filled circles represent average radial velocities in each abundance bin and the vertical bars indicate the velocity dispersion in each bin. Both panels have a bin size of 0.10 dex in abundance.
Figure 14: [Al/Fe] plotted as a function of [Fe/H].
Figure 15: A box plot is shown on top of the [Al/Fe] versus [Fe/H] plot given in Figure 14. The data are binned into 0.10 dex intervals with the boxes centered on each bin. The middle line of each box indicates the median value, the lower and upper bounds of the box are the first and third quartile, the vertical lines are the full data range neglecting outliers, and the open circles indicate data lying 1.5–3.0 times the interquartile range away from either boundary.
Figure 16: Log (Al) is plotted as a function of log (Fe).
Figure 17: Histograms of [Al/Fe] using a bin size of 0.10 dex for multiple metallicity bins.
Comments 0
Request Comment
You are adding the first comment!
How to quickly get a good reply:
  • Give credit where it’s due by listing out the positive aspects of a paper before getting into which changes should be made.
  • Be specific in your critique, and provide supporting evidence with appropriate references to substantiate general statements.
  • Your comment should inspire ideas to flow and help the author improves the paper.

The better we are at sharing our knowledge with each other, the faster we move forward.
""
The feedback must be of minimum 40 characters and the title a minimum of 5 characters
   
Add comment
Cancel
Loading ...
152634
This is a comment super asjknd jkasnjk adsnkj
Upvote
Downvote
""
The feedback must be of minumum 40 characters
The feedback must be of minumum 40 characters
Submit
Cancel

You are asking your first question!
How to quickly get a good answer:
  • Keep your question short and to the point
  • Check for grammar or spelling errors.
  • Phrase it like a question
Test
Test description