1: \documentclass[12pt,preprint]{aastex}
2: %\usepackage{emulateapj}
3: % Version 1.0 (2008.3.4)
4:
5: \begin{document}
6: \def\etal{{\sl et al.~}}
7: \def\Teff{{$\rm{T_{eff}}$}}
8: \def\dTeff{{$\langle \Delta \rm{T_{eff}} \rangle$}}
9:
10: \title{Chemical Abundances in 35 Metal-Poor Stars. I. Basic Data}
11:
12: \author{Jeong-Deok Lee\altaffilmark{1, 2},
13: Sang-Gak Lee\altaffilmark{2},
14: \& Kang-Min Kim\altaffilmark{3}}
15:
16: \email{leejd@astro.snu.ac.kr}
17:
18: \altaffiltext{1}{Astrophysics Research Center for the Structure
19: and Evolution of the Cosmos (ARCSEC), Sejong University, Seoul,
20: 143-747, South Korea}
21:
22: \altaffiltext{2}{Department of Physics and Astronomy, Seoul
23: National University, Seoul 151-742, South Korea}
24:
25: \altaffiltext{3}{Korea Astronomy \& Space Science Institute, 61-1
26: Whaam-dong, Youseong-gu, Taejeon 305-348, South Korea}
27:
28: \received{updated 1/Aug/2007}
29:
30: \shorttitle{Chemical Abundances in Metal-Poor Stars}
31: \shortauthors{Lee, Lee, \& Kim }
32:
33: \begin{abstract}
34: We carried out a homogeneous abundance study for various elements,
35: including $\alpha$-elements, iron peak elements and $n$-capture
36: elements for 35 metal-poor stars with a wide metallicity range
37: ($-3.0\lesssim$[Fe/H]$\lesssim-0.5$). High-resolution
38: ($R\simeq30$k), high signal-to-noise($S/N\geq110$) spectra with a
39: wavelength range of 3800 to 10500 \AA using the Bohyunsan Optical
40: Echelle Spectrograph (BOES). Equivalent widths were measured by
41: means of the Gaussian-fitting method for numerous isolated weak
42: lines of elements. Atmospheric parameters were determined by a
43: self-consistent LTE analysis technique using Fe I and Fe II lines.
44: In this study, we present the EWs of lines and atmospheric
45: parameters for 35 metal-poor stars.
46:
47: \end{abstract}
48:
49: \keywords{stars: abundance - stars: population II - Galaxy:
50: abundance - Galaxy: evolution }
51:
52: \section{INTRODUCTION}
53:
54: The abundance patterns of metal-poor stars provide information on
55: the formation, evolution, and supernova explosions of massive
56: stars formed in the early epoch of the Galaxy formation
57: \citep[e.g.][]{tinsley80, pagel97,matteu01}. They also provide
58: important clues for understanding of the early stage of the
59: chemical evolution of the Galaxy. Many high-resolution
60: spectroscopic abundance studies for metal-poor stars have been
61: carried out \citep[e.g.][]{mac95, rnb96, ful00, burris00,
62: gratton03, cayrel04,honda04, aoki05, francois07} and their results
63: have shown that the abundance patterns of metal-poor stars differ
64: from those of the Sun. In particular, it has been established that
65: the ratio of $\alpha$-elements to iron ratio in stars with
66: [Fe/H]$\lesssim-1.0$ is greater than the solar ratio by $\sim+0.4$
67: dex \citep{wheeler89}. This indicates the complex nature of the
68: chemical enrichment.
69:
70: The general explanation for the chemical evolution of the Galaxy
71: has been based on the products of the two main types of supernovae
72: \citep{tinsley80}. The first, comprising the Type II SN and Type
73: Ib/Ic SNe which are the core-collapse-induced explosions of
74: massive stars. The second, the Type Ia SN is the thermonuclear
75: explosion of an accreting white dwarf in a close binary system.
76: The core-collapse SNe breakout shortly after early Galaxy
77: formation because their progenitors are massive stars with short
78: lifetimes, while Type Ia SNe occur after completing at least one
79: stellar lifetime of $\sim10^9$yr(for an 8$M_\odot$ star) resulting
80: in the formation of a white dwarf \citep{chiosi92}.
81: $\alpha$-elements are the main products of Type II and Type Ib/Ic
82: SNe while Fe is mainly produced by Type Ia SNe. Therefore the
83: ratio of $\alpha$-elements to Fe ratio is high in the halo stars
84: that were formed in the early stage of the Galaxy however this
85: ratio reduces to the solar value as observed in the trends of
86: $\alpha$-elements along metallicity.
87:
88: However, the primary nucleosynthetic sources for most elements are
89: not well known. A relative abundance ratio, for instance, the
90: ratio of an element to Fe is a good tool for investigating the
91: product site of each element. However the inhomogeneous abundance
92: results of previous studies, which showed large scatters and/or
93: some shifts, rule out the use of this technique for this purpose.
94: In order to determine the abundance of a given element, the
95: atmospheric parameters (temperature \Teff, surface gravity $\log
96: g$, micro-turbulent velocity $\xi_v$, and overall metallicity
97: represented by [Fe/H]) need to be derived. Therefore differences
98: in these parameters result in differences in the results of
99: abundance analysis. The choice of the lines used for analysis and
100: the oscillator strength value ($\log gf$) adopted for deriving the
101: abundance also results in uncertainties and systematic differences
102: in the abundance results. A single-lin approximation is found to
103: be in adequate for some spectral lines of odd atomic number
104: elements with significant hyperfine structures \cite{rnb96}. It
105: has been shown that ignoring the hyperfine structure could lead to
106: abundance errors of up to 0.6 dex. Therefore, we aim to obtain
107: abundances of elements in stars with wide metallicity ranges as
108: accurately as possible by using homogeneous methods.
109:
110: We divided the elements into three groups, namely, light elements
111: ($Z\le20$), iron-peak elements ($21\le Z\le30$), and $n$-capture
112: elements ($Z\ge38$). Light elements are good tools for judging the
113: surface chemical contamination due to the internal mixing in
114: giants. Iron-peak elements which are mainly produced by explosive
115: Si-burning can give important information on the supernova
116: explosions. The iron-peak elements particularly in metal-poor
117: stars with [Fe/H]$\le-1.0$ are produced in massive core-collapse
118: supernovae \citep{tinsley80} and provide important constraints for
119: models of type II SNe.
120:
121: The $n$-capture process is responsible for the production of heavy
122: elements beyond the iron-peak elements, however, its site has not
123: yet been determined with certainty. In particular, the
124: nucleosynthesis process of light $n$-capture elements is proposed
125: to be a combination of various processes \citep{montes07}.
126: Further, radioactive elements with sufficiently long half-lives,
127: such as uranium (U, Z=92) and thorium (Th, Z=90) can provide a
128: constraint on the age of metal-poor stars. In order to determine
129: the age of old stars with unstable radioactive elements, a stable
130: element synthesized in the same event is required to be set as
131: reference.
132:
133: The transition probabilities of rare-earth $n$-capture elements
134: and their stable isotopes such as La, Sm, Eu, and Hf have recently
135: been measured with very high accuracy by Lawler and associates
136: \citep[e.g.][]{hartog, lawler01a, lawler01b, lawler07}. This has
137: enabled the derivation of more reliable $n$-capture element
138: abundance, which may provide new insights on the roles of the
139: $n$-capture process in the initial burst of Galactic
140: nucleosynthesis.
141:
142:
143: \section{Observations and reductions}
144:
145: \subsection{Observations}
146: %%%% Target Selection %%%%%%%%
147: We selected the target stars for our study based on the following
148: criteria. Firstly, due to the small aperture (1.8 m) of Bohyunsan
149: Optical Astronomical Observatory(BOAO), we selected stars with a
150: visual magnitude brighter than 10 mag. Secondly, metal-poor stars
151: whose metallicity was determined as [Fe/H]$\le-1.0$ by previous
152: spectroscopic studies were selected. Thirdly, stars that are known
153: to have a high proper-motion but do not have previous abundance
154: studies were also included in our target set. We excluded K- and
155: M-type stars that have strong molecular lines as well as metallic
156: lines that are too strong and blended with other lines.
157:
158: We have obtained high-resolution echelle spectra for 35 metal-poor
159: stars using the Bohyunsan Optical Echelle Spectrograph (BOES)
160: \citep{kim02} in two runs scheduled in November 2005 and April
161: 2006.
162: %%%%%%%%%%% BOES description
163: BOES is a fiber-fed echelle spectrograph mounted on a 1.8-meter
164: telescope at BOAO (Bohyunsan Optical Astronomical Observatory).
165: The BOES detector is an EEV-CCD with $2048\times4102$ pixels and a
166: pixel size of $15\mu m$. The BOES covers a wide wavelength
167: range(3800 to 10500\AA) in a single exposure and has a high
168: efficiency($\sim15\%$). There are three kinds of fiber sets, with
169: apertures of 80$\mu m(1.1\arcsec)$, 200$\mu m(2.9\arcsec)$, and
170: 300$\mu m(4.2\arcsec)$ corresponding to spectral resolving powers
171: of $R\simeq90000$, $44000$, and $30000$, respectively.
172:
173: We observed most of the targets with the largest aperture(300$\mu
174: m$) in order to minimize light loss and obtain spectra with high
175: signal-to-noise ratio because of the poor weather conditions at
176: BOAO(The seeing condition was $\sim3\arcsec$ in the first run
177: while that on the second run was $\sim4\arcsec$) during the
178: observing runs. Six stars(HD 94028, HD93487, HD 101227, HD 118659,
179: HD 148816, and HD159482) were observed using an aperture of
180: 200$\mu m(2.9\arcsec)$, which gives a better resolution. We
181: observed no significant difference on comparing the spectra
182: observed from different apertures. A resolving power of $R\sim30k$
183: was found to be satisfactory for conducting abundance analysis. We
184: also observed B-type stars on every night to obtain telluric line
185: references.
186:
187: We obtained bias frames for zero-correction and tungsten-halogen
188: lamp(THL) frames for flat-fielding, and Th-Ar spectra for
189: wavelength calibration. Since dark current of BOES CCD is very
190: small($\sim2e^- $/hr/pixel) relative to read-out
191: noise($\leq3e^-$), it is not necessary to obtain dark frames. The
192: observation log is presented in Table.\ref{table:log}
193:
194:
195: \subsection{Reductions}
196: The reductions were performed by following the standard steps for
197: echelle spectrum using IRAF. The biases were first corrected by
198: subtracting the median combined bias frame using the IRAF/ccdred
199: task. The dark currents were not corrected because the dark
200: current of the BOES CCD is negligibly small.
201:
202: We adapted a technique for aperture extractions and flattening
203: that differs slightly from the standard reduction procedures. We
204: first prepared a master flat \emph{image} by combining
205: preprocessed flat images. This master flat image is used only as
206: aperture tracing reference. Using this master flat image, we
207: defined apertures and background regions to be used for scattered
208: light subtraction using the IRAF/ECHELLE/apall task. Subsequently,
209: flat spectra and object spectra were extracted. We then produced
210: the master flat\emph{spectrum} by combining the flat spectra using
211: the IRAF/ECHELLE/scombine task.
212:
213: The flat-field correction was carried out by dividing each
214: extracted order of the object spectra by the corresponding order
215: of the master flat \emph{spectrum}. The blaze function is also
216: corrected using this procedure. Since the THL lamp is not a
217: perfect flat source, the flux of the flat-corrected spectra
218: becomes distorted, however this distortion can be corrected by
219: continuum normalization. Compared to the standard echelle
220: reduction procedure, this procedure is faster and simple and it
221: corrects the blaze function prior to the continuum normalization.
222: For the case wherein the information of the flux is not required
223: and only the continuum-normalized spectrum is needed--usually for
224: equivalent width(EW) measurements--this procedure would be more
225: efficient for the reduction process.
226:
227: The wavelength calibration function was obtained using the
228: IRAF/ECHELL/ecidentify task based on the Th-Ar lamp spectrum
229: corresponding to the first night of the observing run. The daily
230: variation in the wavelength of BOES was very small($<0.001\AA$),
231: therefore, the IRAF/ECHELL/ecreidentify task was utilized for the
232: remaining nights of the run. Figure.\ref{fig:sample} shows a small
233: section of the sample spectra around the $H\gamma$ line, with a
234: wavelength range of 4320 to 4350 \AA~for seven stars whose
235: metallicities [Fe/H] range from $-0.63$ to $-2.26$ dex.
236:
237:
238: \subsection{Equivalent Widths}
239: Equivalent widths for most elements were measured using Gaussian
240: fitting in the IRAF/splot task. Local continuum was defined by
241: visual inspection for each line. Weak Fe I and Fe II lines with
242: log(EW/$\lambda$)$< -4.9$ or EWs less than 100m\AA~were used to
243: determine the atmospheric parameters and to carry out abundance
244: analysis, as strong lines lie in the over-saturation region of the
245: curve of growth, where Gaussian fitting is inadequate and an
246: accurate damping constant is required. However, when only strong
247: lines were available for abundance analysis, we measured EWs by
248: means of voigt profile fitting. The sources of adopted atomic
249: information are listed in Table.\ref{table:loggf}. A small sample
250: of measured equivalent widths, along with the corresponding atomic
251: information, is presented in Table.\ref{table:ew}.
252:
253: We compared the measured EWs with those of previous studies. A
254: graphical comparison for the common stars is shown in Figure.
255: \ref{fig:ew_comp1} and Figure. \ref{fig:ew_comp2}. Four stars,
256: namely, BD+29 366, HD 19445, HD 64090, and HD 84937 are common
257: between our study and \cite{gratton03}. The mean differences (this
258: work minus \cite{gratton03}) are $0.50~\pm~0.23~$m\AA~(sdom
259: \footnote{sdom, standard deviation of mean
260: $=\sqrt{\frac{\sigma^2}{N}}$}, 320 lines) for BD+29 366,
261: $0.03~\pm~ 0.29~$m\AA(110 lines) for HD 19445,
262: $0.46~\pm~0.46~$m\AA~(96 lines) for HD 64090, and
263: $-1.36~\pm~0.57~$m\AA~ (20 lines) for HD 84937. Figure.
264: \ref{fig:ew_comp2} shows the comparison of EWs of HD 122563 with
265: those of \cite{cayrel04} and \cite{honda04}. The mean differences
266: corresponding to \cite{cayrel04} and \cite{honda04} are
267: $0.25~\pm~0.27~$m\AA~(78 lines) and $0.33~\pm~0.18~$m\AA~(67
268: lines), respectively. There is no significant offset observed in
269: the measurement of EWs.
270:
271: The expected EW uncertainty derived from Cayrel's formula (1988)
272: for our spectra with $S/N\sim200$ is $\sim2$m\AA. This uncertainty
273: ignores the measurement errors that occur in continuum definition
274: and result from the departure of the Gaussian profile of the
275: observed line. For weak lines, the departure from Gaussian line
276: profile is negligible, while the major error in EW measurements
277: results from the local continuum definition. We repeated the EW
278: measurements for some well isolated lines by setting different
279: local continuum to obtain an error of less than 4m\AA, however, we
280: found that the error in EW depends on the line strength and
281: signal-to-noise ratio of a spectrum.
282:
283:
284: \subsection{Radial Velocities}
285:
286: Radial velocities were measured from the observed centers of the
287: Fe I and Fe II lines that are well-isolated and have well-defined
288: Gaussian profiles. The distribution of radial velocities derived
289: from each line for a star was Gaussian-fitted to obtain the
290: stellar velocity, as Figure. \ref{fig:rad_vel}(a). The radial
291: velocities for program stars are presented in the last column of
292: Table.~\ref{table:atpar}. The typical error of radial velocitiy
293: ($2\sigma$ dispersion in the Gaussian profile) is $\sim0.45$ km/s.
294: A comparison of the radial velocities obtained in this work and
295: those of previous studies \citep{geneva04, latham02} is presented
296: in Figure. \ref{fig:rad_vel}(b) in order to demonstrate the a good
297: agreement between the two.
298:
299: The mean offset is $+0.51~\pm~0.66$ km/s (sdom, N=31), indicted as
300: a solid line in the lower panel of Figure.\ref{fig:rad_vel}(b),
301: the dotted line indicates a $2\sigma$ dispersion. Among the 35
302: stars under consideration, two stars, namely, HD 6755 and BD+17
303: 4708, exhibit velocity differences greater than the 2$\sigma$
304: dispersion, while HD 8724, HD 165908, and BD+29 366 exhibit the
305: velocity differences close to the $2\sigma$ dispersion. These are
306: indicated as filled squares in the lower panel of Figure
307: \ref{fig:rad_vel}(b). BD+17 4708 and HD 6755 are known as binaries
308: in previous radial velocity studies of \cite{latham02} and
309: \cite{carney03}, and our radial velocities are in good agreement
310: with the radial velocity curves calculated from their orbital
311: solutions. HD 165908 is also a star in a binary system(b Her).
312: BD+29 366 is a suspected binary with an unknown period
313: \citep{latham02}. Only one previous radial velocity study
314: \citep{barbier89} exists in relation to HD 8724. Considering its
315: error of 1 km/s and a velocity difference of approximately 2.5
316: km/s from our results, we suspect that HD 8724 may be a binary
317: system, which requires further radial-velocity study.
318:
319:
320: \section{Atmospheric parameters}
321:
322: In order to derive elemental abundances, an appropriate model
323: atmosphere must be constructed prior to conducting abundance
324: analysis. We used Kurucz model atmosphere grids computed with the
325: new opacity distribution function(ODF)(http://kurucz.harvard.edu).
326: A Linux ported ATLAS program \citep{sbordone} was used to
327: interpolate model atmospheres for the atmospheric model between
328: grid points. For the metal-poor stars with [Fe/H]$<-0.5$,
329: $\alpha$-element enhanced models ([$\alpha$/Fe]=+0.4) were used.
330: The adopted $\log gf$ values for the iron lines are presented in
331: Table.~\ref{table:ew}. For the determination of atmospheric
332: parameters, we used only the Fe I and Fe II lines weaker than
333: log(EW/$\lambda$) $<-4.90$ or EW $< 100$m\AA~in order to avoid the
334: saturated lines. The LTE abundance analysis program, MOOG,
335: \citep{sneden73} was used to derive Fe I and Fe II abundances.
336:
337: We determined the atmospheric parameters by following a
338: self-consistent LTE analysis technique. First, the value of
339: \Teff~was adjusted so that there is no trend between the iron
340: abundances from each of the Fe I lines and their lower level
341: excitation potentials. The micro-turbulent velocity($\xi_v$) was
342: then determined to exhibit no correlation with the Fe I line
343: strength and the iron abundance derived from each lines. Finally,
344: the value of log$g$~ was adjusted so that the iron abundance
345: derived from the Fe I lines matches the iron abundance given from
346: the Fe II lines. This process was iterated until the determined
347: parameters (\Teff, log$g$, and $\xi_v$) satisfied the above
348: criteria simultaneously. We used more than 100 Fe I lines and a
349: few tens of Fe II lines to determine the atmospheric parameters.
350: Figure. \ref{fig:fe_abun} shows the Fe abundance of HD 122563 with
351: the optimal atmospheric parameters to be set as \Teff$= 4430$K,
352: $\log g=0.58$, $\xi_v=2.15$km/s, and $\log\epsilon_{Fe}=4.68$. The
353: derived atmospheric parameters derived for our program stars are
354: presented in Table \ref{table:atpar}.
355:
356: We compared the atmospheric parameters derived in our study with
357: those obtained in other studies for common objects. This is
358: depicted in Figure \ref{fig:atcomp}, where each symbol represents
359: a comparison of our atmospheric parameters with those obtained in
360: other literature, using the following key. A filled square
361: represents comparison with \cite{go84}, a plus symbol indicates
362: comparison with \cite{sgc91}, open square, \cite{gs94}, open
363: circle, \cite{ful00}, filled triangle, \cite{mk01}, open triangle,
364: \cite{gratton03}, open star, \cite{honda04}, filled star,
365: \cite{cayrel04}, and filled circle, \cite{johnsell05}. We have
366: examined whether a systematic difference exists in the atmospheric
367: parameters depending on the method employed. The atmospheric
368: parameters in \cite{ful00} were derived in essentially the same
369: manner as in our study, with the exception that the metal
370: abundance of the model atmosphere was enhanced in \cite{ful00} in
371: order to compensate for the increased electron density caused by
372: $\alpha$-element enhancement in metal-poor stars.
373:
374: As shown in Figure. \ref{fig:atcomp} (a), the effective
375: temperatures obtained in this work are in good agreement with
376: those of other studies. The \dTeff~ or the mean of
377: $\Delta$\Teff=\Teff~(this study)$-$\Teff~(other studies) between
378: our study and \cite{ful00}, wherein \Teff~was determined using the
379: same method, is $+54\pm26K$ (sdom, N=26). The value of \dTeff~for
380: \cite{mk01}, wherein \Teff~was derived using wing profile fitting
381: of H$\alpha$, is $+37\pm29K$ (sdom, N=14) and that for
382: \cite{gratton03}, wherein \Teff~was obtained by employing a
383: photometric method using (\bv) and $(b-y)$ color index, is $-59\pm
384: 26K$ (sdom, N=7). There are no significant differences in
385: \Teff~based on the different methods employed. In general, errors
386: of up to $100K$ in effective temperature are acceptable.
387:
388:
389: Figure. \ref{fig:atcomp} (b) shows the comparison of the log $g$
390: values obtained in this study with those obtained in other
391: studies. The mean offset in log $g$ is $+0.08\pm0.05$ in the case
392: of \cite{ful00} and $+0.11\pm0.06$ in \cite{mk01}, wherein surface
393: gravity was determined using the ionization balance between Fe I
394: and Fe II lines. In \cite{gratton03}, surface gravities were
395: obtained using absolute visual magnitudes deduced from Hipparcos
396: parallexes and masses obtained by interpolating the position of
397: the star along the isochrones. The typical values of their errors
398: in surface gravity, resulting mainly from uncertainties in the
399: parallexes, is approximately $\pm0.1$ dex. The mean difference in
400: surface gravities in \cite{gratton03} is $-0.08\pm0.08$ dex. Thus
401: no significant differences are observed. The non-LTE effect may
402: result in a difference between the surface gravities determined by
403: the photometric method and the spectroscopic method, which is
404: based on the LTE assumption involving ionization balance. This
405: non-LTE effects are expected to be significant under the low
406: surface gravity condition. Comparison of common objects common
407: between this study and \cite{gratton03}, whose program stars are
408: subdwarfs or early subgiants, shows no significant difference.
409:
410: Figure. \ref{fig:atcomp} (d) present a comparison of the iron
411: abundance obtained in this study with that obtained in other
412: studies. While \cite{ful00} used the same method for deriving
413: atmospheric parameters, the value of [Fe/H] in the atmospheric
414: model used in \cite{ful00} was set marginally higher than the
415: derived iron abundance in order to compensate for the electron
416: density due to $\alpha$-elements enhancement in metal-poor stars.
417: Using an $\alpha$-enhanced ($+0.4$dex) model in our study, we
418: compared the [Fe/H] value of our model to that of \cite{ful00} and
419: observed no difference ($0.00\pm0.02$~dex). There are no
420: significant systematic errors in [Fe/H], however, our results are
421: metal deficient by $0.06\pm0.03$~dex as compared to
422: \cite{gratton03}, and by $0.10\pm0.03$~dex as compared to
423: \cite{mk01}.
424:
425: \section{Summary}
426: We have obtained high-resolution ($R\simeq30$k), high
427: signal-to-noise($S/N\geq100$) spectra of 35 metal-poor stars using
428: the BOES at BOAO, with a wavelength range of 3,800 to 10,500\AA.
429:
430: EWs were measured using the Gaussian fitting method, and the local
431: continuum for each line was determined by visual inspection. In
432: cases where only strong lines were available for a given element,
433: EWs were measured by the voigt profile fitting method. Thus, we
434: presented the EWs of various elements. The measured EWs are in
435: strong agreement with those of previous studies.
436:
437: Radial velocities were determined with an error of $\sim 0.45$
438: km/s from the Fe lines that were isolated and well-defined as
439: Gaussian profiles. HD 8724 is suspected to be a binary system. The
440: atmospheric parameters were determined by following a
441: self-consistent LTE analysis technique.
442:
443: \Teff~was chosen such that Fe I lines of different excitation
444: potentials yield the same abundance and surface gravity was
445: determined from the ionization balance of Fe I and Fe II. The
446: micro-turbulent velocity was set to exhibit no correlation between
447: Fe abundance and the strength of the Fe I line. Model atmospheres
448: were constructed using Kurucz LTE atmospheric model calculated
449: with the new ODF. The atmospheric parameters we obtained are in
450: good agreement with those of previous studies despite the other
451: studies adapting methods different from ours to obtain these
452: parameters.
453:
454:
455:
456: \begin{thebibliography}{}
457: \bibitem[Aoki \etal(2005)]{aoki05}
458: Aoki, W. \etal 2005, \apj, 632, 611
459:
460: \bibitem[Barbier-Brossat(1989)]{barbier89}
461: Barbier-Brosat, M. 1989, \aaps, 80, 67
462:
463: \bibitem[Bard \& Kock(1994)]{bard94}
464: Bard, A., Kock, M. 1994, \aap, 282, 1014
465:
466: \bibitem[Bard \etal(1991)]{bard91}
467: Bard, A., Kock, A., \& Kock, M. 1991, \aap, 248, 315
468:
469: \bibitem[Bielsky(1975)]{bielsky75}
470: Bielsky, A. 1975, J. Quant. Spectrosc. Radiat. Transfer, 15, 463
471:
472: \bibitem[Bi\'{e}mont \etal(1991a)]{biemont91}
473: Bi\'{e}mont, E., Hibbert, A., Godefroid, M., Vaeck, N., \&
474: Fawcett, B. C. 1991a, \apj, 375, 818
475:
476: \bibitem[Bi\'{e}mont \etal(1991b)]{biemont91b}
477: Bi\'{e}mont, E., Baudoux, M., Kurucz, R. L., Ansbacher, W., \&
478: Pinnington, E. H. 1991b, \aap, 249, 539
479:
480: \bibitem[Bi\'{e}mont \etal(1981)]{biemont81}
481: Bi\'{e}mont, E., Grevesse, N., Hannaford, P., \& Lowe, R. M. 1981,
482: \apj, 248, 867
483:
484: \bibitem[Bizzarri \etal(1993)]{bizzarri93}
485: Bizzarri, A., Huber, M. C. E., Noels, A., Grevesse, N., Bergeson,
486: S. D., Tsekeris, P., \& Lawler, J. E. 1993, \aap, 273, 707
487:
488: \bibitem[Blackwell \etal(1980)]{blackwell80}
489: Blackwell, D. E., Shallis, M. J., \& Simmons, G. J. 1980, \aap,
490: 81, 340
491:
492: \bibitem[Booth \etal(1984)]{booth84}
493: Booth, A. J., Blackwell, D. E., Petford, A. D., \& Shallis, M. J.
494: 1984, \mnras, 208, 435
495:
496: \bibitem[Burris \etal(2000)]{burris00}
497: Burris, D. L., Pilachowsk, C. A., Armandroff, T. A., Sneden, C., Cowan,
498: J. J., \& Roe, H. 2000, \apj, 544, 302
499:
500: \bibitem[Cardon \etal(1982)]{cardon82}
501: Cardon, B. L., Smith, P. L., Scalo, J. M., \& Testerman, L. 1982,
502:
503: \bibitem[Carney \etal(2003)]{carney03}
504: Carney, B. W., Latham, D. W., Stefanik, R. P., Laird, J. B., \&
505: Morse, J. A. 2003, \aj, 125, 293
506:
507: \bibitem[Cayrel \etal(2004)]{cayrel04}Cayrel, R. \etal 2004, A\&A,
508: 416, 1117
509:
510: \bibitem[Chiosi \etal(1992)]{chiosi92}
511: Chiosi, C., Bertelli, G., \& Bressan, A. 1992, ARAR, 30, 235
512:
513: \bibitem[Den Hartog \etal(2006)]{hartog}
514: Den Hartog, E. A., Lawler, J. E., Sneden, C., \& Cowan, J. J.
515: 2006, \apj, 167, 292
516:
517: \bibitem[Den Hartog \etal(2003)]{hartog03}
518: Den Hartog, E. A., Lawler, J. E., Sneden, C., \& Cowan, J. J.
519: 2003, \apjs, 148, 543
520:
521: \bibitem[Ficsher(2002)]{fischer02}
522: Fischer, C. F. 2002, http://www.vuse.vanderbilt,edu/~cff
523:
524: \bibitem[Fields \etal(2002)]{fields02}
525: Fields, B. D., Truran, J. W., \& Cowan, J. J. 2002, \apj, 575, 845
526:
527: \bibitem[Fran\c{c}ois \etal(2007)]{francois07}
528: Fran\c{c}ois, P. \etal 2007, \aap, 476, 935
529:
530: \bibitem[Fulbright(2000)]{ful00}Fulbright, J. P. 2000, AJ,
531: 120, 1841
532:
533: \bibitem[Gilroy \etal(1988)]{gilroy88}
534: Gilroy, K. K., Sneden, C., Pilachowski, C. A., \& Cowan, J. J.
535: 1988, \apj, 327, 298
536:
537: %\bibitem[Goswami \& Prantzos(2000)]{goswami00} Goswami, A. \&
538: %Prantzos, N. 2000, \aap, 359, 191
539:
540: \bibitem[Gratton \& Ortolani(1984)]{go84}
541: Gratton, R. G. \& Ortolani, S. 1984, \aap, 137, 6
542:
543: \bibitem[Gratton \& Sneden(1994)]{gs94}
544: Gratton, R. G. \& Sneden, C. 1994, \aap, 287, 927
545:
546: \bibitem[Gratton \etal(2003)]{gratton03}Gratton, R. G. \etal 2003,
547: A\&A, 404, 187
548:
549: \bibitem[Hannaford \etal(1992)]{hannaford92}
550: Hannaford, P., Lowe, R. M., Grevesse, N., \& Noels, A. 1992, \aap,
551: 259, 301
552:
553: \bibitem[Hannaford \etal(1982)]{hannaford82}
554: Hannaford, P., Lowe, R. M., Grevesse, N., Bi\'{e}mont, E., \&
555: Whaling, W. 1982, \apj, 261, 736
556:
557: %\bibitem[Heger \& Woosley(2002)]{}
558: %Heger, A.\& Woosley, S. E. 2002, \apj, 567, 532
559:
560: \bibitem[Holweger \etal(1990)]{holweger90}
561: Holweger, H., Heise, C., \& Kock, M. 1990, \aap, 232, 510
562:
563: \bibitem[Honda \etal(2004)]{honda04}Honda, S. \etal 2004, \apj, 607,
564: 474
565:
566: \bibitem[Johnsell \etal(2005)]{johnsell05}
567: Johnsell, K. \etal 2005, \aap, 440,321
568:
569: \bibitem[Kim \etal(2002)]{kim02}
570: Kim, K.-M., \etal 2002, JKAS, 35, 221
571:
572: \bibitem[Kurucz(1995)]{kurucz95}
573: Kurucz, R. L. 1995, Kurucz CD-ROM No.23 (Harvard-Smithsonian
574: Center for Astrophysics)
575:
576: \bibitem[Kupka \etal(1999)]{kupka99}
577: Kupka, F., Piskunov, N. E., Ryabchikova, T. A., Stemples, H. C.,
578: \& Weiss, W. W. 1999, \aaps, 138, 119
579:
580: \bibitem[Kusz(1992)]{kusz92}
581: Kusz, J. 1992, \aaps, 92, 517
582:
583: \bibitem[Latham \etal(2002)]{latham02}
584: Latham, D. W. \etal 2002, \aj, 124, 1144
585:
586: \bibitem[Lawler \etal(2001a)]{lawler01a}
587: Lawler, J. E., Bonvallet, G., \& Sneden, C. 2001a, \apj, 556, 452
588:
589: \bibitem[Lawler \etal(2001b)]{lawler01b}
590: Lawler, J. E., Wickliffe, M. E., Cowley, C. R., \& Sneden, C.
591: 2001b, \apjs, 137, 341
592:
593: \bibitem[Lawler \etal(2007)]{lawler07}
594: Lawler, J. E., Den Hartog, E. A., Labby, Z. E., Sneden, C., Cowan,
595: J. J. and Ivans, I. I.2007, \apjs, 169, 120
596:
597: \bibitem[Lawler \etal(2006)]{lawler06}
598: Lawler, J. E., Den Hartog, E. A., Sneden, C., \& Cowan, J. J.
599: 2006, \apjs, 162, 227
600:
601: \bibitem[Matteucci(2001)]{matteu01}
602: Matteucci, F. 2001, The Chemical Evolution of the Galaxy, Kluwer
603: Academic Publishers, Dordrecht
604:
605: \bibitem[McWilliam \etal(1995)]{mac95}McWilliam, A., Preston,
606: G. W., Sneden, C., \& Searle, L. 1995, AJ, 109, 2757
607:
608: \bibitem[McWilliam(1998)]{Mcwilliam98}
609: McWilliam, A. 1998, \aj, 115, 1640
610:
611: %\bibitem[McWilliam(1997)]{mcwil97}
612: %McWilliam, A., 1997, ARA\&A, 35,503
613:
614: \bibitem[Mendoza \etal(1995)]{mendoza95}
615: Mendoza, C., Eissner, W., Zeippen, C. J, \& Galavis, M. E. 1995,
616: http://legacy.nasa.gov/topbase
617:
618: \bibitem[Mishenina \& Kovtyukh(2001)]{mk01}
619: Mishenina, T. V. \& Kovtyukh, V. V. 2001, \aap, 370, 951
620:
621: \bibitem[Montes \etal(2007)]{montes07}
622: Montes, F., Beers, T. C., Cowan, J., Elliot, T., Farouqi, K.,
623: Gallino, R., Heil, M., Kratz, K.-L., Pfeiffer, B., Pignatari, M., \&
624: Schatz, H. 2007, 671, 1685
625:
626: \bibitem[Nordstr\"{o}m \etal(2004)]{geneva04}
627: Nordstr\"{o}m, B. \etal 2004, A\&A, 418, 989
628:
629: \bibitem[O'Brian \& Lawler(1991)]{obrian91}
630: O'Brian, T. R. \& Lawler, J. E. 1991a, Phys. Rev. A, 44, 7134
631:
632: \bibitem[O'Brian \etal(1991)]{obrian91b}
633: O'Brian, T. R., Wickliffe, M. E., Lawler, J. E., Whaling, J. W. \&
634: Brault, W. 1991b, JOSA B, 9, 1185
635:
636: \bibitem[Pagel(1997)]{pagel97}
637: Pagel, B. E. J. 1997, Nucleosynthesis and Chemical Evolution of
638: Galaxies (Cambridge Unv. Press)
639:
640: \bibitem[Palmeri \etal(2000)]{palmeri00}
641: Palmeri, P., Quinet, P., Wyart, J.-F., \& Bi\'{e}mont, \'{E}.
642: 2000, Phys. Scr., 61, 323
643:
644: \bibitem[Ryan, Norris, \& Beers(1996)]{rnb96}
645: Ryan, S. G., Norris, J. E., \& Beers, T. C. 1996, \apj, 471, 254
646:
647: %\bibitem[Samland(1998)]{samland98} Samland, M. 1998, \apj, 496,
648: %155
649:
650: \bibitem[Sbordone \etal(2004)]{sbordone} Sbordone, L., Bonifacio, P.,
651: Castelli, F., \& Kurucz, R. L. 2004, MSAIS, 5, 93
652:
653: \bibitem[Smith \& Raggett(1981)]{smith81}
654: Smith, G., \& Raggett, S. St. J. 1981, Phys. Rev. B, 14, 4015
655:
656: \bibitem[Sneden(1973)]{sneden73}
657: Sneden, C. 1973, ApJ, 184, 839
658:
659: \bibitem[Sneden, Gratton, \& Crocker(1991)]{sgc91}
660: Sneden, C., Gratton, R. G., \& Crocker, D. A. 1991, \aap, 246, 354
661:
662: \bibitem[Tinsley(1980)]{tinsley80}
663: Tinsley, B. M. 1980, Fundamentals of Cosmic Physics Vol.5, p.287
664:
665: \bibitem[Truran \etal(2002)]{truran02}
666: Truran, J. W., Cowan, J. J., Pilachowski, C. A., \& Sneden, C. 2002,
667: PASP, 114, 1293
668:
669: \bibitem[Weiss(1963)]{weiss63}
670: Weiss, A. W. 1963, \apj, 138, 1262
671:
672: \bibitem[Wheeler, Sneden, \& Truran(1989)]{wheeler89}
673: Wheeler, J. C. , Sneden, C, \& Truran, J. W. 1989, ARAA, 27, 279
674:
675:
676: \end{thebibliography}
677: \clearpage
678:
679: \input{tab1.tex}
680:
681: \input{tab2.tex}
682:
683: \input{tab3.tex}
684:
685: \input{tab4.tex}
686:
687: \begin{figure}[!ht]
688: \plotone{f1.eps} \caption{Examples of spectra for seven stars over
689: a wavelength range of 4315-4350\AA. Several identified lines are
690: indicated. This spectral region includes the CH molecular band
691: used for deriving carbon abundance. H$\gamma$ line is observed at
692: 4340\AA.} \label{fig:sample}
693: \end{figure}
694:
695: \begin{figure}[!ht]
696: \plotone{f2.eps} \caption{Comparison between the EWs(in $m\AA$)
697: measured from the BOES spectra and those corresponding to
698: \cite{gratton03}} \label{fig:ew_comp1}
699: \end{figure}
700:
701: \begin{figure}[!ht]
702: \plotone{f3.eps} \caption{Comparison between the EWs measured from
703: the BOES spectra and those corresponding to \cite{cayrel04} and
704: \cite{honda04} for HD 122563} \label{fig:ew_comp2}
705: \end{figure}
706:
707: \begin{figure}[!ht]
708: \plotone{f4.eps} \caption{(a) An example of the determination of
709: radial velocity of HD 58551 prior to heliocentric velocity
710: correction. The dotted line is the Gaussian fitting result.
711: (b){\itshape(Upper panel)} Comparison between radial velocities
712: measured from the BOES spectra and those from other radial
713: velocity studies. {\itshape(Lower panel)} Differences in radial
714: velocities. The solid line represents the mean offset and the
715: dotted line represents a $2\sigma$ dispersion of velocity
716: differences. The velocities of HD 6755, HD 8724, HD 165908, and
717: BD+17 4708, represented as filled squares, show large deviations
718: from previous studies. These stars, except HD 8724, are known as
719: binaries.} \label{fig:rad_vel}.
720: \end{figure}
721:
722: \begin{figure}[!ht]
723: \plotone{f5.eps} \caption{Fe abundances for HD 122563 with derived
724: atmospheric parameters(\Teff$=4430K$, $\log g=0.58$, $\xi_v =
725: 2.15$ km/s). Open circle shows iron abundance derived from the Fe
726: I line and filled circle shows that derived from the Fe II line.
727: }\label{fig:fe_abun}
728: \end{figure}
729:
730: \begin{figure}[!ht]
731: \plotone{f6.eps} \caption{Atmospheric parameter(\Teff, $\log g$,
732: $\xi_v$, $[Fe/H]$) comparisons between this study and various
733: other studies. Filled square indicates comparison with
734: \cite{go84}; plus, indicates comparison with \cite{sgc91}; open
735: square, \cite{gs94}; open circle, \cite{ful00}; filled triangle,
736: \cite{mk01}; open triangle, \cite{gratton03}; open star,
737: \cite{honda04}; filled star, \cite{cayrel04}; and filled circle,
738: \cite{johnsell05} }\label{fig:atcomp}
739: \end{figure}
740:
741:
742: \end{document}
743: