1: %\documentclass[preprint]{/opt/local/aastex/aastex-5.0.2/aastex}
2: \documentclass[preprint,12pt]{aastex}
3: %\usepackage{/opt/local/aastex/aastex-5.0.2/aastexug}
4: %\documentstyle[12pt,aasms4]{article}
5: %\documentstyle[11pt,aaspp4]{article}
6: %\received{}
7: %\accepted{}
8: %\journalid{}{}
9: %\articleid{}{}
10: %\slugcomment{}
11:
12: \lefthead{Cunha et al.}
13: \righthead{Fluorine Abundances in Orion}
14:
15: \received{2005 January 4}
16: \begin{document}
17:
18: \title{Fluorine Abundances in the Orion Nebula Cluster}
19:
20: \author{Katia Cunha}
21: \affil{Observat\'orio Nacional, Rua General Jos\'e Cristino 77, 20921-400,
22: S\~ao Crist\'ov\~ao, Rio de Janeiro, Brazil; katia@on.br}
23:
24: \author{Verne V. Smith}
25: \affil{National Optical Astronomy Observatory, P.O. Box 26732, Tucson,
26: AZ 85726, USA; vsmith@noao.edu}
27:
28: \begin{abstract}
29:
30: This is a pilot study using cool dwarfs as sources with which to
31: probe fluorine abundances via HF. This molecule is detected
32: for the first time in young K-M dwarf members of an OB association.
33: The targets are three low-mass stars (JW22, JW163 and JW433)
34: belonging to the Orion Nebula Cluster.
35: The target stellar parameters were derived to be T$_{\rm eff}$=3650K,
36: 4250K, and 4400K, and log g =3.5, 3.4 and 3.6, with corresponding
37: stellar masses of 0.4, 0.6 and 0.7M$_{\odot}$, for JW22, JW163 and JW433,
38: respectively.
39: Fluorine, oxygen, and carbon abundances were derived from the HF(1--0) R9 line
40: along with samples of OH and CO vibration-rotation lines present in
41: high-resolution infrared spectra
42: observed with the Phoenix spectrograph on the Gemini South Telescope.
43: The fluorine and oxygen results obtained for these targets,
44: still in the pre-main-sequence stage of evolution,
45: agree well with the general trend defined for the Milky Way disk; the latter
46: being deduced from
47: observations of more evolved giant stars. In addition,
48: the carbon and oxygen abundances obtained for the studied
49: stars overlap results from previous studies of the more
50: massive OB stars and FG dwarf members of the Orion Nebula Cluster.
51: We conclude from this agreement
52: that the fluorine abundances derived for the
53: Orion K-M dwarfs (when there is no conspicuous evidence of disks)
54: can be considered a good
55: representation of the current fluorine abundance value for the Milky Way disk.
56:
57: \end{abstract}
58:
59: \keywords{nucleosynthesis--stars: abundances
60: }
61:
62: \section{Introduction}
63:
64: The chemical elements are produced via a variety
65: of nuclear and high-energy processes associated with stars
66: of various masses, cosmic rays, and the big bang itself.
67: One interesting process, whose importance within the
68: overall scheme of nucleosynthesis is not yet well-quantified,
69: involves the inelastic scattering of neutrinos
70: off of nuclei and is referred to as neutrino nucleosynthesis,
71: or the $\nu$-process.
72: Neutrino nucleosynthesis was discussed by
73: Domogatsky, Eramzhyan, \& Nadyozhin (1978), who focussed mainly
74: on its role in the production of the lightest elements, such as
75: $^{2}$H, $^{3}$He, $^{7}$Li, or $^{11}$B. Later, Woosley \& Haxton
76: (1988) expanded on the role of the $\nu$-process in the production
77: of fluorine and subsequently, Woosley et al. (1990) published predicted yields
78: for a number of species, including Li and F.
79: The neutrinos involved in the $\nu$-process
80: are produced as a result of the gravitational collapse of a
81: stellar core to a neutron star during a supernova of Type II
82: (SN II).
83: The $\nu$-process may result in significant synthesis of
84: certain low-abundance isotopes that lie one mass unit below
85: abundant nuclei, such as $^{12}$C giving rise to $^{11}$B,
86: or $^{20}$Ne resulting in $^{19}$F via neutrino-induced spallation.
87: It is the association of $^{19}$F, the only stable isotope of
88: fluorine, with the $\nu$-process that makes its abundance of
89: interest as a potential probe of neutrino nucleosynthesis.
90:
91: In addition to the $\nu$-process, $^{19}$F is possibly produced
92: during He-burning thermal pulses on the asymptotic giant
93: branch (AGB), as first suggested by Forestini et al. (1992).
94: Meynet \& Arnould (2000) also pointed out
95: that Wolf-Rayet stars, via the same sets of reactions as in
96: the AGB stars, may also produce significant amounts of $^{19}$F,
97: with the caveat that large dM/dt stellar winds must remove
98: a significant amount of mass, exposing the
99: inner layers where fluorine is produced before it is burned away.
100: Observations of fluorine abundances, spanning a range of
101: stellar metallicities and across a variety of stellar populations,
102: are needed in order to sort out relative contributions from AGB
103: stars, Wolf-Rayet stars, or neutrino nucleosynthesis.
104:
105: In this paper, we use high-resolution IR spectra from Phoenix
106: on Gemini-South to derive $^{19}$F abundances (from HF), along
107: with $^{12}$C abundances (from CO), and
108: $^{16}$O abundances (from OH) in three low-mass members of the
109: Orion Nebula cluster.
110: Carbon and oxygen provide comparison abundances to those
111: derived from fluorine.
112: Both C and O have been studied
113: previously in Orion members by Cunha \& Lambert (1994); their
114: study analyzed hot O and B stars. In addition, some oxygen
115: abundances in Orion-member F and G main-sequence and pre-main-sequence
116: stars have been published by Cunha, Smith, \& Lambert (1998).
117: These previous studies provide a background on which the abundances
118: derived from K and M type stars can be compared. Apart from the interest
119: in the carbon and oxygen abundances in young stellar systems, C and O
120: provide important comparisons for the fluorine abundances.
121: The Orion Nebula cluster stars are very young and will reflect
122: the current Galactic F/O ratio in the disk. Also, previous stellar
123: $^{19}$F abundances (derived from HF) have come only from red giants
124: (Jorissen, Smith, \& Lambert 1992; Cunha et al. 2003), while here we
125: undergo a pilot study to use cool dwarf stars as sources with which
126: to probe fluorine abundances.
127:
128: \section{Observations}
129:
130: %various $^{12}$C$^{16}$O lines from which carbon abundances can
131: %be derived with $^{16}$OH being studied via OH transitions that fall
132: %near 1.5 $\mu$m.
133:
134: The target stars for this study are K and M dwarfs selected from various
135: studies of the Orion Nebula cluster
136: by Hillenbrand (1997), Hillenbrand \& Hartmann (1998),
137: Hillenbrand et al. (1998), and Rhode, Herbst, \& Mathieu (2001).
138: In particular, the stars were selected for high membership
139: probability (P$\ge$ 93\%), to be cool enough for the HF molecular
140: line to be detectable (cooler than T$_{\rm eff}$ of 4500K),
141: and to have relatively low values of projected rotational
142: velocities (vsin(i)$\le$15 km s$^{-1}$).
143: The selected targets were such that two of them
144: (JW22 and JW433) do not show evidence of flux from a circumstellar disk,
145: while JW163 shows such evidence (primarily from near-IR
146: color excesses from Hillenbrand et al. 1998). Our sample, although small, will allow us to
147: ascertain possible effects of disk emission on an abundance analysis
148: of the stellar photosphere in these young stars.
149:
150: High-resolution infrared spectra of these low-mass stars
151: were obtained with the Gemini South
152: Telescope plus the Phoenix spectrograph (Hinkle et al. 1998) during
153: observing runs in classical mode on 27-28 December 2002 and
154: 20 December 2004.
155: The observed data are single order echelle spectra with a
156: resolution R=$\lambda$/$\Delta$$\lambda$=50,000, corresponding
157: to a resolution element of $\sim$ 4-pixels.
158: In order to obtain two separate spectral regions, one around the H-band
159: (at $\sim$ 15,500\AA, containing $^{16}$OH lines) and another around the
160: K-band (at $\sim$ 23,400\AA, containing the molecular HF(1--0) R9 line
161: and numerous $^{12}$C$^{16}$O lines), two grating tilts were needed. JW22 was observed
162: only in the K-band region.
163:
164: The Orion targets are relatively bright in the near-infrared
165: so that the exposure times
166: needed to obtain co-added spectra with signal-to-noise ratios in
167: excess of $\sim$50--100 were not exceedingly long: for each grating tilt,
168: four target exposures
169: of 15 minutes each were obtained. We adopted the strategy of
170: varying the target position
171: on the slit (positions a-b-b-a) and subtracting one image from another
172: in order to remove dark current and fixed-pattern noise.
173: In addition, for those observations corresponding to the
174: spectral region of the K-band,
175: we also observed hot stars at roughly the same air masses as the target
176: stars.
177: These hot star spectra were needed in order to remove
178: telluric lines from the target star spectra.
179: All the obtained data were reduced to one-dimensional spectra using
180: standard IRAF routines. More details on the observations and
181: reduction procedures adopted can be found in Smith et al. (2004).
182:
183: Phoenix spectra are shown in Figure 1 for two of the target stars, with
184: the two spectra shifted vertically in relative flux. The spectra
185: plotted in this figure were obtained in the december 2002 Gemini run. The top panel
186: shows the spectra in the H band, while the bottom spectra are in the
187: K band. Some of the prominent absorption lines are identified in the figure,
188: with the 15550\AA\ region showing a number of OH lines and the 23400\AA\ spectrum
189: containing numerous $^{12}$C$^{16}$O lines.
190:
191: \section{Stellar Parameters}
192:
193: The effective temperatures for the studied stars were derived from
194: their photometric colors (V-I) and (V-K) and by adopting the empirical
195: calibrations from Bessel, Castelli \& Plez (1998).
196: (V-I) and (V-K) colors are considered to be good temperature
197: indicators for cool stars
198: and have little sensitivity to gravity for the effective
199: temperature range considered in this study. The adopted photometric
200: magnitudes are presented in Table 1.
201: We obtained V-magnitudes and (V-I) colors for our targets from
202: the work by Hillenbrand (1997) that studied the stellar
203: population comprising the Orion Nebula cluster.
204: The target star K-magnitudes were obtained from the
205: Two Micron All Sky Survey (2MASS) database and corrected
206: to the Bessel et al. system
207: by means of the transformation equations presented in Carpenter (2001).
208: De-reddenned colors were
209: computed by adopting A$_{\rm v}$=0 for JW163, A$_{\rm v}$=1.01 mag
210: for JW433, and
211: A$_{\rm v}$=0.8 for JW22 (all values are taken from Hillenbrand 1997).
212:
213: Surface gravities (log g's) for the target stars were calculated from
214: the following standard relation:
215:
216: g/g$_{\odot}$ = (M/M$_{\odot}$) (T$_{eff}$/T$_{\odot}$)$^{4}$
217: (L/L$_{\odot}$)$^{-1}$.
218:
219: The stellar luminosities were taken from Hillenbrand (1997;
220: log L/L$_{\odot}$= 0.22, 0.43, and -0.34 for JW163, JW433, and JW22,
221: respectively). These were derived
222: assuming a distance to the Orion Nebula of 470 pc. As discussed in
223: Hillenbrand (1997), the adopted stellar luminosities are estimated to
224: have an uncertainty of 0.12 in log L/L$_{\odot}$.
225: The effective temperatures adopted to derive log g's were
226: those listed in Table 1.
227: Stellar masses could then be estimated from placement of the
228: targets on an HR diagram
229: and noting their position relative to
230: evolutionary tracks computed by
231: D'Antona \& Mazzitelli (1994).
232: %The following stellar masses were obtained:
233: %M/M$_{\odot}$=0.4, 0.6 and 0.7 for JW22, JW163 and JW433, respectively.
234: Surface gravities were then computed for the sample stars. The derived masses and log g's
235: are presented in Table 1. From the uncertainties in the
236: adopted stellar luminosities
237: and effective temperatures we estimate that our derived surface gravities
238: have a maximum uncertainty of $\sim$0.27 dex.
239:
240: \section{Analysis}
241:
242: LTE fluorine, as well as carbon and oxygen abundances were
243: obtained from spectrum
244: synthesis using the synthesis code MOOG (Sneden, 1973).
245: The model atmospheres employed in the analysis were calculated with
246: the ATLAS9 code (Kurucz 1994) for solar metallicity, adopting a
247: mixing length l=1.25 H$_{p}$ and without convective overshooting.
248:
249: \subsection{Fluorine, Carbon and Oxygen Abundances}
250:
251: Fluorine abundances can be derived from the vibration-rotation
252: lines of HF present in the spectral region around the K-band. In this study,
253: we used the HF(1--0) R9 line at 23357.661\AA. In this same IR spectral
254: window where HF was observed,
255: several other molecular lines that correspond to
256: $^{12}$C$^{16}$O
257: first-overtone vibration rotation lines
258: (3--1 vibration series) are also found and these can be used to obtain
259: the carbon
260: abundances, given that the oxygen abundances are known from other
261: transitions (such as OH).
262: Synthetic spectra were calculated across the entire spectral
263: region of the observed
264: Phoenix spectra, or from $\sim$ 23340\AA\ to 23430 \AA.
265: The linelist adopted was the same as in our previous study of fluorine in
266: red-giants of the Large Magellanic Cloud (see Cunha et al. 2003).
267: The CO lines are typically used to derive the microturbulence in
268: cool giants, where they are much stronger (such as in Smith et al. 2002).
269: Although in the parameter range studied here,
270: the CO lines become weaker, they are still sensitive to the microturbulence
271: and were used in order to estimate this parameter. The adopted
272: values for the stars vary between $\xi$=0.5 and 1.0 km/s (Table 1).
273:
274: Figure 2 shows the synthetic and observed spectra for all three target
275: stars near the HF line. Three synthetic spectra, each with a different
276: fluorine abundance, are plotted in each panel in order to illustrate
277: the sensitivity of the HF line to the fluorine abundance. Just to the
278: blue of the HF(1--0) R9 line is a stellar photospheric line that is
279: probably due to H$_{2}$O. This same unidentified line is clearly visible
280: (in the same approximate relative strength) in the solar sunspot spectrum
281: published by Wallace \& Livingston (1992). They identify a number of
282: other H$_{2}$O lines in this spectral region that have similar strength.
283: Water is not included in the synthesis linelist here, and the offending
284: line blends just a small fraction of the blue edge of the HF line and
285: does not affect significantly the derived $^{19}$F abundance. If triatomic
286: H$_{2}$O is responsible for this line, then it should be much weaker in giant
287: stars of similar effective temperature, and this
288: can be seen in the spectal atlas of Arcturus (Hinkle, Wallace, \&
289: Livingston 1995); although having a similar effective temperature to
290: JW163 or JW433, this feature is not detectable in Arcturus, even though
291: the HF line is clearly visible. Indeed, a tentative identification
292: has been provided by Hinkle (2004) as
293: H$_{2}$O (021)[14,5,9]--(010)[13,11,2]
294: at an air wavelength of about 23356.4\AA.
295:
296: Oxygen abundances can be measured from first-overtone
297: vibration-rotation
298: OH lines present in the H-band region. Three isolated OH lines at 15560,
299: 15569 and 15572\AA\
300: were synthesized to derive oxygen abundances.
301: The studied OH lines and adopted linelist were the same as in
302: Smith et al. (2002).
303: These lines are expected
304: to be reliable oxygen
305: abundance indicators and relatively free from non-LTE effects but to
306: suffer modestly ($\sim$ 0.1 dex) from 3d-effects in solar-type stars
307: (Asplund 2005). However, at
308: the relatively low effective temperatures of the stars studied here,
309: the oxygen abundance differences
310: caused by the adoption of one-dimensional model atmospheres,
311: instead of three-dimensional
312: model atmospheres, are expected to be smaller
313: than at higher effective temperatures (around 6000K; Nissen et al. 2002).
314:
315: Overall, the derived abundances in JW163, relative to JW433 and JW22, are
316: systematically lower for the studied elements, with $^{12}$C 0.30-0.40 dex
317: lower, $^{16}$O 0.25 dex lower, and $^{19}$F 0.10 dex lower. Although
318: it is possible that this difference reflects an underlying abundance
319: spread in the Orion Nebula cluster, it is more likely that
320: the presence of a more substantial circumstellar disk in JW163
321: results in the photospheric absorption lines being filled in
322: by disk emission. It is worth noting that in the (V--I)-(V--K)
323: plane as shown in Figure 11 of Bessell et al. (1998), both JW433
324: and JW22 fall
325: right on the sequence defined by normal dwarf stars, while JW163
326: is displaced several tenths of a magntiude towards redder colors from
327: the standard dwarf relation. In JW433 and JW22, with little if any
328: measurable disk presence, the abundances are near-solar:
329: A($^{12}$C)=8.30, A($^{16}$O)=8.80, and A($^{19}$F)=4.65 for JW433,
330: while for JW22 the abundances are A($^{12}$C)=8.40 and
331: A($^{19}$F)=4.61. Moreover, as will be discussed in Section 5,
332: the carbon (for JW22 and JW433) and oxygen (for JW433)
333: abundances overlap nicely
334: with those values derived from the much hotter B-star members of
335: Orion Id (the sub-association to which the Orion Nebula cluster
336: belongs) by Cunha \& Lambert (1994). The results from JW163, along
337: with JW22 and JW433 illustrate the usefullness of using these cool
338: dwarfs for abundance studies, even those still not on the main sequence,
339: provided that targets are selected carefully, taking into account
340: their circumstellar environments.
341:
342: \subsection{Abundance Uncertainties}
343:
344: The final abundances derived for carbon, oxygen and fluorine are presented in Table 1.
345: The expected uncertainties in the derived effective temperatures and surface gravities
346: are 100K and 0.3 dex, respectively. The errors in the microturbulence parameters
347: are estimated to be around 0.5 km/s.
348: The sensitivity of the derived
349: C, O and F abundances to changes in temperature, gravity and microturbulence
350: for two target stars (the coolest and the hottest in our sample)
351: are presented in Table 2; the total errors ($\Delta$A) in the derived
352: abundances in each case are listed in column 6.
353: Overall, the studied transitions of HF, CO and OH are weak and relatively
354: insensitive to the microturbulence; the exception being the CO lines that are
355: moderately strong at T$_{\rm eff}$'s around 3650K. The fluorine abundances are mostly
356: sensitive to effective temperature, but also, to a lesser extent, to the surface
357: gravity. Concerning OH, the abundance errors are quite insensitive to changes
358: in surface gravity and microturbulence. The total abundance errors for oxygen are dominated
359: by T$_{\rm eff}$ uncertainties. (OH abundances were not calculated for JW22 in this study.)
360: The uncertainties in the derived carbon abundances are shown to have different sensitivities
361: at the "cool" and "hot" regimes studied here. These are more uncertain around 3600K
362: when compared to 4500K.
363:
364: \section{Discussion}
365:
366: Our larger goal is to investigate the behavior of
367: fluorine abundances with metallicity, which is taken here to be represented by
368: the element oxygen, as this element is
369: virtually a pure product of nucleosynthesis in massive stars that end
370: their evolution as core-collapse SN II.
371: In particular, in this study we focus on
372: the fluorine and oxygen contents of the pre-main-sequence stellar members
373: of the Orion Nebula Cluster.
374:
375: As mentioned in the introduction carbon and oxygen abundances derived from
376: molecular transitions (CO and OH) can be used as benchmarks with which
377: to gauge the fluorine abundances obtained from HF.
378: A comparison of carbon and oxygen abundances derived here for JW22
379: and JW433 with abundances from Cunha \& Lambert (1994) and
380: Cunha et al. (1998) is in order; we ignore JW163 in this comparison
381: as there is evidence that the absorption lines may be affected
382: measurably by emission from a circumstellar disk. The Orion
383: Association is divided into four subgroups, labelled Ia, Ib, Ic, and
384: Id. The subgroups are distributed roughly by age, with Ia being the
385: oldest, Id the youngest, and an age difference of about 10Myr.
386: The Orion Nebula stars analyzed in this study are members of Id. The abundances
387: in the two stars are A(C)=8.40 for JW22, and A(C)=8.30 and A(O)=8.80
388: for JW433. For comparison, Cunha \& Lambert (1994) studied three OB
389: stars in Id and their average (NLTE) abundances are A(C)=8.25 and
390: A(O)=8.85: the agreement between two K-M dwarfs and three O-B stars
391: is excellent. In addition, Cunha et al. (1998) have one F-dwarf that
392: is a Id member and its oxygen abundance is A(O)=8.86: again, excellent
393: agreement. Oxygen and carbon abundances in the OB stars come from
394: analyses of O II and C II lines, while oxygen in the F dwarf
395: comes from an analysis of O I lines. The K-M dwarfs use molecular
396: OH and CO lines. This abundance comparison across a large range
397: in T$_{\rm eff}$ utilizing a variety of spectroscopic signatures
398: provides evidence that the abundance analysis of the low-mass stars
399: studied here is robust. We proceed to a discussion of the fluorine
400: abundances.
401:
402: Historically,
403: $^{19}$F abundances were derived initially in a sample of
404: galactic K, M, S, and C red-giants, with metallicities mostly
405: near solar, by Jorissen et al. (1992); this study
406: was restricted to fairly bright stars in the IR. More recently,
407: access to high spectral resolution in the IR using the Phoenix
408: spectrograph on the Gemini-South 8.1m telescope has allowed
409: fluorine abundances to
410: be measured in fainter samples of cool stars, such as in the
411: Large Magellanic Cloud (LMC) or in the globular cluster $\omega$
412: Centauri (Cunha et al. 2003). Renda et al. (2004) used the abovementioned
413: $^{19}$F abundances as comparisons to chemical evolution models.
414: They conclude that both Wolf-Rayet and AGB stars are significant
415: sources of $^{19}$F, with the Wolf-Rayet stars dominating at high
416: metallicities. At low metallicities (less than roughly -0.8 dex)
417: synthesis via the $\nu$ process in SN II dominates fluorine
418: production.
419: Renda et al. (2004) were the first to show that all three $^{19}$F
420: sources can be significant contributors, but with differing relative
421: contributions depending on the metallicity and star formation
422: history of the stellar population.
423: In addition to abundances derived from stellar HF lines, Federman
424: et al. (2005) have obtained detections of interstellar F I lines
425: (at 955\AA) along two lines-of-sight into the Cep OB2 association
426: using the Far Ultraviolet
427: Spectroscopic Explorer (FUSE). Their derived column densities of
428: F and O indicate a near-solar F/O ratio, with some small depletion
429: effects.
430:
431: The Orion stars analyzed here are of particular interest
432: as they represent a different type of star with which
433: to probe fluorine when compared to previous studies since
434: Jorissen et al. (1992) sampled only evolved stars.
435: Our work adds three stars of higher surface gravity to
436: the sample of near-solar metallicity field stars and
437: their $^{19}$F and $^{16}$O abundances are shown in
438: Figure 3 in the form of [F/O] versus A(O).
439: The point represented by JW163 should be viewed with caution
440: as we have noted already that its abundance results have probably been
441: influenced by the presence of a circunstellar disk. Although both O and F
442: could have been affected about the same with the result that the ratio
443: of F/O is changed relatively less.
444: Also, we adopt the same oxygen abundance for JW22 as we derived
445: for JW433 because the 15550\AA\ region with the OH lines was
446: not observed for JW22; however its carbon abundance is the
447: same, within the errors, as found for JW433 and there is no
448: evidence that the Orion stars exhibit large variations in their
449: total C/O values.
450: The Jorissen et al. (1992) K and M giants are also
451: plotted (with the abundances in these stars re-derived
452: by Cunha et al. (2003)); the solar values in Figure 3
453: are taken to be A(O)= 8.69 and
454: A(F)= 4.55. Note that the Orion
455: dwarf abundances overlap very nicely with the Jorissen et al.
456: giants.
457:
458: The stellar abundances that are plotted in Figure 3 represent
459: all the available fluorine abundances that have been measured for members of the
460: Milky Way so far (excluding self-polluted AGB stars).
461: Abundances for LMC red giants from
462: Cunha et al. (2003) are not shown here, as the star formation
463: history and corresponding chemical evolution in the LMC may be
464: quite different from the Milky Way. The solid curve in Figure 3
465: is a chemical evolution model from Renda et al. (2004) for the
466: Milky Way. This particular curve represents their MWc model that
467: includes $^{19}$F yields from all three stellar production sites:
468: SN II with the $\nu$-process, Wolf-Rayet stars, and AGB stars.
469: The Orion Nebula cluster F and O abundances help anchor the Milky
470: Way values at high metallicity, which are fit nicely by the Renda
471: et al. (2004) model with all three sources. The low values of
472: [F/O] in the $\omega$ Cen stars were discussed by both Cunha et al.
473: (2003) and Renda et al. (2004) in the context of the unusual star
474: formation history and chemical evolution within this self-enriched
475: stellar system that has been captured by the Galaxy.
476:
477: \section{Conclusions}
478:
479: We detect the HF molecular line at 23400\AA\ in K-M dwarfs and
480: present the first fluorine
481: abundance measurements in young stars still in
482: the pre-main sequence phase of evolution. These targets are members
483: of the Orion
484: association, in particular from the
485: Orion Nebula cluster, with masses around 0.4-0.8 M$_{\odot}$.
486: Our results for JW433 and JW22, which do not show substantial emission
487: from a circumstellar disk, reveal abundances of C and O that are
488: in excellent agreement with those that are found in B-type stellar
489: members of Orion Id (Cunha \& Lambert 1994). This agreement
490: bolsters confidence in the quantitative spectroscopic abundance
491: analyses techniques used in the hotter B stars, as well as the
492: cool K and M dwarfs. Moreover,
493: JW433 and JW22 also display abundance patterns in
494: agreement with the general behavior of fluorine versus oxygen for the
495: Milky Way
496: disk that has been established from stars in a distinctly
497: different evolutionary state.
498: All previous F abundances had been derived from studies of
499: red giants. The overlap in these abundances provides
500: an important confirmation for the
501: relation of the fluorine abundance with oxygen in the Disk.
502:
503: In the future, larger samples of K and M dwarfs,
504: as well as L and T dwarfs, can
505: be used to measure fluorine and oxygen abundances both in the
506: field and in the nearer star-forming regions.
507:
508: Based on observations obtained at the Gemini Observatory,
509: which is operated by the Association
510: of Universities for Research in Astronomy, Inc., under a cooperative
511: agreement with the NSF on behalf of the Gemini partnership: the National
512: Science Foundation (United States), the Particle Physics and Astronomy
513: Research Council (United Kingdom), the National Research Council (Canada),
514: CONICYT (Chile), the Australian Research Council (Australia),
515: CNPq (Brazil), and CONICRT (Argentina), as program GS-2002B-Q-2.
516: This work is also supported in part by the National Science Foundation through
517: AST03-07534 (VVS) and NASA through NAG5-9213 (VVS).
518:
519: \clearpage
520: \begin{bibliography}{}
521: \noindent
522: Allende Prieto, C., Lambert, D. L., \& Asplund, M. 2000, ApJ, 573, L137\\
523: \smallskip
524:
525: Asplund, M. 2005, to appear in Cosmic Abundances as Records of Stellar
526: Evolution and Nucleosynthesis, ASP Conference Series, eds. F. N.
527: Bash \& T. G. Barnes (San Francisco:ASP)\\
528: \smallskip
529:
530: Bessel, M.S., Castelli, F. \& Plez, B. 1998, A\&A, 333, 231\\
531: \smallskip
532:
533: Carpenter, J. M. 2001, ApJ, 121, 2851\\
534: \smallskip
535:
536: Cunha, K., Smith, V.V., Lambert, D.L. \& Hinkle, K.H. 2003, AJ, 126, 1305\\
537: \smallskip
538:
539: D'Antona, F. \& Mazzitelli, I., 1994, ApJS, 90, 467\\
540: \smallskip
541:
542: Domogatskii, G. V., Eramzhian, R. A., \& Nadezhin, D. K. 1978, Ap.\& Space
543: Sci., 58, 273\\
544: \smallskip
545:
546: Federman, S. R., Sheffer, Y., Lambert, D. L., \& Smith, V. V. 2005,
547: ApJ, in press (astro-ph/0410362)\\
548: \smallskip
549:
550: Forestini, M., Goriely, S., Jorissen, A., \& Arnould, M. 1992,
551: A\&A, 261, 157\\
552: \smallskip
553:
554: Hillenbrand, L. A. 1997, AJ, 1733, 1768\\
555: \smallskip
556:
557: Hillenbrand, L. A., \& Hartmann, L. W. 1998, ApJ, 492, 540\\
558: \smallskip
559:
560: Hillenbrand, L. A., Strom, S. E., Calvet, N., Merrill, K.M.,
561: Gatley, I., Makidon, R.B., Meyer, M.R., \& Skrutskie, M.F., 1998, AJ,
562: 116, 1816\\
563: \smallskip
564:
565: Hinkle, K. H. 2004, private communication\\
566: \smallskip
567:
568: Hinkle, K. H., Wallace, L., \& Livingston, W. 1995, Infrared Atlas
569: of the Arcturus Spectrum, 0.9--5.3$\mu$m (San Francisco:ASP)
570: \smallskip
571:
572: Hinkle, K. H., Cuberly, R., Gaughan, N., Heynssens, J.,
573: Joyce, R., Ridgway, S., Schmitt, P., \& Simmons, J. E. 1998,
574: Proc. SPIE, 3354, 810\\
575: \smallskip
576:
577: Jorissen, A., Smith, V.V., \& Lambert, D.L., A\&A, 261, 164\\
578: \smallskip
579:
580: Kurucz, R. L. 1994, CD-ROM 19 (Cambridge:SAO)\\
581: \smallskip
582:
583: Meynet, G., \& Arnould, M. 2000, A\&A, 355, 176\\
584: \smallskip
585:
586: Nissen, P.E., Primas, F., Asplund, M., \& Lambert, D.L. 2002, A\&A 390, 235\\
587: \smallskip
588:
589: Renda, A. et al. 2004 , MNRAS, 354, 575\\
590: \smallskip
591:
592: Rhode, K.L., Herbst, W., \& Mathieu, R.D. 2001, AJ, 122, 3258\\
593: \smallskip
594:
595: Sneden, C. 1973, ApJ, 184, 839\\
596: \smallskip
597:
598: Smith, V.V., Hinkle, K.H., Cunha, K., Plez, B., Lambert, D.L.,
599: Pilachowski, C.A., Barbuy, B., Melendez, J., Balachandran, S.,
600: Bessel, M.S., Geisler, D.P., Hesser, J.E., \& Winge, C., 2002,
601: AJ, 124, 3241\\
602: \smallskip
603:
604: Wallace, L., \& Livingston, W. 1992, An Atlas of a Dark Sunspot Umbral
605: Spectrum from 1970 to 8640 cm$^{-1}$ (1.16 to 5.1$\mu$m), N.S.O.
606: Technical Report \#92-001\\
607: \smallskip
608:
609: Woosley, S.E., Haxton, W.C., 1988, Nature, 334, 45\\
610: \smallskip
611:
612: Woosley, S.E., Hartmann, D.H., Hoffman, R.D., Haxton, W.C., 1990, ApJ, 356,
613: 272\\
614: \smallskip
615:
616: \end{bibliography}
617:
618: \clearpage
619:
620: \begin{deluxetable}{ cccccccccccc }
621: \setcounter{table}{0}
622: \tablewidth{400pt}
623: \tablecaption{Stellar Parameters and Abundances}
624:
625: \tablehead{ Star &
626: \multicolumn{1}{c} {V$_{0}$} &
627: \multicolumn{1}{c} {I$_{0}$} &
628: \multicolumn{1}{c} {K$_{0}$} &
629: \multicolumn{1}{c} {T$_{\rm eff}$} &
630: \multicolumn{1}{c} {Log g} &
631: \multicolumn{1}{c} {$\xi$} &
632: \multicolumn{1}{c} {M/M$_{\odot}$} &
633: \multicolumn{1}{c} {A($^{12}$C)} &
634: \multicolumn{1}{c} {A($^{16}$O)} &
635: \multicolumn{1}{c} {A($^{19}$F)}
636: }
637: \startdata
638: JW 22 & 15.5 & 13.4 & 11.3 & 3650 & 3.6 & 1.0 & 0.4 & 8.40 & -- & 4.61 \\
639: JW 163 & 13.1 & 11.8 & 10.2 & 4250 & 3.5 & 0.5 & 0.6 & 8.00 & 8.55 & 4.53 \\
640: JW 433 & 12.5 & 11.3 & 9.6 & 4400 & 3.4 & 0.8 & 0.7 & 8.30 & 8.80 & 4.65 \\
641: \enddata
642: \tablecomments{A(X)= log[n(X)/n(H)] + 12.
643: }
644: \end{deluxetable}
645:
646: \clearpage
647:
648: \begin{deluxetable}{ cccccc }
649: \setcounter{table}{1}
650: \tablewidth{400pt}
651: \tablecaption{Abundance Uncertainties}
652:
653: \tablehead{ Star &
654: \multicolumn{1}{c} { Element} &
655: \multicolumn{1}{c} { $\Delta$T$_{\rm eff}$(+100K)} &
656: \multicolumn{1}{c} {$\Delta$Log g(+0.3 dex)} &
657: \multicolumn{1}{c} {$\Delta$$\xi$(+0.5 km/s)} &
658: \multicolumn{1}{c} {$\Delta$A}
659: }
660: \startdata
661: JW 22 & $\Delta$C & 0.08 & 0.0 & 0.10 & 0.13 \\
662: JW 22 & $\Delta$F & 0.08 & 0.05 & 0.0 & 0.09 \\
663: JW 433 & $\Delta$C & 0.15 & 0.15 & 0.0 & 0.21 \\
664: JW 433 & $\Delta$O & 0.10 & 0.02 & 0.0 & 0.10 \\
665: JW 433 & $\Delta$F & 0.13 & 0.05 & 0.0 & 0.14 \\
666: \enddata
667: \end{deluxetable}
668:
669: \clearpage
670:
671: \figcaption[fig1.ps]{Observed Phoenix spectra for the target stars
672: JW163 and JW433 centered at 15545 \AA (top panel) and 23390 \AA.
673: The spectra have been separated vertically, while prominent absorption
674: lines are indicated.
675: \label{fig1}}
676:
677:
678: \figcaption[fig2.ps]{Observed and synthetic spectra for the three target
679: stars in the HF region. The best-fit fluorine abundances (represented
680: by solid lines) are those presented in Table 1. The synthetic
681: spectra represented by the dashed lines were calculated with
682: $\pm$ 0.2dex changes in the $^{19}$F abundance. Note the blending
683: feature just blueward of the HF(1-0) R9 line, which is identified
684: tentatively here as H$_{2}$O.
685: \label{fig2}}
686:
687:
688: \figcaption[fig3.ps]{Values of [F/O] are plotted versus oxygen
689: abundances for field K and M giants from Cunha et al. (2003) and
690: the three Orion Nebular cluster stars. The solar symbol is shown
691: and the solid line illustrates the predictions from one chemical
692: evolution model by Renda et al. (2004) that includes all three
693: suggested sources for $^{19}$F production. Note that the Orion
694: Nebula stars fall within the scatter defined by the red giants
695: of the Milky Way field.
696: \label{fig3}}
697:
698: \clearpage
699:
700: \begin{figure}
701: \plotone{f1.eps}
702: \end{figure}
703:
704: \clearpage
705:
706: \begin{figure}
707: \plotone{f2.eps}
708: \end{figure}
709:
710: \clearpage
711:
712: \begin{figure}
713: \plotone{f3.eps}
714: \end{figure}
715: \end{document}
716: