arXiv subjects
Publications and source records attributed to Timothy C. Beers.
I was asked at the end of this conference to provide a few comments on the present status of the field of cosmic abundances, which I summarize (and extend upon) in this contribution. A discussion of the present ``state of play'' is given, along with some suggested directions for the future.
We present numerical simulations to describe the evolution of pre-Galactic clouds in a model which is motivated by cold dark matter simulations of hierarchical galaxy formation. We adopt a SN-induced star-formation mechanism within a model that follows the evolution of chemical enrichment and energy input to the clouds by Type II and Type Ia supernovae. We utilize metallicity-dependent yields for all elements at all times, and include effects of finite stellar lifetimes. We derive the metallicity distribution functions for stars in the clouds, their age-metallicity relation, and relative elemental abundances for a number of alpha- and Fe-group elements. The stability of these clouds against destruction is discussed, and results are compared for different initial mass functions. We find that the dispersion of the metallicity distribution function observed in the outer halo is naturally reproduced by contributions from many clouds with different initial conditions. The predicted relative abundances of some alpha- and Fe-group elements show reasonable agreement with the observed values down to metallicities below [Fe/H] ~ -4 when the iron yields are reduced relative to stellar models. The observed scatter is also reproduced for most elements including the observed bifurcation in [alpha/Fe] for stars with low [Fe/H]. However, the predicted dispersion may be too large for some elements unless a limited range of progenitor masses contributing to the abundances of these elements is assumed. The results suggest that the low-mass end of SNeII was probably absent at the very lowest metallicities, and that the upper mass limit for the first stars that contributed to nucleosynthesis may be < 40 M_solar.
An abundance analysis for the carbon-enhanced, extremely iron-poor ([Fe/H] -3.5) star CS29498-043 has been obtained using new high-resolution, high signal-to-noise spectra from the Subaru Telescope. The [O I] forbidden line at 6300A and the O I triplet feature at 7771-7776A are both clearly detected. The overabundance of oxygen is significant ([O/Fe]>2). In addition, Na, Co, and Ni abundances have been newly measured. The abundance pattern from C to Ni of this object is quite similar to that of CS 22949--037, another extremely metal-poor star with large excesses of C, N, O, and the alpha-elements. The abundance patterns of these two stars suggest the existence of supernovae progenitors that ejected relatively little material fromtheir iron cores during the very early era of nucleosynthesis in the Galaxy. Themetallicity in these objects, when one includes the elements C, N, and O in the tally of total metals, is not as low as in the most metal-poor stars, suggesting the existence of quite different formation processes for these iron-deficient objects than pertain to the bulk of other metal-deficient stars.
We report a kinematic signature associated with the observed asymmetry in the distribution of thick disk/inner halo stars interior to the Solar circle described in Paper I. In that paper we found a statistically significant excess (20% to 25 %) of stars in quadrant I (l ~ 20 deg to 55 deg) both above and below the plane (b ~ +/- 25 deg to +/- 45 deg) compared to the complementary region in quadrant IV. We have measured Doppler velocities for 741 stars, selected according to the same magnitude and color criteria, in the direction of the asymmetry and in the corresponding fields in quadrant IV. We have also determined spectral types and metallicities measured from the same spectra. We not only find an asymmetric distribution in the V_LSR velocities for the stars in the two regions, but the angular rate of rotation, w, for the stars in quadrant I reveals a slower effective rotation rate compared to the corresponding quadrant IV stars. We use our [Fe/H] measurements to separate the stars into the three primary population groups, halo, thick disk, and disk, and conclude that it is primarily the thick disk stars that show the slower rotation in quadrant I. A solution for the radial, tangential and vertical components of the V_LSR velocities, reveals a significant lag of ~ 80 to 90 km/s in the direction of Galactic rotation for the thick disk stars in quadrant I, while in quadrant IV, the same population has only a ~ 20 km/s lag. The results reported here support a rotational lag among the thick disk stars due to a gravitational interaction with the bar as the most likely explanation for the asymmetry in both the star counts and the kinematics. The affected thick disk stars, however, may be associated with the recently discovered Canis Major debris stream or a similar merger event (abridged).
We report on the first measurement of the Eu isotope fractions (151Eu and 153Eu) in s-process-element-enhanced, metal-poor stars. We use these ratios to investigate the 151Sm branching of s-process nucleosynthesis. The measurement was made by detailed study of Eu II lines that are significantly affected by hyperfine splitting and isotope shifts in spectra of the carbon-rich, very metal-poor stars LP625-44 and CS31062-050, observed with the Subaru Telescope High Dispersion Spectrograph. The 151Eu fractions [fr(151Eu) = 151Eu/(151Eu+153Eu)] derived for LP625-44 and CS31062-050 are 0.60 and 0.55, respectively, with uncertainties of about +/- 0.05. These values are higher than found in solar-system material, but agree well with the predictions of recent s-process models. We derive new constraints on the temperature and neutron density during the s-process based on calculations of pulsed s-process models for the 151Eu fraction.
HE 0107-5240 is a star in more than once sense of the word. Chemically, it is the most primitive object yet discovered, and it is at the centre of debate about the origins of the first elements in the Universe.
The chemical composition of the most metal-deficient stars reflects the composition of the gas from which they formed. These old stars provide crucial clues to the star formation history and the synthesis of chemical elements in the early Universe. They are the local relics of epochs otherwise observable only at very high redshifts; if totally metal-free (``population III'') stars could be found, this would allow the direct study of the pristine gas from the Big Bang. Earlier searches for such stars found none with an iron abundance less than 1/10,000 that of the Sun, leading to the suggestion that low-mass stars could only form from clouds above a critical iron abundance. Here we report the discovery of a low-mass star with an iron abundance as low as 1/200,000 of the solar value. This discovery suggests that population III stars could still exist, that is, that the first generation of stars also contained long-lived low-mass objects. The previous failure to find them may be an observational selection effect.
We present a detailed abundance analysis, including spectral syntheses, of a very metal-poor ([Fe/H]= -2.7), peculiar main sequence star, HE0024-2523 detected during the course of the Keck Pilot Program. Radial velocities of this star were obtained during four different observing runs over a time span of 1.1 years, and demonstrate that it is clearly a short period spectroscopic binary. An orbital solution was obtained, and orbital parameters were determined with high precision. The rotational velocity was also measured (vsin i=9.7$\pm$1.5 kms); rotation appears likely to be synchronous with the orbit. The abundance analysis and spectral syntheses indicate that the object is a CH star characterized by extreme s-process enrichment, likely due to mass accretion from an evolved companion which has now probably become a white dwarf. The lead (Pb) abundance of HE0024-2523 is very high, the same as that of the recently discovered lead-rich metal-poor star CS 29526-110, [Pb/Fe]=+3.3. The abundance ratio of the heavy-s to light-s elements, as characterized by Pb and Ba, [Pb/Ba]=+1.9, is the highest yet found for any metal-poor star, and is about 0.7 dex higher than that of CS29526-110. On the basis of the measured isotopic ratio of carbon (12C/13C about 6) we argue that the mass donor must have had an original mass of at least 3 Msun. The unusually short period of this CH star suggests that it underwent a past common-envelope phase with its evolved companion. Our results are compared to the latest available models for AGB yields and s-process nucleosynthesis. We also discuss the possible connection between HE0024-2523 the lithium depletion of halo stars, and halo blue straggler formation.
We set new limits on the mass of the Milky Way, making use of the latest kinematic information for Galactic satellites and halo objects. In order to bind these sample objects to the Galaxy, their rest-frame velocities must be lower than their escape velocities at their estimated distances. This constraint enables us to show that the mass estimate of the Galaxy is largely affected by several high-velocity objects (Leo I, Pal 3, Draco, and a few FHB stars), not by a single object alone (such as Leo I), as has often been the case in past analyses. We also find that a gravitational potential that gives rise to a declining rotation curve is insufficient to bind many of our sample objects to the Galaxy; a possible lower limit on the mass of the Galaxy is about 2.2 x 10^12 Msolar. To be more quantitative, we adopt a Bayesian likelihood approach to reproduce the observed distribution of the current positions and motions of the sample, in a prescribed Galactic potential that yields a flat rotation curve. This method enables a search for the most likely total mass of the Galaxy, without undue influence in the final result arising from the presence or absence of Leo I, provided that both radial velocities and proper motions are used. The most likely total mass derived from this method is 2.5^+0.5_-1.0 x 10^12 Msolar (including Leo I), and 1.8^+0.4_-0.7 x 10^12 Msolar (excluding Leo I).
The discovery of the second highly $r$-process-enhanced, extremely metal-poor star, CS 31082-001 ([Fe/H] $= -2.9$) has provided a powerful new tool for age determination, by virtue of the detection and measurement of the radioactive species uranium and thorium. One of the serious limitations of this approach, however, is that predictions of the production ratio of U and Th have not been made in the context of a realistic astrophysical model of the $r$-process. We have endeavored to produce such a model, based on the ``neutrino winds'' that are expected to arise from the nascent neutron star of a core-collapse supernova. The mass-integrated $r$-process yields, obtained by assuming a simple time evolution of the neutrino luminosity, are compared to the available spectroscopic elemental abundance data of CS 31082-001. As a result, the ``age'' of this star is determined to be $14.1 \pm 2.5$ Gyr, in excellent agreement with lower limits on the age of the universe estimated by other dating techniques, as well as with other stellar radioactive age estimates. Future measurements of Pt and Pb in this star, as well as expansion of searches for additional $r$-process-enhanced, metal-poor stars (especially those in which both U and Th are measurable), are of special importance to constrain the current astrophysical models for the $r$-process.
We report medium-resolution (1-2 A) spectroscopy and broadband (UBV) photometry for a sample of 39 bright stars (the majority of which are likely to be giants) selected as metal-deficient candidates from an objective-prism survey concentrating on Galactic latitudes below |b| = 30 deg, the LSE survey of Drilling & Bergeron. Although the primary purpose of the LSE survey was to select OB stars (hence the concentration on low latitudes), the small number of bright metal-deficient giant candidates noted during this survey provide interesting information on the metal-weak thick disk (MWTD) population. The kinematics of the LSE giants indicate the presence of a rapidly rotating population, even at quite low metallicity. We consider the distribution of orbital eccentricity of the LSE giants as a function of [Fe/H], and conclude that the local fraction (i.e., within 1 kpc from the Sun) of metal-poor stars that might be associated with the MWTD is on the order of 30%-40% at abundances below [Fe/H] = -1.0. Contrary to recent analyses of previous (much larger) samples of non-kinematically selected metal-poor stars, we find that this relatively high fraction of local metal-poor stars associated with the MWTD may extend to metallicities below [Fe/H] = -1.6, much lower than had been considered before. We identify a subsample of 11 LSE stars that are very likely to be members of the MWTD, based on their derived kinematics; the lowest metallicity among these stars is [Fe/H] = -2.35. Implications of these results for the origin of the MWTD and for the formation of the Galaxy are considered. (abridged)
In this series of two papers we present a high dispersion spectroscopic analysis of 8 candidate extremely metal poor stars selected from the Hamburg/ESO Survey and of 6 additional very metal poor stars. We demonstrate that with suitable vetting using moderate-resolution spectra the yield of this survey for stars with [Fe/H] $\le -3.0$ dex is very high; three out of the eight stars observed thus far at high resolution from the HES are actually that metal poor, three more have [Fe/H] $\le -2.8$ dex, and the remainder are only slightly more metal rich. In preparation for a large scale effort to mine the Hamburg/ESO Survey database for such stars about to get underway, we lay out in this paper the basic principles we intend to use to determine in a uniform way the stellar parameters Teff, surface grav, and reddening.
We present a detailed abundance analysis of 8 stars selected as extremely metal poor candidates from the Hamburg/ESO Survey (HES). For comparison, we have also analysed 3 extremely metal-poor candidates from the HK survey, and 3 additional bright metal-poor stars. With this work, we have doubled the number of extremely metal-poor stars ([Fe/H]$\le 3.0$) with high-precision abundance analyses. Our sample of extremely metal-poor candidates from the HES contains 3 stars with [Fe/H] $\le -3.0$, 3 more with [Fe/H]$\le -2.8$, and 2 stars that are only slightly more metal rich. Thus, the chain of procedures that led to the selection of these stars from the HES successfully provides a high fraction of extremely metal-poor stars. We verify that our stellar parameters, derived in Paper I, lead to acceptable ionization and excitation balances for Fe, ruling out substantial non-LTE effects in Fe. For the $α-$elements Mg, Si, Ca, Ti, the light element Al, the iron-peak elements Sc, Cr, Mn, and the neutron capture elements Sr and Ba, we find trends in abundance ratios [X/Fe] similar to those found by previous studies. However,the scatter in most of these ratios, even at [Fe/H]$\le -3.0$ dex, is surprisingly small. Only Sr and Ba show scatter larger than the expected errors. Future work (the 0Z project) will provide much stronger constraints on the scatter (or lack thereof) in abundances for a greater number of stars. We discuss the implications of these results for the early chemical evolution of the Galaxy, including such issues as the number of contributing SN, and the sizes of typical fragments in which they were born. In addition, we have identified a very metal poor star that appears to be the result of the s-process chain, operating in a very metal-poor environment, with extremely enhanced C, Ba, and Pb, and somewhat enhanced Sr.
We have combined new high-resolution spectra obtained with the Hubble Space Telescope (HST) and ground-based facilities to make a comprehensive new abundance analysis of the metal-poor, halo star BD +17^\circ 3248. We have detected the third r-process peak elements osmium, platinum, and (for the first time in a metal-poor star) gold, elements whose abundances can only be reliably determined using HST. Our observations illustrate a pattern seen in other similar halo stars with the abundances of the heavier neutron-capture elements, including the third r-process peak elements, consistent with a scaled solar system r-process distribution. The abundances of the lighter neutron-capture elements, including germanium and silver, fall below that same scaled solar r-process curve, a result similar to that seen in the ultra-metal-poor star CS 22892--052. A single site with two regimes or sets of conditions, or perhaps two different sites for the lighter and heavier neutron-capture elements, might explain the abundance pattern seen in this star. In addition we have derived a reliable abundance for the radioactive element thorium. We tentatively identify U II at 3859 A in the spectrum of BD +17^\circ 3248, which makes this the second detection of uranium in a very metal-poor halo star. Our combined observations cover the widest range in proton number (from germanium to uranium) thus far of neutron-capture elements in metal-poor Galactic halo stars. Employing the thorium and uranium abundances in comparison with each other and with several stable elements, we determine an average cosmochronological age for BD +17^\circ 3248 of 13.8 +/- 4 Gyr, consistent with that found for other similar metal-poor halo stars.
Observations of eighteen halo main-sequence-turnoff stars, four of which are extremely deficient in Li, show that three of the Li-poor ones have substantial line broadening. We attribute this to stellar rotation. Despite the great ages of halo stars, for G202-65, BD+51:1817 and Wolf 550 we infer vsini = 8.3 +/- 0.4, 7.6 +/- 0.3, and 5.5 +/- 0.6 km/s respectively. The stated errors are 3sigma. For CD-31:19466 we derive a 3sigma upper limit vsini < 2.2 km/s. The three rotating stars are known spectroscopic binaries. We explain the high rotation velocities in terms of mass and angular momentum transfer onto the surface of the turnoff star from an initially more-massive donor. Estimates of the specific angular momentum of accreted material indicate that quite small transfer masses could have been involved, though the unknown subsequent spin-down of the accretor prevents us from assigning definitive values for each star. The accretor is now seen as an ultra-Li-deficient star whose origin makes it a low-mass counterpart of field blue stragglers. The Li could have been destroyed before or during the mass-transfer episode. Such objects must be avoided in studies of the primordial Li abundance and in investigations into the way normal single stars process their initial Li.
Europium isotopic abundance fractions are reported for the very metal-poor, neutron-capture-rich giant stars CS 22892-052, HD 115444, and BD +17 3248. The abundance fractions, derived from analysis of several strong Eu II lines appearing in high-resolution spectra of these stars, are in excellent agreement with each other and with their values in the Solar System: fraction(\iso{Eu}{151}) ~= fraction(\iso{Eu}{153}) ~= 0.5. Detailed abundance studies of very metal-poor stars have previously shown that the total elemental abundances of stable atoms with atomic numbers z >= 56 typically match very closely those of a scaled solar-system r-process abundance distribution. The present results for the first time extend this agreement to the isotopic level.
We report on an analysis of the chemical composition of five carbon-rich, very metal-poor stars based on high-resolution spectra. One star, CS22948-027, exhibits very large overabundances of carbon, nitrogen, and the neutron-capture elements, as found in the previous study of Hill et al.. This result may be interpreted as a consequence of mass transfer from a binary companion that previously evolved through the asymptotic giant branch stage. By way of contrast, the other four stars we investigate exhibit no overabundances of barium ([Ba/Fe]<0), while three of them have mildly enhanced carbon and/or nitrogen ([C+N]+1). We have been unable to determine accurate carbon and nitrogen abundances for the remaining star (CS30312-100). These stars are rather similar to the carbon-rich, neutron-capture-element-poor star CS22957-027 discussed previously by Norris et al., though the carbon overabundance in this object is significantly larger ([C/Fe]=+2.2). Our results imply that these carbon-rich objects with ``normal'' neutron-capture element abundances are not rare among very metal-deficient stars. One possible process to explain this phenomenon is as a result of helium shell flashes near the base of the AGB in very low-metallicity, low-mass (M~< 1M_sun) stars, as recently proposed by Fujimoto et al.. The moderate carbon enhancements reported herein ([C/Fe]+1) are similar to those reported in the famous r-process-enhanced star CS22892-052. We discuss the possibility that the same process might be responsible for this similarity, as well as the implication that a completely independent phenomenon was responsible for the large r-process enhancement in CS22892-052.
We explore the application of artificial neural networks (ANNs) for the estimation of atmospheric parameters (Teff, logg, and [Fe/H]) for Galactic F- and G-type stars. The ANNs are fed with medium-resolution (~ 1-2 A) non flux-calibrated spectroscopic observations. From a sample of 279 stars with previous high-resolution determinations of metallicity, and a set of (external) estimates of temperature and surface gravity, our ANNs are able to predict Teff with an accuracy of ~ 135-150 K over the range 4250 <= Teff <= 6500 K, logg with an accuracy of ~ 0.25-0.30 dex over the range 1.0 <= logg <= 5.0 dex, and [Fe/H] with an accuracy ~ 0.15-0.20 dex over the range -4.0 <= [Fe/H] <= +0.3. Such accuracies are competitive with the results obtained by fine analysis of high-resolution spectra. It is noteworthy that the ANNs are able to obtain these results without consideration of photometric information for these stars. We have also explored the impact of the signal-to-noise ratio (S/N) on the behavior of ANNs, and conclude that, when analyzed with ANNs trained on spectra of commensurate S/N, it is possible to extract physical parameter estimates of similar accuracy with stellar spectra having S/N as low as 13. Taken together, these results indicate that the ANN approach should be of primary importance for use in present and future large-scale spectroscopic surveys.