1: % file /scr2/jlc/hamburg_survey/paper_he1424_2007/ms.tex
2: %
3: % revised to accomodate referee's report, received March 1, 2007
4: %
5:
6:
7: %\documentclass[preprint]{aastex}
8: \documentclass[12pt,preprint]{aastex}
9:
10: \newcommand{\kms}{km~s$^{-1}$}
11: \newcommand{\subsun}{\mbox{$_{\odot}$}}
12: \newcommand{\teff}{$T_{\rm{eff}}$}
13: \newcommand{\grav}{log($g$)}
14: \newcommand{\etal}{{\it et al.\/}}
15: \newcommand{\eqw}{$W_{\lambda}$}
16: \newcommand{\fe}{[Fe/H]}
17: \newcommand{\mtv}{$v_t$}
18: \newcommand{\cband}{C$_2$}
19: \newcommand{\ciso}{$^{12}$C/$^{13}$C}
20: \newcommand{\mystar}{HE1424$-$0241}
21:
22: \begin{document}
23:
24:
25: \title{A New Type of Extremely Metal Poor Star\altaffilmark{1}}
26:
27: \author{Judith G. Cohen\altaffilmark{2},
28: Andrew McWilliam\altaffilmark{3},
29: Norbert Christlieb\altaffilmark{4},
30: Stephen Shectman\altaffilmark{3}, Ian Thompson\altaffilmark{3},
31: Jorge Melendez\altaffilmark{5},
32: Lutz Wisotzki\altaffilmark{6} \& Dieter Reimers\altaffilmark{7} }
33:
34: \altaffiltext{1}{Based in part on observations obtained at the
35: W.M. Keck Observatory, which is operated jointly by the California
36: Institute of Technology, the University of California, and the
37: National Aeronautics and Space Administration.}
38:
39: \altaffiltext{2}{Palomar Observatory, Mail Stop 105-24,
40: California Institute of Technology, Pasadena, Ca., 91125,
41: jlc@astro.caltech.edu}
42:
43: \altaffiltext{3}{Carnegie Observatories of Washington, 813 Santa
44: Barbara Street, Pasadena, Ca. 91101, andy, ian, shec@ociw.edu}
45:
46: \altaffiltext{4}{Current address: Department of
47: Astronomy and Space Physics, Uppsala University, Box 515,
48: 75120 Uppsala, Sweden, formerly at Hamburger Sternwarte, Universit\"at
49: Hamburg, Gojenbergsweg 112, D-21029 Hamburg, Germany, norbert@astro.uu.se}
50:
51: \altaffiltext{5}{Palomar Observatory, Mail Stop 105-24,
52: California Institute of Technology, Pasadena, Ca., 91125,
53: Current address: Australian National University, Australia,
54: jorge@mso.anu.edu.au}
55:
56: \altaffiltext{6}{Astrophysical Institute Potsdam, An der Sternwarte 16,
57: D-14482 Potsdam, Germany, lwisotzki@aip.de}
58:
59: \altaffiltext{7}{Hamburger Sternwarte, Universit\"at
60: Hamburg, Gojenbergsweg 112, D-21029 Hamburg, Germany,
61: dreimers@hs.uni-hamburg.de}
62:
63:
64:
65:
66: \begin{abstract}
67:
68:
69: We present an abundance analysis for the extremely metal poor star \mystar\
70: based on high dispersion spectra from HIRES at Keck.
71: This star is a giant on the lower red giant branch with [Fe/H] $\sim -4.0$~dex.
72: Relative to Fe, \mystar\ has normal
73: Mg, but it shows a very large deficiency of Si,
74: with $\epsilon$(Si)/$\epsilon$(Fe) $\sim$ 1/10
75: and $\epsilon$(Si)/$\epsilon$(Mg) $\sim$ 1/25 that of all previously
76: known extremely metal poor giants or dwarfs. It also has a moderately large deficiency of
77: Ca and a smaller deficit of Ti, combined with enhanced Mn and Co
78: and normal or low C.
79: We suggest that in \mystar\ we see the effect of a
80: very small number of contributing supernovae, and
81: that the SNII contributing to the chemical inventory of \mystar\
82: were biased in progenitor mass or in explosion characteristics
83: so as to reproduce its abnormal extremely low Si/Mg ratio.
84: \mystar\ shows a deficiency of the explosive $\alpha$-burning elements
85: Si, Ca and Ti coupled with a ratio [Mg/Fe]
86: normal for EMP stars; Mg is produced
87: via hydrostatic $\alpha$-burning.
88: The latest models of nucleosynthesis in SNII fail to reproduce
89: the abundance ratios seen in \mystar\ for any combination of
90: the parameter space of core-collapse explosions they explore.
91: % The older models of Woosley \& Weaver
92: % come closer to reproducing our abundance data for \mystar.
93:
94: \end{abstract}
95:
96: \keywords{nuclear reactions, nucleosynthesis, abundances
97: --- stars: abundances --- supernovae: general}
98:
99:
100: \section{Introduction}
101:
102: Extremely metal poor stars provide important clues to the chemical
103: history of our Galaxy: the role and type of early SN, the
104: mode of star formation in the proto-Milky Way, and the formation
105: of the Galactic halo. The classes and properties of EMP stars are summarized by \cite{beers05}.
106: The peculiarities discussed there revolve around enhancements of the elements
107: C and N, which are often accompanied by enhancements of the
108: neutron capture elements beyond the Fe peak.
109: Mass transfer within a binary system which occurred while the former
110: primary was an AGB star is an explanation widely suggested for
111: the majority of these peculiarities, including excesses of both
112: CNO and heavy $s$-process neutron capture elements, see, e.g. \cite{cohen06}.
113:
114:
115: The number of extremely metal poor (EMP)
116: stars known below [Fe/H] $-3.5$ dex\footnote{The
117: standard nomenclature is adopted; the abundance of
118: element $X$ is given by $\epsilon(X) = N(X)/N(H)$ on a scale where
119: $N(H) = 10^{12}$ H atoms. Then
120: [X/H] = log$_{10}$[N(X)/N(H)] $-$ log$_{10}$[N(X)/N(H)]\subsun, and similarly
121: for [X/Fe].} is very small. We have been
122: trying to increase it through
123: data mining of the Hamburg/ESO Survey (HES) \citep{wis00}.
124: In this
125: paper we report our discovery of an extremely metal poor star which
126: shows peculiarities in its chemical abundance distribution
127: not seen in any other such star to date that is known to the authors.
128:
129:
130:
131: \section{Stellar Parameters and Analysis}
132:
133: \mystar\ (R.A.=14 26 40.3, Dec= $-02$ 54 28, J2000)
134: was observed in May 2004 with HIRES \citep{vogt94} at the
135: Keck~I telescope. Based on this high resolution spectrum, whose
136: total exposure time was 3600 sec,
137: it was recognized at that time as an interesting EMP star with very low
138: Fe-metallicity. It was observed again with HIRES in April 2006
139: after the detector upgrade with a total exposure time of 6000 sec. This yielded
140: wider spectral coverage extending far into the UV and
141: a better signal-to-noise ratio than the original data.
142:
143:
144: To determine stellar atmosphere parameters we use the procedures
145: described in \cite{cohen02} and adopted in all
146: subsequent work by our 0Z project published to date.
147: Our \teff\ determinations are based on broad band colors
148: $V-I, V-J$ and $V-K$.
149: The IR photometry is taken from 2MASS \citep{2mass1,2mass2}.
150: We have
151: obtained new photometry
152: at $V$ and at $I$ for \mystar\ ($V = 15.45\pm0.03$ mag and
153: $I = 14.54\pm0.03$~mag) from
154: ANDICAM images taken for this purpose over the past two years via a service
155: observing queue on
156: the 1.3m telescope at CTIO operated
157: by the SMARTS consortium\footnote{See http://www.astronomy.ohio-state.edu/ANDICAM and
158: http://www.astro.yale.edu/smarts.}.
159: We derive surface gravities through combining these \teff\ with
160: an appropriate
161: 12 Gyr isochrone from the grid of \cite{yi01}.
162: We thus derive \teff = 5195~K and \grav = 2.50~dex. The narrow Balmer
163: lines do not permit the star to be a dwarf below the main sequence turnoff.
164:
165: The abundance analysis was carried out in a manner similar to those
166: described in \cite{cohen04}. Full details will be
167: given in an upcoming paper which will present the most metal poor
168: stars we have found thus far.
169: If \teff\ for \mystar\ were to be increased by 100~K, the deduced [Fe/H]
170: would increase by 0.15~dex, but the abundance ratios [X/Fe] would
171: be essentially unchanged.
172:
173: \section{Abundances in \mystar}
174:
175: The abundances we derive for \mystar\ are given in Table~\ref{table_abund}.
176: The number of lines used and the $\sigma$ of the derived
177: log[$\epsilon$(X)] is given for each species for which absorption
178: lines could be detected; upper limits for some key
179: elements are included. These results are compared to the
180: evaluation at [Fe/H] $-4.0$~dex of linear fits to the
181: abundance ratios determined by our 0Z project of
182: stars from the HES from the 0Z project (many still unpublished)
183: with \teff $< 6000$~K and without substantial
184: carbon enhancement ([C/Fe] $< +1.0$~dex).
185: In the last column of the table we give same as determined
186: by \cite{cayrel04} for EMP giants.
187: The dispersion about their regression lines
188: for giants with $-4.2 <$ [Fe/H] $< -3.1$~dex is
189: small, only 0.11 dex for Mg, 0.20 dex
190: for Si, and 0.11 dex for Ca. The extremely good agreement between
191: the abundance ratios for EMP stars found by these two independent large survey
192: projects, our 0Z project and the First Stars VLT project,
193: and listed in the table is very gratifying, and provides support for our
194: statements about the extreme peculiarities of \mystar.
195:
196:
197:
198: The anomalies seen in \mystar\ are many. The most extreme and most
199: peculiar is the very large deficit of Si, with [Si/Fe] $\sim -1.0$~dex
200: and [Si/Mg] $\sim -1.4$~dex,
201: while all other known EMP stars have [Si/Fe] $\sim +0.3$~dex
202: and [Si/Mg] $\sim -0.3$~dex.
203: [Si/Fe] is low in \mystar\ by more than 6$\sigma$\footnote{$\sigma$ here is
204: the sum in quadrature of the uncertainty in [X/Fe] for \mystar\
205: and that of the uncertainty of the linear regression for the
206: ``normal'' EMP giants.} compared to all
207: other known EMP giants, as is shown in Fig.~\ref{figure_sica_feh}.
208: \mystar\ also has a moderately large deficiency of
209: Ca (significant at the 5$\sigma$ level) and a smaller deficit of Ti.
210: It has enhanced Mn and strongly enhanced Co (significant at the 4$\sigma$
211: level), both odd atomic number elements. Copper (another
212: odd atomic number Fe-peak element) may also be enhanced but
213: the single detected line is the rarely observed resonance line at 3274~\AA.
214: Carbon is not enhanced and the heavy neutron capture elements
215: Sr and Ba
216: have low abundances relative to Fe, suggesting that mass transfer
217: in a binary system involving an AGB star is not the cause of
218: the peculiar abundance ratios found in \mystar.
219: Each of these anomalies are seen in both the May 2004 and April 2006
220: HIRES spectra. For example, the equivalent width of the only detected
221: Si~I line (at 3905~\AA)
222: is 17.7~m\AA\ from the 2004 spectrum
223: and 13.9~m\AA\ from the latter one.
224:
225:
226: No other EMP star shows the low Si/Fe and
227: Ca/Fe ratios seen in \mystar.
228: With one minor exception,
229: no other EMP dwarf or giant that is not C-enhanced is known to show
230: highly statistically significant abundance ratio deviations for any elements
231: between
232: Mg and Ni. (C-enhanced EMP stars sometimes show large enhancements
233: of the light elements, for example CS22949$-$037, found by
234: McWilliam et al 1995, analyzed again by Depagne et al al 2002.)
235: The exception is
236: the dwarf HE2344$-$2800 with [Fe/H] $\sim -2.7$~dex,
237: found in the Keck Pilot Project
238: (Cohen et al 2002, Carretta et al 2002) to have an excess of Mn,
239: with $\epsilon$(Mn)/$\epsilon$(Fe) $\sim$ twice the prevailing
240: value among EMP stars. This has been confirmed by a better HIRES
241: spectrum acquired in 2004; this dwarf also has a small excess of Ti relative to Fe.
242: A few C-normal EMP stars (CS22169$-$035 and CS22952$-$015,
243: for example, both of which are
244: included in Fig.~\ref{figure_sica_feh}), have slightly low $\alpha$-elements, but,
245: as the figure clearly illustrates, in
246: no case do they approach the anomalies seen in \mystar.
247:
248:
249:
250: \section{Comparison With Predicted SNII Yields}
251:
252:
253: At least several SN contribute to the chemical inventory of stars
254: with [Fe/H] $\gtrsim -3$~dex, and the
255: observed ratios of the chemical elements are determined by a
256: sum over an assumed initial mass function of predicted SNII yields.
257: SNIa and AGB stars also contribute at
258: still higher metallicity and later times.
259: But given the very low metallicity of \mystar,
260: ejected material from only a very small number of core collapse SN
261: are presumed to have contributed to the material in this star. We must
262: therefore find a model SNII whose predicted nucleosynthetic yields
263: match the abundance ratios seen in this star.
264: $^{28}$Si is formed
265: largely in regions interior to where the bulk of
266: the $^{24}$Mg is produced, although of course nothing of
267: either of these species remains in the central region of the
268: SN, which is mostly $^{56}$Ni. Thus the details of the SN explosion
269: model are important in determining the Si/Mg ratio in the ejected
270: material. We require a range in the ratio of $^{28}$Si/$^{24}$Mg
271: in the ejected material of at least a factor of 10 to reproduce
272: the behavior of both \mystar, with its strong deficit of
273: explosive $\alpha$-burning elements but normal Mg (from hydrostatic
274: $\alpha$-burning) and of all previously
275: known ``normal'' EMP stars.
276:
277: The older models of \cite{woosley}
278: are much more effective at reproducing the observed distribution
279: of abundance ratios in \mystar. Mg/Si production varies
280: by a factor exceeding 10 in these models, with Mg yields
281: highest at masses near 35~$M$\subsun, while Si yields
282: reach their maximum in SNII with lower progenitor masses
283: near 20~$M$\subsun. These yields can qualitatively reproduce the
284: behavior seen in \mystar.
285:
286: However, none of the SNII models in the grids
287: recently calculated by \cite{chieffi} and by \cite{kobayashi}
288: comes close to
289: reproducing the abundance ratios among the $\alpha$-elements
290: seen in \mystar. Both studies provide predictions
291: of explosive
292: yields for SNII progenitors with a wide range of initial masses
293: from 13 to 35 or 50~$M$\subsun\ with a wide range of metallicities.
294: They included an extensive network
295: of nuclear reactions. For the mass cut adopted in each of these
296: two studies, they each predict
297: yields after the radioactive decays
298: for $^{28}$Si/$^{24}$Mg whose range over the entire
299: set of model explosions does not
300: exceed a factor of two. However, the two studies differ
301: in which mass range of SN progenitors produces larger
302: ratios of $^{28}$Si/$^{24}$Mg, \cite{chieffi} favoring
303: lower mass progenitors, while \cite{kobayashi} suggests
304: progenitor masses at the upper end of the range they consider.
305: In no case does the predicted production ratio [Si/Fe] become
306: less than $-0.1$~dex.
307:
308:
309: The most recent predictions of nucleosynthesis yields
310: in SNII have undoubtedly been tuned to reproduce the
311: behavior of the previously known stars EMP stars with [Fe/H]
312: reaching down to $\sim -4$ dex.
313: The abundance ratios we have derived for \mystar, however, demonstrate
314: that these models do not reproduce the full range of
315: the behavior of nucleosynthesis achieved in real SNII
316: and seen among the most peculiar of the the large sample of EMP stars
317: we have studied in the 0Z project.
318:
319:
320:
321:
322: Production of the odd atomic number elements Mn and Co occurs
323: through incomplete Si-burning for Mn and complete Si-burning for Co.
324: \cite{kobayashi} point out that the odd-to-even ratio
325: among the Fe-peak elements depends on the mixing-fallback process, the
326: explosion energy, and the neutron excess $Y_e$. While again
327: no published model can reproduce the large excess of Co and Mn relative
328: to Fe seen in \mystar, we must hope that some combination
329: of these parameters can be found that will accomplish this task.
330:
331:
332:
333: \section{Behavior of The $\alpha$-Elements }
334:
335:
336:
337: Differential analyses of large samples of stars within a small range
338: of \teff\ in the thin disk of the Galaxy as compared
339: to stars in the thick disk such as those of \cite{edvard93},
340: \cite{bensby04}, and \cite{reddy06} have been able to achieve
341: very high precision. These surveys have
342: demonstrated that the trends of [X/Fe] versus [Fe/H] are not identical
343: between the various stellar populations of the Galaxy.
344: But in such studies to date, all the $\alpha$-elements are believed to
345: have varied together and to show the same trends.
346:
347: A few moderately metal-poor halo field stars have
348: been found that appear to be $\alpha$-poor; \cite{fulbright02}
349: suggests that lower [$\alpha$/Fe] stars are found among those with high
350: space velocities with respect to the local standard of rest, while
351: \cite{stephens02} suggest such stars are associated with the outer halo.
352: The most extreme $\alpha$-poor stars, including that found by
353: \cite{carney97}, were reviewed by \cite{ivans}. However, these
354: stars show depletions of Na, Al, Mg, Si and Ca with respect to Fe.
355: They are sufficiently metal-rich compared to \mystar\ that their
356: chemical inventory has a composite origin, with SNIa, SNII and AGB
357: stars all contributing, and can be qualitatively explained by
358: varying the SNIa/SNII ratio, an explanation which
359: cannot be applied to \mystar.
360:
361: All this, while interesting, is not the key issue for the abundance
362: distribution of the EMP giant \mystar.
363: \cite{woosley} find that Si, Ca, and Ti are formed by explosive
364: $\alpha$ burning in SNII, while O and Mg are produced by hydrostatic
365: $\alpha$ burning. \mystar\ shows a clear large deficiency
366: of the former elements, but no apparent deficiency of the hydrostatic
367: $\alpha$ burning element Mg.
368:
369: As abundance analyses have reached higher levels of accuracy
370: (or at least of internal accuracy) and as sample sizes have increased,
371: there have been reports
372: of small differences, much smaller than those we find in \mystar,
373: between the behavior of the explosive and hydrostatic
374: $\alpha$-elements in certain specific environments.
375: The recent analyses of \cite{fulbright07} of a
376: sample of 27 red giants with Keck/HIRES spectra in Baade's Window
377: in the Galactic bulge found
378: that the explosive $\alpha$-elements
379: Si, Ca and Ti have similar trends of
380: [X/Fe] as a function of [Fe/H]. However they found
381: that the hydrostatic $\alpha$-elements
382: O and Mg show a different behavior in the bulge giants.
383: This separation within the Galactic bulge sample is small;
384: $\sim$0.2~dex in total, much smaller than what we observe in \mystar.
385: % Gal bulge: small; for [Fe/H] $\sim -1.4$~dex, [Mg/(expl.$\alpha$)] $\sim 0.0$~dex,
386: % while for [Fe/H] $\sim +0.4$~dex, [Mg/exp.$\alpha$] $\sim +0.2$~dex.
387: % This is a much smaller effect that what we observe in \mystar.
388: \cite{fulbright07} detected similar effects, again on
389: a much smaller scale than in \mystar,
390: in a second environment,
391: among stars in the Milky Way dwarf spheroidal satellite galaxies. They
392: used the compilation of data from the literature by \cite{venn}, which relies
393: heavily on the work of Shetrone, see, e.g. \cite{shetrone}.
394: % These giants have lower ratios of [Ca/Fe] and [Ti/Fe] by $\sim$0.2~dex
395: % than of [Mg/Fe] at [Fe/H] $\lesssim -2$~dex.
396:
397:
398: These examples demonstrate that the production ratios of
399: the explosive to hydrostatic $\alpha$ elements are not fixed;
400: they must depend
401: on environment, the IMF, the star formation history, or other
402: relevant factors. The subtle differences seen in the Galactic bulge
403: and in dSph giants between the behavior of these two groups of
404: $\alpha$ elements, with the the explosive $\alpha$-elements
405: being more depleted than the hydrostatic ones, are seen in a much
406: more dramatic fashion in \mystar.
407: \mystar\ is a very extreme example of this phenomenon
408: in a situation where only
409: a very few SN contributed to the chemical inventory of this star
410: and where, because of the very low metallicity of
411: \mystar, most other possible explanations for this become irrelevant.
412:
413:
414: \section{Summary}
415:
416: All C-normal EMP giants studied to date in the two major surveys,
417: our 0Z project \citep{cohen04} and the First Stars VLT project \citep{cayrel04},
418: show smooth trends of abundance ratios [X/Fe] with Fe-metallicity with
419: modest dispersion around these trends and no strong outliers.
420: \mystar, with [Fe/H] $\sim -4.0$~dex,
421: breaks this paradigm. It is a many $\sigma$
422: outlier in several of the abundance ratios,
423: with $\epsilon$(Si)/$\epsilon$(Fe) $\sim$ 1/10
424: and $\epsilon$(Si)/$\epsilon$(Mg) $\sim$ 1/25
425: that of all previously
426: known extremely metal poor giants or dwarfs, but normal
427: [Mg/Fe]. It also has a moderately large deficiency of
428: Ca and a smaller deficit of Ti, combined with enhanced Mn and
429: highly enhanced Co, both odd atomic number elements.
430: With respect to Fe, C is normal or low in \mystar\
431: (the G band of CH was not
432: detected) and the heavy neutron capture elements are low.
433:
434:
435: From the point of view of SNII nucleosynthesis, \mystar\ is deficient
436: in the explosive $\alpha$-elements, but has a normal [Mg/Fe] ratio, where
437: Mg is produced in hydrostatic $\alpha$-burning. Recent models
438: of production yields in SNII fail completely to reproduce the behavior of
439: the $\alpha$-elements \mystar,
440: whose chemical inventory presumably resulted from a very small number
441: of previous SNII combined with any contributions from a hypothesized
442: Pop. III. They also fail to reproduce the huge excess of Co with
443: respect to Fe.
444: These predicted yields are sensitive to the mass cut, the adopted electron
445: excess profile, and to other explosion
446: characteristics assumed in the calculations for model SN. They
447: presumably were tuned to reproduce the behavior
448: of the previously known EMP stars, so their failure to come close
449: to reproducing the highly anomalous abundance distribution in \mystar\ is perhaps
450: understandable.
451:
452: \mystar\ thus provides important clues as to the details
453: of SNII explosions and their nuclear production yields.
454: It is so metal poor that no explanation other than unusual core collapse
455: SN nucleosynthesis yields can be invoked to explain its unique
456: abundance ratios.
457: Modifications to standard
458: SNII models will need to be made to find explosion parameters which
459: can reproduce the properties we have derived for the peculiar
460: EMP giant \mystar.
461:
462: \acknowledgements
463:
464: We are grateful to the many people
465: who have worked to make the Keck Telescope and HIRES
466: a reality and to operate and maintain the Keck Observatory.
467: The authors wish to extend special thanks to those of Hawaiian ancestry
468: on whose sacred mountain we are privileged to be guests.
469: Without their generous hospitality, none of the observations presented
470: herein would have been possible.
471: This publication makes use of data from the Two Micron All-Sky Survey,
472: which is a joint project of the University of Massachusetts and the
473: Infrared Processing and Analysis Center, funded by the
474: National Aeronautics and Space Administration and the
475: National Science Foundation.
476: J.G.C. is grateful to NSF grant AST-0507219 for partial support.
477: N.C. is a Research Fellow of the Royal Swedish Academy of
478: Sciences supported by a grant from the Knut and Alice
479: Wallenberg Foundation. He also acknowledges financial
480: support from Deutsche Forschungsgemeinschaft through grants
481: Ch~214/3 and Re~353/44.
482:
483: \clearpage
484:
485: \begin{thebibliography}{}
486:
487: \bibitem[Beers \& Christlieb(2005)]{beers05}
488: Beers, T.~C. \& Christlieb, N., 2005, \araa, 43, 531
489:
490: \bibitem[Bensby, Feltzing \& Lundstrom(2004)]{bensby04}
491: Bensby, T., Feltzing, S. \& Lundstr\"{o}m, I., 2004,
492: \aap, 415, 155
493:
494: \bibitem[Carney \etal(1997)]{carney97}
495: Carney, B.~W., Wright, J.~S., Sneden, C., Laird, J.~B.,
496: Aguilar, L.~A. \& Latham, D.~W., 1997, \aj, 114, 363
497:
498: \bibitem[Carretta \etal(2002)]{carretta02} Carretta, E., Gratton, R.~G.,
499: Cohen, J.~G., Beers, T.~C. \& Christlieb, N., 2002, \aj, 124, 481
500: % Keck pilot project, abundances
501:
502: \bibitem[Cayrel \etal(2004)]{cayrel04} Cayrel, R. \etal\, 2004, \aap, 416, 1117
503:
504: \bibitem[Chieffi \& Limongi(2004)]{chieffi}
505: Chieffi, N. \& Limongi, M., 2004, \apj, 608, 405
506:
507: % \bibitem[Clayton(2003)]{clayton}
508: % Clayton, D., 2003, {\it{Handbook of Isotopes in the Cosmos}},
509: % Cambridge U. Press
510:
511: \bibitem[Cohen \etal(2002)]{cohen02} Cohen, J.~G., Christlieb, N.,
512: Beers, T.~C., Gratton, R.~G. \& Carretta, E., 2002, \aj, 124, 470
513:
514: \bibitem[Cohen \etal(2004)]{cohen04}
515: Cohen, J.~G., Christlieb, N., McWilliam, A., Shectman, S.,
516: Thompson, I., Wasserburg, G.~J., Ivans, I., Dehn, Karlsson, T. \&
517: Melendez, J., 2004, \apj, 612, 1107
518:
519: \bibitem[Cohen \etal(2006)]{cohen06}
520: Cohen, J.~G. \etal, 2006, \aj, 132, 137
521: C-star abundance long paper
522:
523: \bibitem[Cutri \etal(2003)]{2mass2} Cutri, R.~M. \etal, 2003,
524: ``Explanatory Supplement to the 2MASS All-Sky Data Release,
525: http://www.ipac.caltech.edu/2mass/releases/allsky/doc/explsup.html
526:
527: \bibitem[Depagne \etal(2002)]{depagne}
528: Depagne, E. \etal, 2002, \aap, 390, 187
529: % First Stars II, CS29497-037, sprocess at low metallicity
530:
531: \bibitem[Edvardsson \etal(1993)]{edvard93}
532: Edvardsson, B., Andersen, J., Gustafsson, B., Lambert, D.~L.,
533: Nissen, P.~E. \& Tomkin, J., 1993, \aap, 275, 101
534:
535: \bibitem[Fulbright(2002)]{fulbright02} Fulbright, J.~P., 2002,
536: \aj, 123, 404
537:
538: \bibitem[Fulbright, McWilliam \& Rich(2007)]{fulbright07}
539: Fulbright, J.~P., McWilliam, A. \& Rich, R.~M., 2007,
540: \apj, submitted
541:
542: \bibitem[Ivans \etal(2003)]{ivans}
543: Ivans, I.~I. \etal, 2003, \apj, 592, 906
544:
545: \bibitem[Kobayashi \etal(2006)]{kobayashi}
546: Kobayashi, C., Umeda, H., Nomoto, K., Tominaga, N.
547: \& Ohkubo, W., 2007, \apj, 653, 1145
548:
549: \bibitem[McWilliam \etal(1995b)]{mcwilliam95b}
550: McWilliam, A., Preston, G.~W., Sneden, C. \& Searle, L., 1995, \aj,
551: 109, 2757
552:
553: \bibitem[Reddy \etal(2006)]{reddy06}
554: Reddy, B.~E., Tomkin, J., Lambert, D.~L. \& Allende Prieto, C.,
555: 2006, \mnras, 367, 1329
556:
557: \bibitem[Shetrone \etal(2003)]{shetrone}
558: Shetrone, M.D., Venn, K.A., Tolstoy, E., Primas, F.,
559: Hill, V. \& Kaufer, A., 2003, \aj, 125, 684
560:
561: \bibitem[Skrutskie \etal(2006)]{2mass1}
562: Skrutskie, M.~F. \etal, 2006, \aj, 131, 1163
563:
564: \bibitem[Stephens \& Boesgaard(2002)]{stephens02}
565: Stephens, A. \& Boesgaard, A.~M., 2002, \aj, 123, 1647
566:
567: \bibitem[Tominaga, Umeda \& Nomoto(2007)]{tominaga}
568: Tominaga, N., Umeda, H. \& Nomoto, K., 2007, \apj, 660,
569: in press
570:
571: \bibitem[Venn \etal(2004)]{venn}
572: Venn, K.~A., Irwin, M., Shetrone, M.~D., Tout, C.~A., Hill, V.
573: \& Tolstoy, E., 2004, \aj, 128, 1177
574:
575: \bibitem[Vogt \etal(1994)]{vogt94} Vogt, S.~E. \etal\, 1994, SPIE, 2198, 362
576:
577: \bibitem[Wisotzki \etal(2000)]{wis00} Wisotzki, L., Christlieb, N.,
578: Bade, N.,Beckmann, V., K\"ohler, T., Vanelle, C. \& Reimers, D., 2000,
579: \aap, 358, 77
580:
581: \bibitem[Woosley \& Weaver(1995)]{woosley}
582: Woosley, S.~E. \& Weaver, T.~A., 1995, \apjs, 101, 181
583:
584: \bibitem[Yi \etal(2002)]{yi01}
585: Yi, S., Demarque, P., Kim, Y.-C. , Lee, Y.-W., Ree, C.
586: Lejeune, Th. \& Barnes, S., 2001, \apjs, 136, 417
587:
588: \end{thebibliography}{}
589:
590: \clearpage
591:
592: \begin{deluxetable}{l rrrrr | rr}
593: % \tablenum{8}
594: \tablewidth{0pt}
595: \tablecaption{Abundances for \mystar \label{table_abund}}
596: \tablehead{\colhead{Species} & \colhead{Log[$\epsilon$(X)]} &
597: \colhead{[X/H]}& \colhead{[X/Fe]} & \colhead{$\sigma$} & \colhead{Number of} &
598: \colhead{[X/Fe](0Z)\tablenotemark{a}} &
599: \colhead{[X/Fe](VLT)\tablenotemark{b}} \\
600: \colhead{} & \colhead{(dex)} &
601: \colhead{(dex)} & \colhead{(dex)} & \colhead{(dex)} &
602: \colhead{Lines} & \colhead{(dex)} & \colhead{(dex)}
603: }
604: \startdata
605: C & $<5.26$ & $<-3.33$ & $<+0.62$ & \nodata & CH
606: & \nodata & $\sim$+0.20 \\
607: N & $<5.10$ & $<-2.83$ & $<+1.12$ & \nodata & NH
608: & \nodata & \nodata \\
609: NaI\tablenotemark{c} &
610: 2.32 & $-4.00$ & $-0.05$ & 0.07 & 2 & $-0.15$ & $-$0.20 \\
611: MgI & 4.03 & $-3.51$ & +0.44 & 0.12 & 3 & +0.49 & +0.24 \\
612: AlI\tablenotemark{d} &
613: 2.35 & $-4.13$ & $-0.18$ & 0.15 & 2 & $-0.13$ & $-0.12$ \\
614: SiI & 2.59 & $-4.96$ & $-1.01$ & \nodata & 1 & +0.45 & +0.41 \\
615: CaI & 1.84 & $-4.52$ & $-0.58$ & \nodata & 1 & +0.32 & +0.27 \\
616: CaII & 2.10 & $-4.26$ & $-0.31$ & \nodata & 1 & \nodata & \nodata \\
617: ScII & $-0.95$ & $-4.05$ & $-0.10$ & \nodata & 1 & +0.13 & +0.04 \\
618: TiII & 0.85 & $-4.14$ & $-0.17$ & 0.17 & 8 & +0.29 & +0.24 \\
619: VII & $<0.64$ & $<-3.36$ & $<+0.59$ & \nodata & 1 & \nodata &
620: \nodata \\
621: CrI & 1.33 & $-4.34$ & $-0.38$ & 0.09 & 5 & $-0.45$ & $-0.46$ \\
622: MnI\tablenotemark{e} & 1.59 & $-3.80$ & +0.15 & 0.02 & 2 &
623: $-0.42$ & $-0.47$ \\
624: MnII & 1.69 & $-3.70$ & +0.25 & 0.13 & 2 &
625: \nodata & \nodata \\
626: FeI & 3.49 & $-3.96$ & 0.00 & 0.18 & 39 & 0.00 & 0.00 \\
627: FeII & 3.58 & $-3.87$ & +0.09 & 0.22 & 4 & 0.00 & 0.00 \\
628: CoI & 1.98 & $-2.94$ & +1.01 & 0.21 & 4 & +0.50 & +0.40 \\
629: NiI & 2.52 & $-3.73$ & +0.22 & 0.01 & 2 & $-0.08$ & $-0.04$ \\
630: CuI & $-0.41$ & $-4.62$ & $-0.67$ & \nodata & 1 & \nodata & \nodata \\
631: SrII & $<-2.75$ & $<-5.65$ & $<-1.70$ & \nodata & 2
632: & \nodata & \nodata \\
633: YII & $<-1.46$ & $<-3.70$ & $<+0.25$ & \nodata & 2
634: & \nodata & \nodata \\
635: BaII\tablenotemark{f} & $-2.74$ & $-4.87$ & $-0.92$ & \nodata & 1
636: & \nodata & \nodata \\
637: EuII & $<-1.95$ & $<-2.46$ & $<+1.49$ & \nodata & 1
638: & \nodata & \nodata \\
639: \enddata
640: \tablenotetext{a}{Regression lines for C-normal giants from our 0Z survey,
641: evaluated at $-4.0$~dex.}
642: \tablenotetext{b}{Regression lines from \cite{cayrel04}, Table~9, evaluated
643: at $-4.0$~dex.}
644: \tablenotetext{c}{Non-LTE correction of $-$0.2 dex has been applied for
645: [Na/Fe] from the Na~D lines.}
646: \tablenotetext{d}{Non-LTE correction of +0.6 dex has been applied for
647: [Al/Fe] from the 3950~\AA\ doublet.}
648: \tablenotetext{e}{ The
649: adjustment of +0.4~dex for the 4030~\AA\ Mn~I triplet
650: suggested by \cite{cayrel04} and by our own work has been applied here.}
651: \tablenotetext{f}{The Ba II line at 4554\AA\ is only
652: marginally detected and therefore the Ba abundance may also
653: be interpreted as an upper limit.}
654: \end{deluxetable}
655:
656: \clearpage
657:
658:
659:
660:
661: \begin{figure}
662: \epsscale{0.7}
663: % Comment out the following line to embed the PS figure into the manuscript
664: % \plotone{/scr2/jlc/hamburg_survey/paper_lowest_2006/plots/he1424_sica.ps}
665: \plotone{f1.ps}
666: \caption[]{[Si/Fe] (upper panel) and
667: [Ca/Fe] (lower panel) are shown as a function of [Fe/H] for EMP giants with
668: [Fe/H] $< -2.4$~dex. Filled circles denote HES stars from our 0Z
669: project and star symbols are giants from the First Stars VLT project \citep{cayrel04}.
670: \mystar\ is shown as the large filled circle, and is the only outlier,
671: being very low in both panels. The dashed horizontal lines represent
672: the Solar abundance ratios.
673: \label{figure_sica_feh}}
674: \end{figure}
675:
676: \end{document}
677:
678: ===================
679: No room for this figure
680:
681: \begin{figure}
682: \epsscale{0.9}
683: % Comment out the following line to embed the PS figure into the manuscript
684: \plotone{/scr2/jlc/hamburg_survey/paper_lowest_2006/plots/si_he1424.ps}
685: % \plotone{f2.ps}
686: \caption[]{The region of the Si~I line at 3905~\AA\ is shown in HE~1424$-$0241 and
687: in a comparison star from our work with
688: the same \teff. Although the Fe-abundance
689: is roughly 4 times higher in the comparison star, the ratio of line strengths
690: clearly demonstrates that [Si/Fe] is abnormally low
691: in \mystar.
692: \label{figure_si3905}}
693: \end{figure}
694: