1: %\documentclass{emulateapj}
2: %\documentclass[12pt,preprint]{aastex} % para submeter.
3: \documentclass[10pt,preprint]{aastex} % para submeter.
4: %\documentstyle[12pt,aasms4]{aastex} % para submeter PcTEx
5: %\documentstyle[11pt, aaspp4]{aastex} % para 2 colunas UNIX *** Iam using this ***
6: %\documentstyle[11pt, aaspp4]{article} % para preprint UNIX
7: %\documentstyle[11pt, aaspp4]{aastex} % para preprint PCtEx
8: %\documentstyle[apjpt4]{article} % para tables
9:
10: \shorttitle{Mg isotopes in Halo Dwarfs}
11: \shortauthors{Mel\'endez \& Cohen}
12:
13: \begin{document}
14: \title {Magnesium Isotopes in Metal-Poor Dwarfs, the Rise of AGB Stars
15: and the Formation Timescale of the Galactic Halo\altaffilmark{1}}
16:
17:
18: \newcommand{\teff}{T$_{\rm eff}$ }
19: \newcommand{\tsin}{T$_{\rm eff}$}
20: % Next 5 lines define \simless and \simgreat: "less than or approximately
21: % equal to" and "greater than or approximately equal to".
22: \newbox\grsign \setbox\grsign=\hbox{$>$} \newdimen\grdimen \grdimen=\ht\grsign
23: \newbox\simlessbox \newbox\simgreatbox
24: \setbox\simgreatbox=\hbox{\raise.5ex\hbox{$>$}\llap
25: {\lower.5ex\hbox{$\sim$}}}\ht1=\grdimen\dp1=0pt
26: \setbox\simlessbox=\hbox{\raise.5ex\hbox{$<$}\llap
27: {\lower.5ex\hbox{$\sim$}}}\ht2=\grdimen\dp2=0pt\def\simgreat{\mathrel{\copy\simgreatbox}}
28: \def\simless{\mathrel{\copy\simlessbox}}
29: % Next lines define "approximately proportional to"
30: \newbox\simppropto
31: \setbox\simppropto=\hbox{\raise.5ex\hbox{$\sim$}\llap
32: {\lower.5ex\hbox{$\propto$}}}\ht2=\grdimen\dp2=0pt
33: \def\simpropto{\mathrel{\copy\simppropto}}
34:
35: \author{Jorge Mel\'endez}
36: \affil{Research School of Astronomy \& Astrophysics,
37: Australian National University, Mt. Stromlo Observatory, Weston ACT 2611,
38: Australia} \email{jorge@mso.anu.edu.au}
39:
40: \and
41:
42: \author{Judith G. Cohen}
43: \affil{Palomar Observatory, MC 105--24, California Institute of Technology, Pasadena, CA
44: 91125}\email{jlc@astro.caltech.edu}
45:
46:
47: \altaffiltext{1}{The data presented herein were obtained at the W.M. Keck Observatory,
48: which is operated as a scientific partnership among the California Institute of Technology,
49: the University of California and the National Aeronautics and Space Administration.}
50:
51:
52: \slugcomment{Submitted to the The Astrophysical Journal}
53: \slugcomment{Send proofs to: J. Melendez}
54:
55: \begin{abstract}
56:
57: We have determined magnesium isotopic ratios ($^{25,26}$Mg/Mg) in metal-poor
58: ($-2.6 \leq$ [Fe/H] $\leq -1.3$)
59: halo dwarfs employing high S/N (90-280) high spectral resolution
60: (R = 10$^5$) Keck HIRES spectra.
61: Unlike previous claims of an important contribution from intermediate-mass
62: AGB stars at low metallicities, we find that the rise of the AGB contribution
63: in the Galactic halo did not occur until intermediate metallicities ([Fe/H] $\gtrsim -$1.5).
64: \end{abstract}
65:
66: \keywords{Stars: Population II - stars: AGB - stars: atmospheres - stars: abundances -
67: - Galaxy: halo}
68:
69:
70: \section{Introduction}
71:
72: Magnesium is composed of three stable isotopes $^{24}$Mg, $^{25}$Mg and $^{26}$Mg,
73: which can be formed in massive stars (e.g. Woosley \& Weaver 1995; hereafter WW1995).
74: The lightest isotope is formed as a primary isotope from H,
75: while $^{25,26}$Mg are formed as secondary isotopes.
76: The heaviest Mg isotopes are also produced in intermediate-mass
77: AGB stars (Karakas \& Lattanzio 2003),
78: so the isotopic ratios $^{25,26}$Mg/$^{24}$Mg increase
79: with the onset of AGB stars.
80: Therefore, Mg isotopic ratios in halo stars
81: could be used to constrain the rise of AGB stars in our Galaxy.
82:
83: It is important to know when AGB stars begin to enrich the halo
84: in order to disentangle the contribution of elements produced by
85: intermediate-mass stars from those produced by masssive stars.
86: For example, the high nitrogen abundances observed in metal-poor stars
87: can be explained by fast-rotating massive stars
88: (Chiappini, Matteucci \& Ballero 2005; Chiappini et al. 2006)
89: or alternatively by intermediate-mass stars, although
90: the latter option may be unlikely because those stars
91: may not have had time to enrich the halo due to their longer lifetime.
92:
93: Mg isotopic abundances can be obtained from the analysis
94: of MgH lines in cool stars.
95: After the early work of Boesgaard (1968) and Bell \& Branch (1970),
96: other studies have increased the coverage in metallicity
97: down to [Fe/H] = $-$1.8 (Tomkin \& Lambert 1980; Lambert \& McWilliam 1986;
98: Barbuy 1985, 1987; Barbuy, Spite \& Spite 1987; McWilliam \& Lambert 1988;
99: Gay \& Lambert 2000).
100:
101: In order to reach lower metallicities ([Fe/H] $< -$2), very metal-poor
102: cool dwarfs have to be discovered. This work was undertaken
103: by Yong \& Lambert (2003a,b), who found a number of metal-poor ([Fe/H] $< -$2)
104: cool dwarfs (\teff $<$ 5000 K) useful for Mg isotopic studies.
105: Employing that sample, Yong, Lambert \& Ivans (2003, hereafter YLI2003)
106: were able to study $^{25,26}$Mg/$^{24}$Mg ratios down to [Fe/H] = $-$2.5.
107: Surprisingly, they found metal-poor stars with
108: relatively high $^{25,26}$Mg/$^{24}$Mg ratios, suggesting thus
109: an important contribution by intermediate-mass AGB stars even
110: at such low metallicities (Fenner et al. 2003, hereafter F2003).
111:
112: In this work, we determine Mg isotopic ratios in
113: cool halo dwarfs and constrain the rise of intermediate-mass AGB stars
114: by comparing the observed ratios with chemical evolution models.
115:
116:
117: \section{Sample stars and Observations}
118:
119: The sample was selected from previous spectroscopic analyses
120: of metal-poor cool dwarfs (Yong \& Lambert 2003a,b).
121: Five metal-poor stars were chosen
122: covering the range $-2.6 \leq$ [Fe/H] $\leq -$1.3:
123: G 69-18 (LHS 1138), G 83-46 (LHS 1718), G 103-50,
124: G 63-40 (LHS 2765) and the well-known moderately metal-poor dwarf HD 103095,
125: with [Fe/H] = $-$2.6, $-$2.6, $-$2.2, $-$1.9 and $-$1.3, respectively.
126:
127: The observations were obtained with HIRES (Vogt et al. 1994) at the
128: Keck I telescope.
129: The first set of spectra was taken in August 2004, just a few days after
130: a HIRES upgrade, taking thus advantage of improvements in efficiency,
131: spectral coverage and spectral resolution. A resolving power of
132: R $\approx 10^5$ was achieved using a 0.4\arcsec-wide slit.
133: Additional observations were obtained in November 2004 and June 2005.
134:
135: The spectral orders were extracted with MAKEE\footnote{MAKEE was developed
136: by T.A. Barlow specifically for reduction of Keck HIRES data. It is
137: freely available at http://www2.keck.hawaii.edu/inst/hires/data\_reduction.html}
138: and IRAF was used for further data reductions (Doppler correction,
139: continuum normalization and combining spectra).
140:
141: Two sample stars (G 83-46 and G 103-50) turned out to be
142: double lined stars, with G 103-50 being a spectroscopic binary
143: (Latham et al. 1988).
144: These two stars were discarded from the analysis.
145:
146:
147: \section{Atomic and Molecular Data}
148:
149: Three wavelength regions at 5134.6, 5138.7 and 5140.2 \AA\ are
150: usually employed to determine the isotopic abundance ratios
151: $^{25,26}$MgH/$^{24}$MgH (e.g. McWilliam \& Lambert 1988;
152: YLI2003). For these regions we adopted laboratory FTS measurements
153: of the isotopic $^{24,25,26}$MgH lines obtained by
154: Bernath, Black \& Brault (1985).
155:
156: In addition, outside the recommended regions, laboratory wavenumbers
157: for $^{24}$MgH were taken from Bernath et al. (1985), and the
158: corresponding $^{25,26}$MgH line positions were computed by
159: adding the theoretical isotopic shifts to the laboratory $^{24}$MgH wavenumbers.
160: The $^{25,26}$MgH isotopic shifts were calculated using the
161: relative reduced mass of $^{25,26}$MgH to $^{24}$MgH.
162:
163: The energy levels were calculated using molecular constants by
164: Shayesteh et al. (2004) and we adopted a dissociation energy of
165: 1.27 eV (Balfour \& Lindgren 1978). Oscillator strengths were obtained
166: from transition probabilities given by Weck et al. (2003).
167:
168: Molecular C$_2$ lines are also present in the same region, so a line list
169: of C$_2$ lines was also implemented. Laboratory wavenumbers were
170: taken from Amiot (1983) and Prasad \& Bernath (1994).
171: The rotational strengths (H\"onl-London factors) were computed
172: following Kovacs (1961), and we adopted an oscillator band strength of
173: $f_{00}$ = 0.03 (see Grevesse et al. 1991).
174: Excitation potentials were computed using the molecular
175: constants by Prasad \& Bernath (1994) and a dissociation energy of
176: 6.297 eV (Urdahl et al. 1991) was adopted.
177:
178: Atomic lines present in the region were also included.
179: The initial line list was based in the work of Barbuy (1985),
180: and lines were added or discarded based on spectral synthesis
181: of both the Sun and Arcturus spectra.
182:
183: In previous works the macroturbulence has been determined mainly
184: using two lines: \ion{Ni}{1} 5115.4 \AA\ and \ion{Ti}{1} 5145.5 \AA\
185: (e.g. McWilliam \& Lambert 1988; YLI2003). For the two more metal-poor
186: stars in our sample these lines became too weak, so we
187: additionally used lines of \ion{Fe}{1} and \ion{Ca}{1} present
188: in the 5569-5601 \AA\ region.
189:
190: For the atomic lines we adopted transition probabilities from
191: the NIST database\footnote{http://physics.nist.gov/PhysRefData/ASD/};
192: astrophysical {\it gf}-values were derived when no entry was available.
193:
194: \section{Spectral Synthesis Analysis}
195:
196: The stellar parameters (\tsin, log g, [Fe/H], v$_t$)
197: were initially adopted from Mel\'endez \& Barbuy (2002) for HD 103095,
198: and from Yong \& Lambert (2003b) for the other two stars. A check of the
199: stellar parameters was done employing the IRFM \teff calibrations by
200: Ram\'{\i}rez \& Mel\'endez (2005), Hipparcos parallaxes, Y$^2$ isochrones
201: (Demarque et al. 2004) and our HIRES spectra. E(B-V) was estimated
202: both using interstellar \ion{Na}{1} D lines and reddening maps ($\S$4.1 of Mel\'endez et al. 2006).
203: Reasonable agreement was found with respect to the stellar parameters
204: given in the above references. Our final adopted values are given in Table 1.
205:
206: Once the stellar parameters were set, the macroturbulence was determined
207: employing the \ion{Ni}{1} 5115.4 \AA\ and \ion{Ti}{1} 5145.5 \AA\ lines,
208: as well as \ion{Fe}{1} and \ion{Ca}{1} lines around 5569-5601 \AA.
209:
210: The contribution of C$_2$ lines was constrained by spectral synthesis of the
211: weak feature around 5135.7 \AA, which is a blend of C$_2$ lines
212: (5135.57 and 5135.69 \AA) sometimes blended with an unidentified line
213: in the red side (Gay \& Lambert 2000).
214: Fortunately the observations are of such high resolution that it is
215: possible to constrain the contribution of C$_2$ employing the
216: blue side of this feature, imposing thus an upper limit to blends by C$_2$ lines.
217:
218: The Mg isotopic ratios were determined using spectral synthesis.
219: After the first trials it was clear that the $^{25,26}$Mg isotopic
220: ratios were lower than 5\%, i.e., much lower than the
221: terrestrial ratios (79:10:11). The computed synthetic spectra
222: have isotopic ratios ranging from $^{24}$Mg:$^{25}$Mg:$^{26}$Mg=100:0:0 to 90:5:5.
223:
224: Initially the Mg isotopic ratios were determined by an eye-fit of the
225: synthetic spectra to the HIRES observed data of the three recommended regions
226: (see $\S$3). The results are shown in Table 1.
227: After the fits by eye were completed we performed a $\chi^2$ fit by computing
228: $\chi^2$ = $\Sigma$($O_i - S_i$)/$\sigma^2$, where $O_i$ and $S_i$ represents
229: the observed and synthetic spectrum, respectively, and $\sigma$ = (S/N)$^{-1}$.
230: As an example of the $\chi^2$ fits we show in Fig. 1 the fits for
231: the recommended region at 5140.2 \AA\ in the most metal-rich
232: and most metal-poor stars of our sample.
233: The results of the $\chi^2$ fits are shown in Table 1. As can be seen,
234: the eye fit compares well to the $\chi^2$ fit. The errors given in Table 1
235: are due to statistical errors (the standard deviation between the isotopic ratios
236: of the three recommended regions) and systematic errors of 1\% (due to errors
237: in the atmospheric parameters, see e.g. YLI2003).
238: Our results for HD 103095 ($^{24}$Mg:$^{25}$Mg:$^{26}$Mg = 94.8:2.4:2.8)
239: compare very well with previous visual (eye-fit)
240: determinations in the literature.
241: For HD 103095 both Tomkin \& Lambert (1980) and Barbuy (1985)
242: obtained isotopic ratios of 94:3:3.
243: More recently Gay \& Lambert (2000) determined 93:4:3.
244:
245:
246: \section{Discussion}
247:
248: Here we discuss how our isotopic ratios compare with chemical evolution models.
249: We compare only the ratio $^{26}$Mg/$^{24}$Mg, since the isotopic ratio
250: for $^{25}$Mg is more uncertain due to the smaller isotopic shift.
251:
252: In Fig. 2 (left panel) we compare our results with the models computed
253: by F2003, both including and neglecting the contribution of
254: intermediate-mass AGB stars. Another model (Alib\'es, Labay \& Canal 2001, hereafter ALC2001)
255: which includes only massive stars is also shown.
256: A comparison with other models (Ashenfelter, Mathews \& Olive 2004, hereafter AMO2004;
257: Goswami \& Prantzos 2000) is shown in the right panel.
258: As can be seen, our low isotopic ratios can be explained mostly
259: by massive stars, thus we find no need to invoke the contribution of
260: intermediate-mass AGB stars at low metallicities. HD 103095 lies slightly
261: above the predicted F2003 curve of massive stars nucleosynthesis (although
262: it is in perfect agreement with the ALC2001 model), so this may
263: indicate that at [Fe/H] $\gtrsim -$1.5 the contribution from AGB stars begins.
264:
265: The high isotopic ratios found in metal-poor stars by YLI2003
266: were interpreted by F2003 as an important contribution of
267: intermediate-mass AGB stars at low metallicities. However, most of the stars with high
268: isotopic $^{25, 26}$Mg/Mg ratios in YLI2003 are not
269: bona fide halo dwarfs. We computed the probability of halo membership following
270: Bensby, Feltzing \& Lundstr\"om (2003)
271: and found that a fraction of the metal-poor stars in
272: YLI2003 are actually thick disk stars. Furthermore,
273: some halo stars have abundance anomalies (e.g. CH stars) or their
274: spectra are abnormal (e.g. double lined), and they should be removed
275: for a fair comparison with chemical evolution models.
276:
277: After eliminating the probable thick disk stars, as well as
278: halo stars with anomalies, we find that only 4 bona fide halo dwarfs remain
279: from the YLI2003 sample:
280: G 39-36, LHS 3780, G 113-40 and G 86-39.
281: As can be seen in Fig. 2 (right panel), the results of YLI2003
282: are in excellent agreement with ours\footnote{YLI2003 report that
283: G 39-36 was observed with a resolving power of R = 60 000, but their
284: other 3 metal-poor dwarfs were observed with a lower R = 35 000.
285: For G 39-36 we adopted the typical error of 3\% quoted by YLI2003,
286: but for the other 3 dwarfs in their sample we increased the error to 4\%
287: due to the lower resolving power of the observations}.
288: We have done a similar exercise with the sample of Gay \& Lambert (2000)
289: and found that the only good unevolved halo star is HD 103095, which is
290: already included in our sample. Lambert \& McWilliam (1986) have analyzed
291: the metal-poor ([Fe/H] = $-$1.5) subgiant $\nu$ Ind, for which they
292: obtained only upper limits of $^{25, 26}$Mg/Mg $\leq$ 3\%.
293:
294: Thus both our results and YLI2003 suggest a
295: small (or none) $^{25, 26}$Mg contribution of intermediate-mass AGB stars to the Galactic halo.
296: Perhaps the $^{25, 26}$Mg yields from AGB stars are lower
297: than in current models (Karakas \& Lattanzio 2003).
298: If this is the case, then intermediate-mass
299: AGB stars can not be invoked to explain the possible variation of the
300: fine-structure constant $\alpha$ (AMO2004).
301: The chemical evolution models used to
302: explain variations in $\alpha$ require an ad-hoc AGB-enhanced IMF
303: in order to produce large amounts of $^{25, 26}$Mg (AMO2004).
304: If the correct yields are lower than present calculations,
305: then much larger ad-hoc modifications to the IMF would
306: be required.
307:
308: Calculations of Karakas \& Lattanzio (2003) show that
309: the AGB stars that contribute significant amounts of $^{25, 26}$Mg are stars with
310: initial masses of 3-6 M$_\odot$. Since these stars have lifetimes
311: considerably shorter than the age of the universe, they can be
312: used to constrain the timescale for the formation of the
313: Galactic halo. According to the
314: Padova evolutionary tracks\footnote{http://pleiadi.pd.astro.it/}
315: the lifetime of 3-6 M$_\odot$ metal-poor stars ([Fe/H] = $-$1.5)
316: are 0.1-0.3 Gyr, so the halo timescale formation should be
317: of the order of 0.3 Gyr. This short timescale probably explains why
318: recent studies of age spread in Galactic globular clusters have
319: shown that most clusters from intermediate to low metallicity are
320: coeval within the uncertainties (e.g. Rosenberg et al. 1999; De Angeli et al. 2005).
321:
322: \section{Conclusions}
323:
324: We have shown that the $^{25, 26}$Mg/Mg ratios in halo
325: dwarfs are low and that there is no need to invoke a
326: contribution from intermediate-mass AGB stars at low metallicities.
327:
328: Further high S/N high spectral resolution observation
329: of a larger sample will help constrain the rise of AGB stars
330: in the Galaxy and will be useful to better constrain the
331: formation timescale of the Galactic halo.
332:
333: \acknowledgements
334:
335:
336: JM thanks C. Chiappini for useful discussions and
337: D. Yong for providing data to check our MgH line list,
338: as well as for useful discussions on Mg isotopic ratios.
339: JGC is grateful for partial support to NSF grant AST-0507219.
340: We have made use of data from the SIMBAD database operated at CDS.
341: The authors wish to recognize and acknowledge the very significant
342: cultural role and reverence that the summit of Mauna Kea has always had
343: within the indigenous Hawaiian community.
344: We are most fortunate to have the opportunity to conduct observations from this mountain.
345:
346: \begin{references}
347:
348: \reference{} Alib\'es, A., Labay, J., Canal, R. 2001, A\&A, 370, 1103 (ALC2001)
349:
350: \reference{} Amiot, C. 1983, ApJS, 52, 329
351:
352: \reference{} Ashenfelter, T., Mathews, G. J., Olive, K. A. 2004, ApJ, 615, 82 (AMO2004)
353:
354: \reference{} Balfour, W. J. \& Lindgren, B. 1978, Can. J. Phys., 56, 767
355:
356: \reference{} Barbuy, B. 1985, A\&A, 151, 189
357:
358: \reference{} Barbuy, B. 1987, A\&A, 172, 251
359:
360: \reference{} Barbuy, B., Spite, F., Spite, M. 1987, A\&A, 178, 199
361:
362: \reference{} Bell, R. A. \& Branch, D. 1970, ApL, 5, 203
363:
364: \reference{} Bensby, T., Feltzing, S., Lundstr\"om, I. 2003, A\&A, 410, 527
365:
366: \reference{} Bernath, P. F., Black, J. H., Brault, J. W. 1985, ApJ, 298, 375
367:
368: \reference{} Boesgaard, A. M. 1968, ApJ, 154, 185
369:
370: \reference{} Chiappini, C., Matteucci, F. \& Ballero, S. K. 2005, A\&A, 437, 429
371:
372: \reference{} Chiappini, C. et al. 2006, A\&A, 449, L27
373:
374: \reference{} De Angeli, F. et al. 2005, AJ, 130, 116
375:
376: \reference{} Demarque, P., Woo, J.-H., Kim, Y.-C., Yi, S. K.
377: 2004, ApJS, 155, 667
378:
379: \reference{} Fenner, Y. et al. 2003, PASA, 20, 340 (F2003)
380:
381: %\reference{} Fenner, Y., Murphy, M. T. \& Gibson, B. K. 2005, MNRAS, 358, 468
382:
383: \reference{} Gay, P. L. \& Lambert, D. L. 2000, ApJ, 533, 260
384:
385: \reference{} Goswami, A. \& Prantzos, N. 2000, A\&A, 359, 191
386:
387: \reference{} Grevesse, N. et al. 1991, A\&A, 242, 488
388:
389: \reference{} Karakas, A. I., Lattanzio, J. C. 2003, PASA, 20, 279
390:
391: \reference{} Kovacs, I. 1961, Rotational structure in the spectra of Diatomic
392: Molecules (London: Hilger)
393:
394: \reference{} Lambert, D. L. \& McWilliam, A. 1986, ApJ, 304, 436
395:
396: \reference{} Latham, D. W. et al. 1988, AJ, 96, 567
397:
398: \reference{} McWilliam, A. \& Lambert, D. L. 1988, MNRAS, 230, 573
399:
400: \reference{} Mel\'endez, J. \& Barbuy, B. 2002, ApJ, 575, 474
401:
402: \reference{} Mel\'endez, J., Shchukina, N. G.,
403: Vasiljeva, I. E. \& Ram\'{\i}rez, I. 2006, ApJ, 642, 1082
404:
405: \reference{} Prasad, C. V. V. \& Bernath, P. F. 1994, ApJ, 426,812
406:
407: \reference{} Ram\'{\i}rez, I. \& Mel\'endez, J. 2005, ApJ, 626, 465
408:
409: \reference{} Rosenberg, A., Saviane, I., Piotto, G., Aparicio, A. 1999, AJ, 118, 2306
410:
411: \reference{} Shayesteh, A. et al. 2004, J. Chem. Phys., 120, 10002
412:
413: \reference{} Urdahl, R. S., Bao, Y., Jackson, W. M. 1991,
414: Chem. Phys. Letters, 178, 425
415:
416: \reference{} Tomkin, J. \& Lambert, D. L. 1980, ApJ, 235, 925
417:
418: \reference{} Vogt, S. S., et al. 1994, Proc. SPIE, 2198, 362
419:
420: \reference{} Weck, P. F. et al. 2003, ApJ, 582, 1059
421:
422:
423: \reference{} Woosley, S. E.\& Weaver, T. A. 1995, ApJS, 101, 181 (WW1995)
424:
425: \reference{} Yong, D. \& Lambert, D. L. 2003a, PASP, 115, 22
426:
427: \reference{} Yong, D. \& Lambert, D. L. 2003b, PASP, 115, 796
428:
429: \reference{} Yong, D., Lambert, D. L., Ivans, I. I. 2003, ApJ, 599, 1357 (YLI2003)
430:
431: \end{references}
432:
433:
434: \clearpage
435:
436: \begin{deluxetable}{lllllllllllll}
437: %\tablenum{1}
438: %\tablewidth{0pt}
439: \footnotesize
440: \tablecaption{Atmospheric Parameters and Mg isotopic ratios}
441: \tablehead{
442: \colhead{ID} & \colhead{$E_{B-V}$} & \colhead{\teff} &
443: \colhead{log $g$} & \colhead{[Fe/H]} & \colhead{v$_{mic}$} & \colhead{v$_{mac}$} &
444: \multicolumn{2}{c}{$\chi^2$ fit*} & \multicolumn{2}{c}{eye fit*} \\
445: \colhead{} & \colhead{(mag)} & \colhead{(K)} &
446: \colhead{(dex)} & \colhead{(dex)} & \colhead{(km s$^{-1}$)} & \colhead{(km s$^{-1}$)} &
447: \colhead{$^{25}$Mg (\%)} & \colhead{$^{26}$Mg (\%)} &
448: \colhead{$^{25}$Mg (\%)} & \colhead{$^{26}$Mg (\%)} }
449: \startdata
450: HD 103095 & 0.000 & 5010 & 4.60 &-1.35 & 0.5 & 3.0 & 2.4$\pm$1.3 & 2.8$\pm$1.6 & 3.5$\pm$1.5 & 3.3$\pm$1.5 \\
451: G 63-40 & 0.005 & 4686 & 4.81 &-1.86 & 0.3 & 2.0 & 0.5$\pm$2.2 & 1.2$\pm$1.4 & \nodata & 1.0$\pm$2.0\\
452: G 69-18 & 0.030 & 4480 & 4.75 &-2.60 & 0.3 & 1.5 & 0.9$\pm$2.0 & 1.5$\pm$2.2 & \nodata & 1.0$\pm$2.0\\
453: \enddata
454: \tablenotetext{*}{Mg isotopic ratios are given with respect to $^{24}$Mg+$^{25}$Mg+$^{26}$Mg
455: and are expressed as percentages.
456: }
457: \label{tabiso}
458: \end{deluxetable}
459:
460: \begin{figure}
461: \epsscale{}
462: \plottwo{f1a.eps}{f1b.eps}
463: \caption{Fits for the 5140.2 \AA\ region in the stars HD 103095 and G 69-18.
464: Observed spectra are represented with filled circles, and synthetic
465: spectra with solid lines. The calculations were performed for
466: $^{25,26}$Mg/Mg ratios of 0-5\%. The relative variation of the $\chi^2$ fits
467: are shown as a function of the isotopic abundance.
468: }
469: \label{mghfit}
470: \end{figure}
471:
472:
473: \begin{figure}
474: \epsscale{}
475: \plottwo{f2a.eps}{f2b.eps}
476: \caption{$^{26}$MgH/$^{24}$MgH as a function of [Fe/H] in halo dwarfs.
477: Filled and open circles represent our results and those by
478: YLI2003, respectively. All models include yields of massive stars
479: (mostly by WW1995). Models including massive stars and intermediate-mass
480: AGB stars (F2003 and AMO2004) are also shown.
481: Note that AMO2004 extrapolated AGB yields for Z $<$ 0.004,
482: so their results may be unrealiable.
483: The model that agrees better
484: with the observed data is the ALC2001 model, which does not include
485: intermediate-mass stars.
486: }
487: \label{mghfeh}
488: \end{figure}
489:
490:
491: \end{document}
492: