1: %\documentclass{emulateapj}
2: \documentclass[preprint]{aastex}
3: %\usepackage{emulateapj5}
4: %\usepackage{apjfonts}
5:
6: %\received{2003 May 22}
7: \begin{document}
8: \title{Europium Isotope Ratios in s-Process-Element-Enhanced, Metal-Poor Stars: A
9: New Probe of the $^{151}$Sm Branching\footnote{Based on data collected
10: at the Subaru Telescope, which is operated by the National Astronomical
11: Observatory of Japan.}}
12:
13: \author{Wako Aoki\altaffilmark{2}, Sean G. Ryan\altaffilmark{3},
14: Nobuyuki Iwamoto\altaffilmark{2,8}, Timothy
15: C. Beers\altaffilmark{4},
16: John E. Norris\altaffilmark{5}, Hiroyasu Ando\altaffilmark{2},
17: Toshitaka Kajino\altaffilmark{2},
18: Grant J. Mathews\altaffilmark{6},
19: Masayuki Y. Fujimoto\altaffilmark{7}}
20:
21: \altaffiltext{2}{National Astronomical Observatory, Mitaka, Tokyo,
22: 181-8588 Japan; email: aoki.wako@nao.ac.jp, ando@optik.mtk.nao.ac.jp,
23: kajino@nao.ac.jp}
24: \altaffiltext{3}{Department of Physics and Astronomy, The Open
25: University, Walton Hall, Milton Keynes, MK7 6AA, UK; email:
26: s.g.ryan@open.ac.uk}
27: \altaffiltext{4}{Department of Physics and Astronomy, Michigan State
28: University, East Lansing, MI 48824-1116; email: beers@pa.msu.edu}
29: \altaffiltext{5}{Research School of Astronomy and Astrophysics, The
30: Australian National University, Mount Stromlo Observatory, Cotter
31: Road, Weston, ACT 2611, Australia; email: jen@mso.anu.edu.au}
32: \altaffiltext{6}{Department of Physics, Center for Astrophysics,
33: University of Notre Dame, Notre Dame, IN 46556, USA; email:
34: gmathews@nd.edu}
35: \altaffiltext{7}{Department of Physics, Hokkaido University, Sapporo,
36: Hokkaido 060-0810, Japan; email: fujimoto@astro1.sci.hokudai.ac.jp}
37: \altaffiltext{8}{Present address: Department
38: of Astronomy, School of Science, University of Tokyo, Hongo,
39: Bunkyo-ku, Tokyo, 133--0033 Japan, niwamoto@astron.s.u-tokyo.ac.jp}
40:
41: \begin{abstract}
42:
43: We report on the first measurement of the Eu isotope fractions
44: ($^{151}$Eu and $^{153}$Eu) in s-process-element-enhanced, metal-poor
45: stars. We use these ratios to investigate the $^{151}$Sm branching of
46: s-process nucleosynthesis. The measurement was made by detailed study
47: of \ion{Eu}{2} lines that are significantly affected by hyperfine
48: splitting and isotope shifts in spectra of the carbon-rich, very
49: metal-poor stars LP~625--44 and CS~31062--050, observed with the
50: Subaru Telescope High Dispersion Spectrograph. The $^{151}$Eu
51: fractions [fr($^{151}$Eu) = $^{151}$Eu/($^{151}$Eu+$^{153}$Eu)]
52: derived for LP~625--44 and CS~31062--050 are 0.60 and 0.55,
53: respectively, with uncertainties of about $\pm 0.05$. These values are
54: higher than found in solar-system material, but agree well with the
55: predictions of recent s-process models. We derive new constraints on
56: the temperature and neutron density during the s-process based on
57: calculations of pulsed s-process models for the $^{151}$Eu fraction.
58:
59: \end{abstract}
60: \keywords{nuclear reactions, nucleosynthesis, abundances --- stars: abundances --- stars: AGB and post-AGB --- stars: Population II}
61:
62:
63: %\keywords{nuclear reactions, nucleosynthesis, abundances -- stars:
64: %abundances -- stars: AGB and post-AGB --stars: Population II}
65:
66: \section{Introduction}\label{sec:intro}
67:
68: The slow neutron capture process (s-process) is
69: responsible for about half of the abundances of elements heavier than
70: the iron peak in solar-system material. The s-process has
71: traditionally been described by two approaches: the schematic
72: classical approach \citep[e.g., ][]{kappeler89}, and modeling of the
73: s-process in thermally pulsing asymptotic giant branch (AGB) stars
74: \citep[e.g., ][]{straniero95}. In the classical approach, the neutron
75: exposure is estimated from the abundance patterns of
76: pure s-process isotopes in solar-system material, or the elemental
77: abundance patterns in stars enriched in s-process-elements, while the
78: temperature and neutron density during the s-process can be estimated
79: by an analysis of the branchings of the
80: neutron-capture chains where the rates for beta decay and neutron
81: capture are comparable.
82:
83: For $^{151}$Sm, whose half-life is about 90 years, the $\beta$-decay
84: rate is strongly dependent on temperature, while the neutron capture
85: rate is not. This makes the $^{151}$Sm branching an excellent
86: thermometer \citep[e.g., ][]{wisshak95}. Previously, this
87: branching has been analyzed using the $^{152}$Gd and $^{154}$Gd
88: isotope ratios in solar-system material, which are believed to be
89: significantly affected by this branching \citep[e.g., ][]{beer88,
90: wisshak95}. One difficulty in this approach is that these Gd isotopes
91: are affected by a small amount of contamination from the p-process,
92: though they are shielded from the r-process. Hence, an independent
93: observational constraint on the branching is desired.
94:
95: The Eu isotope ratio ($^{151}$Eu/$^{153}$Eu) is one possible
96: constraint. More than 90~\% of Eu in the solar system is provided by
97: r-process nucleosynthesis \citep[e.g., ][]{kappeler89, arlandini99,
98: burris00}, which suggests that it is difficult to constrain the
99: s-process from measurement of the Eu isotopes in {\it solar-system
100: material}. However, the Eu isotope ratios can also be measured in
101: individual stars, as high-resolution spectroscopy can partially
102: resolve the expected hyperfine splitting and isotope
103: shifts. \citet{sneden02} and \citet{aoki03a} analyzed the Eu isotopes
104: for r-process-element-enhanced stars. They showed that the isotope
105: ratios derived for a total of four of these objects agree well with
106: that in solar-system material. In this Letter, we apply this analysis
107: to two s-process-element-enhanced, metal-poor stars, and report the
108: first results of our analysis of Eu isotopes to analyze the $^{151}$Sm
109: branching.
110:
111: \section{Observations}\label{sec:obs}
112:
113: We selected two subgiants (LP~625--44 and CS~31062--050), which were
114: found to be enriched in s-process-elements in our previous studies
115: \citep{aoki00,aoki02b} as most suitable for this investigation. These
116: stars are both very metal-poor ([Fe/H] = --2.7 and [Fe/H] = --2.4,
117: respectively\footnote{[A/B] $\equiv \log(N_{\rm A}/N_{\rm B})-
118: \log(N_{\rm A}/N_{\rm B})_{\odot}$, and $\log \epsilon_{\rm A} \equiv
119: \log(N_{\rm A}/N_{\rm H})+12$ for elements A and B.}) and have quite
120: similar atmospheric parameters, as shown in Table~\ref{tab:obs}. The
121: observed enhancement of s-process-elements in these stars is believed
122: to be due to nucleosynthesis in extinct AGB stars which transfered
123: mass to the surviving luminous companions of the binary systems. The
124: binarity of these stars is discussed below. The enhancement of the
125: bulk of the s-process elements in these two stars is also similar
126: (e.g., [Ba/Fe]$\sim +2.5$). One important difference between these two
127: objects is their Pb abundance: the Pb/Ba ratio in CS~31062--050 is
128: five times higher than that in LP~625--44. For comparison purposes, we
129: also selected a very metal-deficient giant, HD~6268 ([Fe/H] = --2.5),
130: which exhibits a moderate enhancement of r-process-elements, and has
131: strong Eu lines, comparable to those in the two
132: s-process-element-enhanced stars.
133:
134: The [Ba/Eu] values of LP~625--44 and CS~31062--050 derived in our
135: previous studies \citep{aoki02a,aoki02b} are 1.09 and 0.46,
136: respectively. These are more than 1~dex higher than the value of the
137: r-process component in solar-system material ($-0.69$, Arlandini et
138: al. 1999). While the [Ba/Eu] value of LP~625--44 is similar to that of
139: the solar-system s-process component ($+1.15$), the value of
140: CS~31062--050 is significantly lower. However, low [Ba/Eu] values
141: ($\sim 0.4$) are predicted by some AGB nucleosynthesis models
142: \citep[e.g., ][]{goriely00} for metal-deficient stars, for which high
143: neutron-exposure is expected due to the high ratios of neutrons per
144: seed nuclei. Indeed, the value of [Ba/Eu]$\sim 0.4$ is not unusual in
145: carbon-rich, metal-deficient stars like CS~31062--050 \citep[e.g.,
146: ][]{aoki02b,johnson02}. To explain these [Ba/Eu] values by
147: contamination from the r-process, we must assume a very large
148: enhancement of r-process elements ([Eu/Fe]$\sim$ +1.5--1.8), similar
149: to that in extremely r-process-enhanced stars, such as CS~31082--001
150: \citep{cayrel01}, which are known to be quite rare. For these
151: reasons, we assume in the following discussion that the majority of
152: the Eu in CS~31062--050, as well as in LP~625--44, originates from the
153: s-process.
154:
155: The observations we report here were made with the Subaru Telescope
156: High Dispersion Spectrograph (HDS; Noguchi et al. 2002) in August
157: 2002. The wavelength range covered was 3100-4700~{\AA}, with a
158: resolving power $R = 90,000$. The total exposure times and
159: signal-to-noise ratios obtained are listed in Table~\ref{tab:obs}.
160:
161:
162: The heliocentric radial velocities measured for our spectra are given
163: in Table~\ref{tab:obs}. \citet{aoki00} already reported a variation of
164: radial velocities for LP~625--44, which is confirmed by the present
165: measurement. For CS~31062-050, a change of radial velocity from the
166: previous measurement ($V_{r} = 7.75$~km~s$^{-1}$ by Aoki et al. 2002b)
167: was also detected. These results suggest the binarity of these stars,
168: and strongly support the mass-transfer scenario for the enrichment of
169: s-process elements.
170:
171: \section{Analysis and Results}\label{sec:ana}
172:
173: We adopted the Eu line data, including the hyperfine splitting and
174: isotope shifts, provided by \citet{lawler01}. The isotope fractions of
175: $^{151}$Eu [fr($^{151}$Eu) = $^{151}$Eu / ($^{151}$Eu + $^{153}$Eu)]
176: were measured by fitting observed spectra with synthetic ones
177: calculated for the Eu lines, using the model atmospheres of
178: \citet{kurucz93} and atmospheric parameters derived in the previous
179: studies \citep{aoki02a,aoki02b}. The macro-turbulent velocity was
180: estimated by fitting a gaussian profile to clean Fe and Ti lines
181: detected in the same spectrum for each object (see
182: Table~\ref{tab:obs}). Our careful investigation of the HDS
183: instrumental profiles around the Eu lines used in the analysis showed
184: that the profile is quite symmetric. Hence, a gaussian profile
185: approximation is sufficient for the analysis of stellar spectra that
186: are further broadened by macro-turbulence.
187:
188:
189: Figure~\ref{fig:sp} shows the spectra of the \ion{Eu}{2} 4205~{\AA}
190: line in our three program stars. In the calculation of synthetic
191: spectra, blending by other species is included using the comprehensive
192: line list of \citet{kurucz95} and the CH and CN line lists produced in
193: \citet{aoki02b}. The dot-dashed lines show the synthetic spectra
194: calculated without the \ion{Eu}{2} contribution. The absorption around
195: 4205~{\AA} is dominated by \ion{Eu}{2}. In the top panel, the line
196: positions of both Eu isotopes are shown. Since the hyperfine splitting
197: of $^{151}$Eu is larger than that of $^{153}$Eu, the asymmetry of the
198: line profile increases with increasing fr($^{151}$Eu).
199:
200: Our measurements were carried out using the same procedure as was
201: utilized for r-process-enhanced stars in \citet{aoki03a}. We searched
202: for the isotope fractions that minimize the value of $\chi_{\rm
203: r}^{2}$ for a given Eu abundance. We estimated the uncertainty of the
204: derived Eu isotope fractions and total abundances by considering the
205: range over which the $\chi_{\rm r}^{2}$ is twice as large as the
206: best-fit case\footnote{This corresponds to approximately a $3 \sigma$
207: confidence level for 25 degrees of freedom, which is the case of our
208: analysis for Eu lines. We do not, however, insist on the exact value
209: of the confidence level, because the statistical properties are not
210: proven for spectra re-binned to a constant wavelength step (Bonifacio
211: \& Caffau 2003).}. The error in the isotope fraction due to the
212: uncertainty in the total Eu abundance was estimated from the range of
213: the isotope fraction allowed within the adopted abundance uncertainty.
214:
215: We estimated the errors arising from the following factors using the
216: same procedures as in \citet{aoki03a}: (1) The macro-turbulent
217: velocity (given in Table~\ref{tab:obs}); (2) The continuum-level
218: uncertainty estimates, which are assumed to be 0.5\% for HD~6268 and
219: 1\% for the other two stars; and (3) The wavelength calibration of the
220: spectrum and the Eu line position. The wavelength shift was estimated
221: from the longer wavelength (redder) part of the spectral line, which
222: is insensitive to the Eu isotope ratio; the error is typically a few
223: m{\AA}.
224:
225: We analyzed three \ion{Eu}{2} lines at 3819, 4129, and 4205~{\AA}.
226: Table~\ref{tab:res} gives the derived $^{151}$Eu fraction and total
227: error ($\sigma_{\rm total}$), estimated from the quadrature sum of the
228: individual errors mentioned above. The Eu ($^{151}$Eu + $^{153}$Eu)
229: abundance derived from each line is also given in the table. The
230: 4205~{\AA} line is most sensitive to the Eu isotope
231: fractions. Therefore, the fractions deduced from this line have the
232: smallest uncertainty, even though the strong CH line at 4204.75~{\AA}
233: affects the bluest part of the \ion{Eu}{2} line. The 3819~{\AA} line
234: is strongest among the three lines. However, the uncertainty in the
235: derived isotope ratio due to the error in the Eu abundance estimation
236: is significant \citep{aoki03a}. The \ion{Eu}{2} 4129~{\AA} line, as
237: for \ion{Eu}{2} 4205~{\AA}, has suitable strength for this
238: analysis. However, our synthetic spectra fail to well reproduce the
239: blue portion ($\sim$ 4129.6~{\AA}) of the absorption line in the two
240: carbon-rich stars. There seems to exist some unidentified absorption
241: lines. Excluding the blue wing from the fitting, we derived the
242: isotope fractions, given in Table~\ref{tab:res}, for the range of
243: 4129.65--4129.82~{\AA}. Though the number of data points within this
244: range is sufficient, the results are rather sensitive to the choice of
245: wavelength range used for the analysis.
246:
247: For these reasons, we prefer to adopt the isotope fractions derived
248: from the \ion{Eu}{2} 4205~{\AA} line as the best determination. We
249: note, for comparison purposes, that the weighted means of the results
250: from all three lines for HD~6268, LP~625--44, and CS~31062--050 are
251: 0.47, 0.61, and 0.57, respectively, which agree well with the results
252: from the 4205~{\AA} line alone.
253:
254: The fr($^{151}$Eu) of HD~6268 (0.48$\pm 0.04$) perfectly agrees with
255: that of solar-system material \citep[0.478; ][]{anders89}, as found in
256: other r-process-element-enhanced stars \citep{sneden02,aoki03a}. In
257: contrast, the fr($^{151}$Eu) of the two stars enriched in s-process
258: elements are higher than the solar-system value.
259:
260:
261: \section{Discussion}
262:
263: The yields of the s-process nuclides which explain the abundances of
264: pure s-process isotopes in solar-system material have been calculated
265: by \citet{arlandini99}. The fr($^{151}$Eu) values deduced from their
266: best-fit stellar and classical models are 0.541 and 0.585,
267: respectively. The first conclusion of the present investigation is
268: that {\it the agreement between these values and those derived from
269: our observations is quite good}.
270:
271: In order to further investigate what can be learned about the
272: s-process from the measured Eu isotopes, we have made an analysis
273: using the thermally pulsed s-process models described in Howard et al.
274: (1986). We utilized updated neutron-capture rates \citep[e.g.,
275: ][]{bao00}. In particular, in order to follow the nuclear flow in the
276: Sm-Eu-Gd region, we included the rates for $^{151-155}$Eu and
277: $^{151}$Sm obtained by \citet{best01} and \citet{toukan95}. The
278: electron-capture rate of $^{153}$Gd was adopted from
279: \citet{takahashi87}. We found that the fr($^{151}$Eu) value has almost
280: no dependence on the mean neutron exposure ($\tau_{0}$) for a fixed
281: temperature in the range of $0.2\leq \tau_{0} \leq 0.8$, which well
282: covers the $\tau_{0}$ ($\sim 0.5$~mb$^{-1}$) estimated by
283: \citet{aoki01} for LP~625--44. Our results are calculated with
284: $\tau_{0}=$ 0.3~mb$^{-1}$, which gives a good fit to solar abundances.
285:
286: Figure~\ref{fig:fr151eu} shows the fr($^{151}$Eu) values calculated by
287: our model. They are plotted as a function of neutron density ($N_{\rm
288: n}$) for four temperatures ($kT=$ 10, 15, 20, and 30~keV). Also
289: shown for comparison by the hatched area is the fr($^{151}$Eu) range
290: deduced for the s-process-element-enhanced star LP~625--44. As can be
291: seen in this figure, the fr($^{151}$Eu) value is rather sensitive to
292: the ambient temperature and neutron density during the s-process. The
293: fr($^{151}$Eu) value is maximized in the range of neutron density from
294: $N_{\rm n} =$ 5$\times$ $10^7$ to $10^9$~cm$^{-3}$.
295:
296:
297:
298: The branching factor at $^{151}$Sm is given by
299: $f=\lambda_n/(\lambda_\beta+\lambda_n)$ , where $\lambda_n = N_{\rm n}
300: v_{T} <\sigma>$ and $\lambda_\beta = \ln 2/t_{1/2}$ are the
301: neutron-capture rate and the $\beta$-decay rate, respectively. In
302: these formulations, $v_{T}$, $<\sigma>$, and $t_{1/2}$ are the mean
303: thermal velocity, the Maxwellian averaged cross section, and the
304: half-life, respectively. For $N_{\rm n} > 10^{7}$~cm$^{-3}$, the branching
305: factor is higher than 0.9. This indicates that the nuclear flow
306: bypasses $^{151}$Eu. In this case, the $^{151}$Eu abundance is
307: produced by the decay of $^{151}$Sm after the neutron-capture flow
308: ceases. As the neutron density increases to $N_{\rm n} > 10^{9}$~cm$^{-3}$,
309: another branching at $^{153}$Sm becomes important. Since we adopt a
310: much smaller cross section for $^{153}$Sm than those for $^{151}$Sm
311: and $^{153}$Eu, the $^{153}$Sm abundance relative to $^{151}$Sm and
312: $^{153}$Eu {\it during} the neutron capture reactions increases with
313: increasing neutron density. This results in a decrease of the {\it
314: final} value of fr($^{151}$Eu) with increasing neutron density for
315: $N_{\rm n} > 10^{9}$~cm$^{-3}$, as found in Figure~\ref{fig:fr151eu}. A
316: measurement of the cross section for $^{153}$Sm is thus highly
317: desirable to constrain the high neutron density branching.
318:
319:
320: In contrast to the high neutron-density conditions, $^{151}$Sm
321: $\beta$-decay {\it during} the s-process is increasingly important
322: with decreasing neutron density ($N_{\rm n}\lesssim 10^{7}$
323: ~cm$^{-3}$). For the typical temperature ranges thought to apply in
324: the s-process (10-30~keV), the neutron-capture rate on $^{151}$Eu is
325: faster than that for $^{153}$Eu. Therefore, the nuclear flow passes
326: through $^{151}$Eu rather easily, and creates $^{153}$Gd via
327: $^{151}$Eu($n$, $\gamma$) $^{152}$Eu($\beta,\nu$)$^{152}$Gd($n$,
328: $\gamma$)$^{153}$Gd. In low neutron-density conditions, the
329: electron-capture on $^{153}$Gd is comparable with, or faster than, the
330: neutron capture. This contributes to the production of $^{153}$Eu and
331: explains the decrease of fr($^{151}$Eu) with decreasing neutron
332: density in the low ($N_{\rm n}\lesssim 10^{7}$ cm$^{-3}$) range. This trend
333: appears more clearly for the higher temperature case, because the
334: effect of a higher neutron-capture rate on $^{151}$Eu relative to that
335: on $^{153}$Eu is large, while the branching factor decreases with
336: increasing temperature.
337:
338: The comparison of the calculations with the observational results
339: indicates that s-processes with high neutron density ($\log N_{\rm n}
340: \gtrsim 9$) and low temperature ($kT \lesssim 20$~keV), or those with
341: quite low neutron density ($N_{\rm n}<10^{7}$cm~s$^{-1}$), are not
342: allowed. This is a new constraint on s-process nucleosynthesis
343: provided by the Eu isotope analysis.
344:
345: Recent models of AGB stars show that the abundance patterns of nuclei
346: in branchings are affected by the s-process both during the thermal
347: pulses and between pulses \citep[e.g., ][]{arlandini99}. The reaction
348: which provides neutron in the former phase is
349: $^{22}$Ne$(\alpha,n)^{25}$Mg, which produces a neutron density as high
350: as $10^{8}$--$10^{10}$~cm$^{-3}$. In the interpulse phase,
351: $^{13}$C$(\alpha,n)^{16}$O is assumed to be the neutron source
352: reaction, which leads to a lower neutron density ($N_{\rm n}\sim
353: 10^{7}$~cm$^{-3}$). The observational errors of the derived Eu isotope
354: fractions are still too large to constrain the contribution of each
355: process to the final abundance patterns produced by nucleosynthesis in
356: AGB stars. However, our analysis shows that Eu isotopes can be a new
357: probe to determine the temperature and neutron density in s-process
358: nucleosynthesis.
359:
360: We note that the fr($^{151}$Eu) values derived by using the previous
361: nuclear data of Sm and Eu isotopes, in particular the $^{151}$Sm
362: neutron-capture rate, are significantly lower than the results shown
363: above. The previous neutron-capture rate recommended by \citet{bao00},
364: e.g., 2377~mb at 30~keV, is significantly higher than the rate used in
365: the present work (1585~mb at 30~keV, Best et al. 2001). The high
366: neutron-capture rate causes the nuclear flow to bypass $^{151}$Eu, and
367: results in lower fr($^{151}$Eu) values by 0.05-0.08. The low
368: neutron-capture rate is clearly preferable to explain our
369: observational results. New experiments to determine the neutron
370: capture cross section of $^{151}$Sm are highly desirable to fix ratios
371: of $^{151}$Eu/$^{153}$Eu and also $^{152}$Gd/$^{154}$Gd.
372:
373: It should also be noted that the Eu isotope fractions are quite
374: similar between LP~625--44 and CS~31062--050, even though their Pb/Ba
375: abundance ratios are significantly different \citep{aoki02a,aoki02b}.
376: This observational fact indicates that nuclear flow is very fast, and
377: fr($^{151}$Eu) easily obtains asymptotic values in the Sm-Eu-Gd
378: region, once the flow passes through the $N=82$ neutron magic
379: nuclei. This situation is quite likely to occur in these very
380: metal-poor stars because of the expected high neutron-to-seed
381: ratio. However, the Pb abundance is very sensitive to the neutron
382: exposure \citep[e.g., ][]{gallino03}, and large variations of the
383: Pb/Ba ratio are expected. Similar studies of additional
384: s-process-enhanced stars would be useful to derive more clear
385: conclusions on production of heavy nuclei by the s-process.
386:
387: \acknowledgments
388:
389: \begin{thebibliography}{}
390: \bibitem[Anders \& Grevesse (1989)]{anders89} Anders, E. \& Grevesse,
391: N. 1989, Geochem. Cosmochim. Acta, 53, 197
392: \bibitem[Aoki et al. (2002a)]{aoki02a} Aoki, W., et al. 2002a, PASJ,
393: 54, 427
394: \bibitem[Aoki et al. (2003)]{aoki03a} Aoki, W., Honda, S., Beers, T.C.,
395: \& Sneden, C. 2003, \apj, 586, 506
396: %\bibitem[Aoki et al. (2003b)]{aoki03b} Aoki, W., Inoue, S.,
397: %Kawanomoto, S., Ryan, S.G., Smith, I.M., Suzuki, T., \& Takada-Hidai,
398: %M. 2003b, in preparation
399: \bibitem[Aoki et al. (2000)]{aoki00} Aoki, W., Norris, J. E., Ryan,
400: S. G., Beers, T. C., \& Ando, H. \ 2000, \apj, 536, L97
401: \bibitem[Aoki et al. (2001)]{aoki01} Aoki, W., Ryan, S.G., Norris, J.E., Beers, T.C., Ando, H., Iwamoto, N., Kajino, T., Mathews, G. J. \& Fujimoto, M.Y., 2001, \apj, 561, 346
402: \bibitem[Aoki et al. (2002b)]{aoki02b} Aoki, W., Ryan, S. G., Norris,
403: J. E., Beers, T. C., Ando, H., \& S. Tsangarides, 2002b, \apj, 580,
404: 1149
405: \bibitem[Arlandini et al. (1999)]{arlandini99} Arlandini, C.,
406: K\"{a}ppeler, F., Wisshak, K., Gallino, R., Lugaro, M., Busso, M. \&
407: Straniero, O. 1999, \apj, 525, 886
408: \bibitem[Bao et al. (2000)]{bao00} Bao, Z. Y., Beer, H., K\"{a}ppeler,
409: F., Voss, F., Wisshak, K., Rauscher, T. 2000, At. Data Nucl. Data
410: Tables, 76, 70
411: \bibitem[Beer \& Macklin (1988)]{beer88} Beer, H., \& Macklin, R.L. 1988, 331, 1047
412: \bibitem[Best et al. (2001)]{best01} Best, J., Stoll, H., Arlandini,
413: C., Jaag, S., K\"{a}ppeler, F., Wisshak, K., Mengoni, A., Reffo, G.,
414: \& Rauscher, T. 2001, Phys. Rev. C, 64, 015801
415: \bibitem[Bonifacio \& Caffau (2003)]{bonifacio03} Bonifacio \& Caffau
416: 2003, \aap, 399, 1183
417: \bibitem[Burris et al. (2000)]{burris00} Burris, D. L., Pilachowski,
418: C. A., Armandroff, T. E., Sneden, C., Cowan, J. J., \& Roe, H. 2000,
419: \apj, 544, 302
420: \bibitem[Cayrel et al. (2001)]{cayrel01} Cayrel, R. et al., 2001,
421: Nature, 409, 691
422: \bibitem[Gallino et al. (2003)]{gallino03} Gallino, R. et al. 2003,
423: Nucl. Phys. A, 718, 181c
424: \bibitem[Goriely and Mowlavi (2000)]{goriely00} Goriely, S., \&
425: Mowlavi, N. 2000, \aap, 362, 599
426: \bibitem[Howard et al. (1986)]{howard86} Howard, W. M., Mathews,
427: G. J., Takahashi, K., \& Ward, R. A. 1986, ApJ, 309, 633
428: \bibitem[Johnson \& Bolte (2002)]{johnson02} Johnson, J. A., \& Bolte,
429: M. 2002, \apj, 579, L87
430: \bibitem[K\"{a}ppeler, Beer, \& Wisshak (1989)]{kappeler89}
431: K\"{a}ppeler, F., Beer, H.,\& Wisshak, K., 1989, Rep. Prog. Phys., 52,
432: 945
433: \bibitem[Kurucz (1993)]{kurucz93} Kurucz, R. L., 1993, CD-ROM 13,
434: ATLAS9 Stellar Atmospheres Programs and 2km/s Grid (Cambridge:
435: Smithsonian Astrophys. Obs.)
436: \bibitem[Kurucz \& Bell (1995)]{kurucz95} Kurucz, R. L. \& Bell, B.,
437: 1995, CD-ROM 23, Atomic Line List (Cambridge: Smithsonian
438: Astrophys. Obs.)
439: \bibitem[Lawler et al. (2001)]{lawler01} Lawler, J. E., Wickliffe,
440: M. E., Den Hartog, E. A. \& Sneden, C. 2001, \apj, 563, 1075
441: \bibitem[Noguchi et al. (2002)]{noguchi02} Noguchi, K., Aoki, W.,
442: Kawanomoto, S., et al. 2002, PASJ, 54, 855
443: \bibitem[Sneden et al. (2002)]{sneden02} Sneden, C., Cowan, J. J.,
444: Lawler, J. E., Burles, S., Beers, T. C. \& Fuller, G. M. 2002, \apj,
445: 566, L25
446: \bibitem[Straniero et al. (1995)]{straniero95} Straniero, O., Gallino,
447: R., Busso, M., Chieffi, A., Raiteri, C. M., Limongi, M. \& Salaris,
448: M., 1995, \apj, 440, L85
449: \bibitem[Takahashi \& Yokoi (1987)]{takahashi87} Takahashi, K., \&
450: Yokoi, K. 1987, At. Data. Nucl. Data Tables, 36, 375
451: \bibitem[Toukan et al. (1995)]{toukan95} Toukan, K. A., Debus, K.,
452: K\"{a}ppeler, F., \& Reffo, G. 1995, Phys. Rev. C, 51, 1540
453: \bibitem[Wisshak et al. (1995)]{wisshak95} Wisshak, K., Voss, F.,
454: K\"{a}ppeler, F., Guber, K., Kazakov, L., Kornilov, N., Uhl, M., \&
455: Reffo, G. 1995, Phys. Rev. C, 52, 2762
456: \end{thebibliography}
457:
458: %\clearpage
459: %\begin{figure}
460: %\includegraphics[width=8cm]{f1.eps}
461: %\caption[]{The s-process reaction path around the $^{151}$Sm branching.
462: %Unstable nuclei are shown by boxes with dotted lines.}
463: %\label{fig:sm151br}
464: %\end{figure}
465: %
466: %\clearpage
467:
468: \begin{figure}
469: \includegraphics[width=8cm]{f1.eps}
470: %\plotone{eu4205.eps}
471: \caption[]{Comparison of the observed spectra (dots) and synthetic
472: ones (lines) for the \ion{Eu}{2} 4205~{\AA} line. The name of the
473: object and the adopted fr($^{151}$Eu) value are presented in each
474: panel. The solid line shows the synthetic spectra for the adopted
475: fr($^{151}$Eu); the dotted and dashed lines show those for ratios
476: which are smaller and larger by 0.10 in fr($^{151}$Eu),
477: respectively. The dot-dashed lines show the synthetic spectra for no
478: Eu. The weak blending at 4205.09~{\AA} in the spectrum of HD~6268
479: (top) is due to \ion{V}{2}, whose effect is negligible in the two
480: subgiants.. The wavelengths and relative strength of the hyperfine
481: components for $^{151}$Eu and $^{153}$Eu are shown in the top panel.}
482: \label{fig:sp}
483: \end{figure}
484: %
485: %\clearpage
486:
487: \begin{figure}
488: \includegraphics[width=8cm]{f2.eps}
489: \caption[]
490: {The fr($^{151}$Eu) values calculated for four temperatures (squares:
491: $kT$ = 30~keV, triangles: 20~keV, circles: 15~keV, and asterisks:
492: 10~keV) as a function of neutron density ($N_{\rm n}$). The
493: fr($^{151}$Eu) of LP~625--44 and its uncertainty is shown by the
494: hatched area. }
495: \label{fig:fr151eu}
496: \end{figure}
497:
498: %\clearpage
499: \begin{deluxetable}{@{}l@{\extracolsep{\fill}}c@{\extracolsep{\fill}}c@{\extracolsep{\fill}}c@{\extracolsep{\fill}}c@{\extracolsep{\fill}}c@{\extracolsep{\fill}}c}
500: \tabletypesize{\small}
501: %\begin{deluxetable}{lcccccc}
502: %\tablewidth{0pt}
503: %\caption{PROGRAM STARS AND OBSERVATIONS}\label{tab:obs}
504: \tablecaption{PROGRAM STARS AND OBSERVATIONS\label{tab:obs}}
505: \startdata
506: %\noalign{smallskip}
507: \tableline
508: %\hline
509: \tableline
510: Star & Exp.\tablenotemark{a} & S/N\tablenotemark{b} & Obs. date (JD) & Radial velocity ~ & $T_{\rm eff}/\log g/$[Fe/H]/$v_{\rm micro}$ & $v_{\rm macro}$ \\
511: & & & & (km s$^{-1}$) & & (km s$^{-1}$) \\
512: \tableline
513: LP~625--44 & 270 (9) & 137 & 22/08/2002 (2,452,509) & $28.41\pm0.19$ & 5500/2.5/$-2.7$/1.2 & $6.69\pm 0.13$ \\
514: CS~31062--050 ~ & 390 (13) & 124 & 23/08/2002 (2,452,510) & $11.51\pm0.17$ & 5600/3.0/$-2.4$/1.3 & $5.30\pm 0.40$ \\
515: HD~6268 & 100 (5) & 304 & 22/08/2002 (2,452,509) & $40.13\pm0.15$ & 4600/1.0/$-2.5$/2.1 & $7.27\pm 0.27$ \\
516: \tableline
517: \enddata
518: %~ \\
519:
520: \tablenotetext{a}{
521: Total exposure time (minute) and number of exposures.
522: }
523:
524: \tablenotetext{b}{
525: S/N ratio per 0.012~{\AA} pixel at 4000~{\AA}.
526: }
527:
528: \end{deluxetable}
529: %\clearpage
530:
531: \begin{deluxetable}{ccccccccc}
532: %\tablewidth{0pt}
533: \tablecaption{RESULTS\label{tab:res}}
534: %\caption{RESULTS}\label{tab:res}
535: \startdata
536: \tableline
537: \tableline
538: line & \multicolumn{2}{c}{HD~6268} & & \multicolumn{2}{c}{LP~625--44} & & \multicolumn{2}{c}{CS~31062--050} \\
539: \cline{2-3}\cline{5-6}\cline{8-9}
540: & fr($^{151}$Eu) & $\log \epsilon$(Eu) & & fr($^{151}$Eu) & $\log \epsilon$(Eu) & & fr($^{151}$Eu) & $\log \epsilon$(Eu) \\
541: \tableline
542: 3819{\AA} & 0.44 $\pm$ 0.11 & $-1.55$ & & 0.68 $\pm$ 0.10 & $-0.44$ & & 0.65 $\pm$ 0.14 & $-0.02$ \\
543: 4129{\AA} & 0.49 $\pm$ 0.07 & $-1.57$ & & 0.56 $\pm$ 0.07 & $-0.45$ & & 0.56 $\pm$ 0.09 & $+0.01$ \\
544: 4205{\AA} & 0.48 $\pm$ 0.04 & $-1.54$ & & 0.60 $\pm$ 0.06 & $-0.43$ & & 0.55 $\pm$ 0.05 & $+0.04$ \\
545: \tableline
546: \enddata
547: \end{deluxetable}
548:
549: \end{document}
550:
551:
552:
553: