Beryllium, Oxygen and Iron Abundances in Extremely Metal-Deficient Stars
aa r X i v : . [ a s t r o - ph . GA ] J un Beryllium, Oxygen and Iron Abundances in ExtremelyMetal-Deficient Stars
Jeffrey A. Rich & Ann Merchant Boesgaard Institute for Astronomy, University of Hawai‘i at Manoa,2680 Woodlawn Drive, Honolulu, HI 96822 [email protected]@ifa.hawaii.edu
ABSTRACT
The abundance of beryllium in the oldest, most metal-poor stars acts as aprobe of early star formation and Galactic chemical evolution. We have analyzedhigh-resolution, high signal-to-noise Keck/HIRES spectra of 24 stars with [Fe/H]from − − − − ± − σ level. A single relationship between Beand [O/H] has a slope of 1.21 ± − − − Visiting Astronomer, W. M. Keck Observatory jointly operated by the California Institute of Technologyand the University of California.
Subject headings: stars: abundances; stars: evolution; stars: late-type; starsPopulation II; Galaxy: halo; Galaxy: disk
1. Introduction
The abundances of the rare light elements, Li, Be, and B, have revealed an amazingamount of information about an array of astrophysical issues. These include cosmology andnucleosynthesis in the Big Bang, Galactic chemical evolution, cosmic ray theory, and stellarinteriors and stellar evolution. It is important to add more observations of Be to the largenumber of Li observations in order to make new progress in these research fields. Berylliumis the least complicated of the three light elements because it has only one known source ofproduction and only one stable isotope, Be. It is also less susceptible to depletion than Liin stellar interiors.Observations of Primas et al. (2000a, 2000b) of Be in very metal poor stars led them tosuggest the possibility of a Be plateau at very low [Fe/H], possibly similar to the Spite lithiumplateau (Spite & Spite 1982). The Li plateau in older stars is attributed to nucleosynthesisof Li during the Big Bang. If such a plateau exists, could it be similar to the Li plateau?Whereas the Li plateau is produced by nuclear reactions during the Big Bang, very little Beis produced in standard Big Bang nucleosynthesis (BBN): N(Be)/N(H) = 10 − (Thomaset al. 1994). Models that include inhomogeneities in the early universe could produce adifferent mixture of light elements from the standard models (e.g. Malaney & Mathews 1992).Inhomogeneous models can create far more Be than standard models and the predictions ofOrito et al. (1997) are 10 − to 10 − ; the implied plateau corresponds to 5 x 10 − , whichis tantalizingly close to the observed values.Another possible cause of a Be “plateau” could be the creation of Be through spallationreactions in contained superbubbles created by multiple supernovae (Parizot 2000) or nearhypernovae (Fields et al. 2002, Nakamura et al. 2006). Enrichment of the ISM in Be by suchprocesses early in the history of the galaxy could result in a detectable plateau.The lowest metal stars ([Fe/H] < − ∼
150 MeV) protons and neutronsbombard interstellar nuclei of C, N and O creating lighter isotopes. It has also been suggestedthat the “bullets” and the “targets” might be reversed near supernovae where C, N, and O 3 –nuclei could be accelerated into the local interstellar gas including protons and neutrons (see,for example, Duncan et al. 1997, 1998, Lemoine, Vangioni-Flam & Cass´e 1998).Recent studies of Be in low metal stars include Boesgaard, Levesque & Rich (2008), Tan,Shi & Zhao (2009) and Smiljanic et al. (2009). All of these studies examine the Galacticevolution of Be through the relationships between the abundances of Be with Fe and Bewith O and/or alpha-elements. The large study of Smiljanic et al. (2009) includes 90 starsand the determination of Be abundances of 73 halo and thick disk stars. They relied onHipparcos parallaxes to find log g , and adopted published values for temperature, [Fe/H],[ α /Fe], and microturbulent velocity. Their stars range in [Fe/H] from − − − − − − −
2. Observations and Data Reduction
The Be II resonance lines, located in the UV spectral region at 3130.421 and 3131.065˚A, were used to determine the Be abundances. High S/N and high-resolution spectra arerequired to observe the Be doublet, which is weak in low metallicity stars. The upgradedHigh Resolution Echelle Spectrograph (HIRES) instrument on the Keck I telescope achievesthe necessary requirements (Vogt et al. 1994). The spectral resolution for our setup was48,000, and the light collecting area of the 10 m Keck telescope enables us to obtain S/Nnear 100 per pixel for our metal-poor sample. The pixel size on the upgraded CCD is 15 µ .Such stars are uncommon and thus most are faint, especially in the UV wavelength region ofthe Be II lines. The observations reported here were carried out over 11 nights using the bluecross-disperser. The spectral range is approximately 3000 to 6000 ˚A. The new HIRES CCDis composed of three chips, one each for UV/blue (on which the Be doublet is located), greenand red wavelengths. The quantum efficiency near 3100 ˚A is about 93%, making exposuresof stars down to V = 12 possible.Stars were selected based on their metallicity and observability. Stars with [Fe/H] valuesof less than − − < δ < +60; the latitude ofMauna Kea is +19 deg ∼
3. Data Analysis
The data were analyzed using IRAF and MOOG, a stellar synthesis and analysis pro-gram (Sneden 1973). We used the 2002 version, which includes the UV opacity edges of themajor atomic species. Equivalent widths of spectral lines were measured using IRAF rou-tines. MOOG was used for two purposes: stellar parameter determination and abundancedeterminations. MOOG’s “abfind” driver was used with equivalent widths of Fe I, Fe II,Ti I, and Ti II lines to measure the metallicity, [Fe/H], and determine T eff and the surfacegravity, log g , of the stars in our sample. We used the “synth” driver to create syntheticspectra in order to measure Be and O abundances. http://verdi.as.utexas.edu/moog.html Stellar parameters are needed to generate the stellar models used for spectral synthesisand abundance determination. We have chosen to determine our stellar parameters spec-troscopically. We are dealing with spectral lines and need to know the temperature in theregion of the atmosphere where the lines are formed. These temperatures seem more relevantfor abundance work than the continuum-based color temperatures. It is generally the casethat spectroscopic temperatures are lower than color temperatures because the continuumis formed deeper in the stellar atmosphere where the temperatures are higher.Several Fe I and Fe II and Ti I and Ti II lines fall within the range of the HIRES CCDand can be used to determine [Fe/H], T eff and log g , using an iterative method similar to thatof Stephens (1999). Microturbulence, ξ , has a negligible effect on any of the measurementsin this project, so a standard value of 1.5 km/s was used in all models (Magain 1984). Wemeasured equivalent widths of 30-60 Fe I and 10 Fe II lines on the green and red chips of theHIRES CCD as well as 3 Fe I lines from the UV chip near the Be doublet. The Fe I linescovered a range in excitation potential from 0.86 to 3.98 eV and so could be used to derivethe temperature. In addition we measured equivalent widths of 10-15 Ti I lines and 8-12 TiII lines. A list of the lines measured, along with the excitation potential and the gf values,are given in Table 2. The values for gf come from the compilation presented in Stephens(1999) and Stephens & Boesgaard (2002) which gives the appropriate source references.We have been careful to use only weak lines which would be on the linear portion of thecurve of growth. This corresponds to log (W/ λ ) < − ∼
75 m˚A at 5000 ˚A. Forsome stars we applied a even stricter limit of log (W/ λ ) < − ∼
35 m˚A at 5000˚A. The measured equivalent widths are given in the tables in the Appendix. For two stars,G 64-12 and G 64-37, we adopted the parameters of Stephens & Boesgaard (2002) whichwere also determined spectroscopically; they used the same method as we did here, but theyhad a larger wavelength coverage toward the red and thus more useable lines.Using an initial estimate for the stellar parameters and the measured equivalent widths,MOOG’s “abfind” driver was then used to calculate an Fe abundance for each line, as wellas an average [Fe/H] from the equivalent widths. MOOG also calculates the slope of eachline’s calculated abundance versus its excitation potential (EP). If the temperature is correct,there should be no trend of abundance with EP.Atomic abundances derived from two ions of the same species should give similar results.If the abundances do not agree, the value of log g can be adjusted and the abundancesrecalculated with MOOG. For the purposes of this project, MOOG was run iteratively byfirst calculating the abundance from Ti I lines using the temperature found via the Fe I lines, 6 –then finding the abundance from Ti II and comparing the two sets of abundances. The valuefor log g is adjusted and the process can then be run iteratively between Fe I and Ti I plusTi II abundances until [Fe/H], T eff and log g are calculated. (We have used the ionizationbalance between Ti I and Ti II, rather than Fe I and Fe II out of concern about NLTE effectson Fe I lines in metal-poor stars, e.g. Th´evenin & Idiart 1999.) The stellar parameters used tocalculate abundances are shown in the second, third and fourth columns of Table 3. Typicalerrors are T eff ±
80 K, log g ± ± ±
80 K, we point out that there maybe large (100-250 K) systematic differences in temperature determinations by other methodssuch as UBV and other photometric indices (and the potential uncertainties in the reddeningcorrections), the infrared flux method, Balmer-line profiles.Although we have relied on the Magain (1984) value of 1.5 km s − for the microturbu-lence, we have checked the effect of using 1.0 km s − on the stellar parameters. On averagewith ξ = 1.0 km s − , the value of T eff is lower by 66 K, log g is lower by 0.15 and [Fe/H] islower by 0.03 dex. All of these are within the range of the uncertainities in our parameterdeterminations.Ten of our stars have been observed for Li by Hosford et al. (2009). They use theexcitation balance of Fe I to constrain their temperature scale and they constrain log g usingtheoretical isochrones. They present two sets of parameters for most of the stars appropriatefor a main-sequence star and a subgiant-branch star. Our temperatures for the 10 stars incommon are typically lower than theirs by −
163 K on average. Our log g values are usuallymore in alignment with their subgiant branch gravities and on average are lower by − g , the Be abundances are quite insensitive to temperature, butare sensitive to log g (see § The difficulty in determining A(Be) in metal poor stars can be seen in Figure 1, whichshows a comparison of a very metal poor star, BD +3 740 at [Fe/H] = − − ± gf of − − T eff and log g to see how it affected the abundancecalculation. The largest effect on the Be abundances of any of the stellar parameters is dueto uncertainties in log g . For instance, a change in T eff in ±
80 K results in a change in A(Be)of ± ± ± g of ± ± ± T eff has the largest effect on the determined abundance, followed by log g . A change in T eff of ±
80 K changes [O/H] by ± g of +0.2 only results in achange of [O/H] by +0.05 to +0.07 dex. These uncertainties were added in quadrature andappear in Table 3.Although the “missing” UV opacity has been blamed for inaccuracies in the Be abun-dances (Balachandran & Bell 1998), in these stars of such low metallicity, there is no problemof continuum placement and no issue of missing opacity. Similarly, the O abundances fromthe UV lines of OH are basically unaffected by opacity or continuum placement issues. Wepoint out again that the 2002 version of MOOG includes the UV opacity edges of the dom-inant elements. In considering this issue Smiljanic et al. (2009) calculated the effect of anincrease of 0.20 in [Fe/H] (a factor of 1.6 as suggested by Balachandran & Bell 1998) on Befor a star with [Fe/H] = − − − We have reanalyzed the Keck/HIRES observations in Boesgaard et al. (1999a, 1999b) touse the newer version of MOOG from 2002. These observations were made with the originalHIRES and due to the limited wavelength coverage of that CCD, we could not determinethe stellar parameters spectroscopically. Instead we used the the parameters from the 1999papers with temperatures on the Carney (1983) scale. That temperature scale is consistentwith the spectroscopic temperatures derived for the stars in this paper. For three stars wehave lowered the temperature and achieve better fits. The new Be abundances were foundby the same method used here - spectrum synthesis - and with the same line list. We haverederived the O abundances but use only three OH features (3130, 3139, and 3140 ˚A), aswas done here for the newly observed stars. The new abundance results and the 1 σ errorestimates are presented here in Table 4 along with the stellar parameters and the 1 σ errorson those.The Be abundances do not differ much from those presented in Boesgaard et al. (1999a):on average they differ by − ± − − − ± We have incorporated in this study the results on Be in metal-poor stars published byBoesgaard & Novicki (2006) (four stars), and the three Li-Be normal stars from Boesgaard(2007). These stars, their parameters and abundances are given in Table 4. We have deter-mined O abundances in these stars in the same manner as above, from 3 OH features. Thoseresults are presented in Table 5.
There are two stars in common in this study with Boesgaard et al. (1999a): BD +3 ◦
740 and BD − ◦ ◦
740 the parameters determined spectroscopicallyand photometrically agree well. The temperatures differ by −
80 K, log g by 0.21, [Fe/H] by − − − − ◦ g from Boesgaard et al. (1999a) is veryuncertain ( ± − − − − − g values derived from parallax and ours are fromionization balance, with the exception of one star (HD 219617) the agreement in log g is goodwith a mean difference of +0.01 ± − ± − ± − ±
4. Results and Discussion
In this paper we discuss the abundance relationships between Be and Fe, between Beand O, and between Fe and O. We have acquired additional observations of Be in 34 morestars with [Fe/H] between − − The Be abundances from Tables 3, 4, and 5 are shown in Figure 4 plotted against thevalues of [Fe/H]. The abundances show a trend of increasing A(Be) with increasing [Fe/H].The straight line between A(Be) and [Fe/H] isA(Be) = 0.92 ( ± ± ± ± ± ± ± ± < − ± > − > − − ± ± − − < − − ± − ± ± − − g forHD 94028 and thus a lower Be abundance, putting it in alignment with the other stars.) The nucleosynthesis of Be is directly related to O as the spallation of abundant atomslike C, N, and O produce smaller atoms like Li, Be, and B. The spallation can occur 1) ashigh energy cosmic rays bombard CNO atoms in the ambient interstellar gas or 2) duringsupernovae explosions where atoms of CNO and protons are accelerated to high energies inthe vicinity. These two mechanisms predict different relationships between Be and O.In the more traditional GCR spallation the slope between A(Be) and [O/H] would be2 because the number of O atoms is proportional to the cumulative number of SN IIa ( N )while the energetic protons are proportional to the instantaneous number of SN II ( dN ).The integral of N dN is kN .In the immediate vicinity of SN II the number of Be atoms would be proportional to 13 –the number of O atoms; this would result in a slope of 1 between A(Be) and [O/H].In Figure 7 we show the least-squares fit between A(Be) and [O/H]. The equation forthe straight line isA(Be) = 1.21 ( ± ± < − ± > − ± − ± ± ± ± Our data enable us to reenter the controversy regarding the form of the relation betweenO and Fe. Our sample contains 49 stars which are all dwarfs and subgiants - no giants; therange in log g is 3.20 to 4.52. The temperatures range from turn-off stars at 6400 K todwarf stars at 5500 K. Our star sample covers 3 orders of magnitude in [Fe/H] from − − − ≤ − g ranging from 0.7 to 2.7; they find a mean [O/Fe] of 0.7 with a dispersion of 0.17 dex and0.47 after an uncertain correction for stellar surface inhomogeneities.In Figure 9 we show the tight, linear relationship between [O/H] and [Fe/H] with awell-fitting slope of 0.70 ± ± − ± ∼ − − ± − ± The linearity of the relation between [O/H] and [Fe/H] in Figure 9 is striking. And itimplies that the 2 slope fit between A(Be) and [O/H] (Figure 8) probably requires the 2slope fit between A(Be) and [Fe/H] (Figure 5). In turn this indicates that the upturn in 15 –[Be/Fe] in Figure 6 at the lowest metallicities is significant. It is not evidence for a “plateau”in the Be abundance, but rather a change in the chemical history of the formation of Be. Asdiscussed in §
5. Summary and Conclusion
We have used the upgraded HIRES on the Keck I 10m telescope to obtain high reso-lution, high signal-to-noise spectra of 24 very metal deficient stars from [Fe/H] = − − < − − − − − T eff , and Ti Iand Ti II for log g. An iterative procedure was followed for the parameter determinationand the Fe abundance. The stellar synthesis program MOOG was used in both abfind and synth modes to find the elemental abundances of Be, O, and Fe.The full sample of 49 stars reveals a tight correlation between [O/H] and [Fe/H] with aslope of 0.69 ± − − − − − ≤− < − − ± ± ± ± − − − − − − − − ± − ± REFERENCES
Alibes, A., LaBay, J. and Ramon, C. 2002, ApJ, 571, 326Asplund, M. & Garc´ ia Per´ez, 2001, A&A, 372, 601Balachandran, S.C. & Bell, R.A. 1998, Nature, 392, 791Boesgaard, A. M. 2000, IAU symp. 198, 389Boesgaard, A. M. 2001, NewAr, 45, 525Boesgaard, A. M. 2004, in the Origin and Evolution of the Elements ed. A. McWilliam &M. Rauch, (Cambridge: Cambridge University Press), p. 117Boesgaard, A. M., 2007, ApJ, 667, 1196Boesgaard, A. M., Deliyannis, C. P., King, J. R., Ryan, S. G., Vogt, S. S., Beers, T. C.1999a, AJ, 117, 1549Boesgaard, A. M., King, J. R., Deliyannis, C. P., Vogt, S. S., 1999b, AJ, 117, 492Boesgaard, A.M., Levesque, E.M. & Rich, J.A. 2008, First Stars III, ed. B.W. O’Shea, AHeger & T. Abel [AIP: Melville NY]., p. 164Boesgaard, A. M. & Novicki, M. C., 2006, ApJ, 641, 1122Boyd, R. N. and Kajino, T., 1989, ApJ, 336, L55Carney, B. 1983, AJ, 83, 623Cayrel R., Depagne, E., Spite, M., Hill, V., Spite, F., Francois, P. et al. 2004, A&A, 416,1117Duncan, D.K., Primas, F., Rebull, L.M., Boesgaard, A.M., Deliyannis, C.P., Hobbs, L.M.,King, J.R. & Ryan, S.G. 1997, ApJ, 488, 338Fields, B. D., and Daigne, F., Casse, M., and Vangioni-Flam, E., 2002, ApJ, 581, 389Fulbright, J. P. & Johnson, J. A. 2003, ApJ, 595, 1154Garc’/i a P´erez, A.E., Asplund, M., Primas, F., Nissen, P.E., & Gustafsson, B. 2006, A&A,451, 621Hosford, A., Ryan, S.G., Garc´ıa P´erez, A.E., Norris, J.E. & Olive, K.A. 2009, A&A, 493,601Israelian, G., Garc´ıa L´opez, R. & Rebolo, R. 1998, ApJ, 507, 805Israelian, G., Rebolo, R., Garc´ıa L´opez, R., Bonifacio, P., Molaro, P. Basri, G. & Shchukina,N. 2001, ApJ, 551, 833Jedamzik, K. and Rehm, J. B., 2001, PhRvD, 64, 3510 19 –Magain, P. 1984, A&A, 132, 208Malaney, R. .A. and Fowler, W. A., 1989, ApJ, 345, L5Meneguzzi, M., Audouze, J., and Reeves, H. 1971, A&A, 15, 337Nakamura, K., Inoue, S., Wanajo, S., and Shigeyama, T., 2006, ApJ, 643, 115Orito, M., Kajino, T., Boyd, R. N., and Mathews, G. J., 1997, ApJ, 488, 515Parizot, E., 2000, A&A 362, 786Pasquini, L., Galli, D., Gratton, R. G., Bonifacio, P., Randich, S., Valle, G. 2005, A&A, 436,L57Pasquini, L., Bonifacio, P., Randich, S., Galli, D., Gratton, R. G., 2004, A&A, 426, 651Primas, F., Asplund, M., Nissen, P. E., and Hill, V., 2000a, A&A, 364, L42Primas, F., Molaro, P., Bonifacio, P. and Hill, V., 2000b, A&A, 362, 666Reeves, H., Fowler, W.A. and Hoyle, F. 1970, Nature, 226, 727Smiljanic, R., Pasquini, L., Bonifacio, P., Galli, D., Gratton, R.G., Randich, S. & Wolff, B.2009, A&A, 499, 103Sneden, C. A. 1973, Ph.D. thesis, Univ. Texas, AustinSpergel, D. N., Bean, R., Dor´e, R., Nolta, M. R., Bennett, C. L. et al. 2007, ApJS, 170, 377Spite, F., and Spite, M. 1982, A&AS, 115, 357Stephens, A., 1999, A&A, 117, 1771Stephens, A. & Boesgaard, A. M. 2002, AJ, 123, 1647Suzuki, T.K. & Yoshii, Y. 2001, ApJ, 549, 303Tan, K.F., Shi, J.R., & Zhao, G. 2009, MNRAS, 392, 205Th´evenin, F. & Idiart, T.P. 1999, ApJ, 521, 753Thomas, D., Schramm, D., Olive, K., Mathews, G., Meyer, B. S. & Fields, B.D. 1994, ApJ,430, 291Vogt, S. S. et al., SPIE, 2198, 362
This preprint was prepared with the AAS L A TEX macros v5.2.
20 –Table 1. Log of the Beryllium ObservationsStar Code a R.A. Dec. V Date(s) U.T. Exp. min. S/NLP 651-4 1 02 44 −
05 26 12.04 2006 Jan 2 65 83G 75-56 2 03 00 −
05 57 11.94 2004 Sep 72004 Nov 7 232 107LP 831-70 3 03 06 −
22 19 11.40 2008 Jan 16 90 52BD +3 740 4 05 01 +04 06 9.80 2004 Nov 7 60 166G 108-58 5 07 10 −
01 17 11.82 2008 Jan 16 90 91G 88-10 6 07 10 +24 20 11.86 2007 Nov 192008 Jan 16 150 121BD +20 2030 7 08 16 +19 41 11.20 2006 Jan 2 60 122BD +9 2190 8 09 29 +08 38 11.16 2005 Jan 31 120 110BD +1 2341p 9 09 40 +01 00 10.48 2006 Jan 2 40 139BD +44 1910 10 09 49 +44 17 10.95 2008 Jan 16 60 112BD −
13 3442 11 11 46 −
14 06 10.37 2006 Jan 2 80 194LP 553-62 12 11 54 +02 57 11.68 2008 Jan 16 90 97G 11-44 13 12 10 +00 23 11.08 2008 Jan 16 30 76G 59-24 14 12 34 +15 16 12.02 2008 Jan 16 56 73G 64-12 15 13 40 −
00 02 11.49 2005 May 15 210 109G 64-37 b
14 02 −
05 39 11.14 2003 May 27 270 87G 201-5 16 14 36 +55 33 11.52 2007 Jun 10 120 115BD +26 2621 17 14 54 +25 33 11.00 2007 Jun 10 60 123G 206-34 18 18 35 +28 41 11.40 2005 May 15 105 97LP 752-17 19 19 25 −
11 56 11.88 2005 Jul 62005 Sep 272006 Jun 19 240 94G 92-6 20 19 29 +01 01 11.75 2007 Jun 10 120 100LP 635-14 21 20 26 −
00 37 11.33 2005 Jul 62007 Jun 10 162 114LP 815-43 22 20 38 −
20 25 10.91 2004 Sep 72006 Jun 19 120 98G 275-4 23 23 07 −
23 52 12.18 2005 Jul 52005 Jul 62005 Sep 27 220 61 a Code is an ID number for the star in the Appendix Table of equivalent widths. b G64-37 observed on Subaru with HDS. See Boesgaard & Novicki (2006). 21 –Table 2. Lines MeasuredElement λ (˚A) Ex. Pot.(eV) gfFe I λ (˚A) Ex. Pot.(eV) gf5098.697 2.1760 0.00941905123.720 1.0110 0.00086505150.840 0.9900 0.00095505171.596 1.4850 0.01749855192.344 2.9980 0.37931505194.942 1.5580 0.00880005198.711 2.2230 0.00770905202.336 2.1760 0.01398005215.182 3.2660 0.13458605216.274 1.6080 0.00765605217.390 3.2110 0.07656005227.190 1.5570 0.05929255232.940 2.9400 0.80167805242.491 3.6350 0.12488205250.646 2.1980 0.00767405263.305 3.2660 0.11898705269.537 0.8590 0.04758835328.039 0.9150 0.03423735332.900 1.5570 0.00138705339.930 3.2660 0.20893005341.024 1.6080 0.00985105393.167 3.2410 0.15399305397.128 0.9150 0.01029205405.775 0.9900 0.01406055429.696 0.9580 0.01318255434.524 1.0110 0.00755105455.609 1.0110 0.00804455497.516 1.0110 0.00145505501.465 0.9580 0.00100505569.618 3.4170 0.30690205572.841 3.3970 0.50991805576.090 3.4300 0.10000005586.756 3.3680 0.7177940 23 –Table 2—ContinuedElement λ (˚A) Ex. Pot.(eV) gf5658.816 3.3970 0.1458810Fe II I II λ (˚A) Ex. Pot.(eV) gf4589.958 1.2369 0.01621814657.203 1.2430 0.00582104779.985 2.0478 0.04265805129.152 1.8918 0.04073805154.070 1.5659 0.01202265188.680 1.5819 0.06165955226.543 1.5659 0.05011875336.781 1.5819 0.02162725381.018 1.5658 0.0094406 25 –Table 3. Stellar Parameters and AbundancesStar T eff (K) log g [Fe/H] σ ([Fe/H]) a [O/H] σ ([O/H]) A(Be) σ (Be)LP 651-4 6030 4.26 − − − − − − − − < − − − − − − − − − − − − − − − − − − − − − − −
13 3442 6090 4.11 − − − − − − − − − − − − − − − − − − − − − − − − − − − − − − − − − − − − − − − − − − a These are the standard deviations based on the agreement of the different Fe I lines used. 26 –Table 4. New Beryllium and Oxygen Abundances from Boesgaard et al. (1999a)Star T eff (K) σ log g σ [Fe/H] σ [O/H] σ A(Be) σ HD 19445 5853 40 4.41 0.23 − − − − − − − − − − − − − − − − − − − − − − − − − − − − − − ◦ − − − ◦ − − − ◦ − − ◦ − − − ◦ − − − ◦
740 6110 82 3.64 0.33 − − − ◦ − − − − ◦ − − − − ◦ − − − T eff (K) σ log g σ [Fe/H] σ [O/H] σ A(Be) σ HD 24289 a − − − a − − − b − − b − − a − − b − − b − − a Keck data in Boesgaard (2007) b Subaru data in Boesgaard & Novicki (2006) 28 –Fig. 1.— The Be region of BD +3 740 ([Fe/H] = − − ± ± ± > − < − − ∼ −
6. Appendix
39 –Table 6a. Measured Equivalent Widths λ (˚A) Ele. 1 2 3 4 5 6 7 8 9 10 11 123100.304 Fe I ... ... ... 41.9 ... ... ... ... 41.5 ... 44.8 ...3100.665 Fe I ... ... ... 46.4 ... ... ... ... 43.3 ... ... ...3116.631 Fe I 20.0 ... 22.2 16.3 ... 34.7 ... 17.6 15.8 ... 20.4 20.14494.563 Fe I 12.7 31.8 8.3 11.7 ... 24.1 15.8 11.2 11.3 20.6 11.0 15.54528.614 Fe I 24.6 ... 17.4 20.1 ... 36.9 29.0 18.1 23.0 ... 21.0 26.14531.148 Fe I 6.4 16.2 6.1 6.1 19.7 11.3 9.6 7.4 6.8 16.1 5.9 7.04556.126 Fe I ... ... ... ... 6.5 ... ... ... ... 5.8 ... ...4592.651 Fe I ... ... ... ... 12.3 ... ... ... ... 8.2 ... ...4602.941 Fe I 4.6 16.4 ... 4.2 20.3 12.7 7.9 6.4 6.3 12.7 5.3 8.64647.434 Fe I ... ... ... ... 6.0 ... ... ... ... 7.6 ... ...4678.846 Fe I ... ... ... ... 7.0 ... ... ... ... ... ... ...4691.411 Fe I ... ... ... ... 5.7 ... ... ... ... ... ... ...4733.591 Fe I ... ... ... ... 5.3 ... ... ... ... ... ... ...4736.773 Fe I 5.3 ... ... ... ... ... ... ... ... ... ... ...4871.318 Fe I 16.7 14.3 ... ... 14.3 10.9 7.9 ... 4.9 9.9 4.9 6.54872.137 Fe I 12.9 28.2 8.6 7.6 27.9 19.0 12.8 8.5 9.2 17.4 9.9 12.24890.750 Fe I 16.3 34.3 10.9 14.1 ... 25.9 20.4 11.5 17.3 24.1 14.0 17.64891.490 Fe I 25.9 ... 18.5 21.4 ... 38.3 29.0 20.7 26.3 ... 23.2 26.84918.994 Fe I 17.6 ... 13.4 13.9 ... 31.2 19.2 12.0 15.4 26.7 14.5 20.74920.503 Fe I 32.1 ... 27.2 31.4 ... 46.7 ... 31.9 31.3 ... 29.0 ...4985.253 Fe I ... ... ... ... 5.7 ... ... ... ... ... ... ...4985.547 Fe I ... ... ... ... 6.9 ... ... ... ... ... ... ...4994.130 Fe I ... ... ... ... 15.0 ... ... ... ... ... ... ...5001.862 Fe I ... ... ... ... 14.1 ... ... ... ... ... ... ...5006.119 Fe I 9.7 25.8 7.9 10.6 28.9 20.5 13.2 11.4 9.6 18.2 6.7 13.05012.068 Fe I 10.2 23.0 7.4 9.7 30.9 16.7 10.9 7.5 7.4 19.2 6.9 10.15022.236 Fe I ... ... ... ... 6.3 ... ... ... ... ... ... ...5049.819 Fe I ... ... ... ... 22.4 ... ... ... ... ... ... ...5051.635 Fe I ... ... ... ... 24.2 ... ... ... ... ... ... ...5068.766 Fe I ... ... ... ... 10.5 ... ... ... ... ... ... ...5079.224 Fe I ... ... ... ... 8.2 ... ... ... ... ... ... ...5079.740 Fe I ... ... ... ... 8.2 ... ... ... ... ... ... ... 40 –Table 6a—Continued λ (˚A) Ele. 1 2 3 4 5 6 7 8 9 10 11 125083.339 Fe I ... ... ... ... 15.3 ... ... ... ... ... ... ...5098.697 Fe I ... ... ... ... 11.4 ... ... ... ... ... ... ...5123.720 Fe I 3.2 7.9 ... ... 11.9 7.1 4.4 ... ... ... ... ...5150.840 Fe I ... ... ... ... 11.4 ... ... ... ... ... ... ...5171.596 Fe I 13.9 33.1 8.9 17.7 ... 34.0 16.6 12.6 10.8 24.2 13.0 17.45192.344 Fe I 10.1 27.5 8.9 11.4 30.1 20.6 12.7 8.6 11.6 19.1 8.9 13.75194.942 Fe I ... ... ... ... ... ... ... ... ... ... ... ...5198.711 Fe I ... ... ... ... 8.0 ... ... ... ... ... ... ...5202.336 Fe I ... ... ... ... 13.1 ... ... ... ... ... ... ...5215.182 Fe I ... ... ... ... 7.4 ... ... ... ... ... ... ...5216.274 Fe I ... ... ... ... 20.1 ... ... ... ... ... ... ...5217.390 Fe I ... ... ... ... 6.2 ... ... ... ... ... ... ...5227.190 Fe I 32.5 ... 22.8 30.7 ... 50.2 ... 28.0 25.6 ... 30.0 36.65232.940 Fe I 23.0 ... 15.2 20.9 ... 37.1 27.9 19.2 18.6 35.4 19.4 25.85242.491 Fe I ... ... ... ... 5.2 ... ... ... ... ... ... ...5250.646 Fe I ... ... ... ... 6.8 ... ... ... ... ... ... ...5263.305 Fe I ... ... ... ... 8.5 ... ... ... ... ... ... ...5269.537 Fe I ... ... 51.2 57.6 ... 77.2 ... 53.7 54.2 ... 59.0 ...5328.039 Fe I ... ... 43.1 49.0 ... 68.5 ... 45.0 44.7 ... 50.4 ...5332.900 Fe I ... ... ... ... 5.1 ... ... ... ... ... ... ...5339.930 Fe I ... ... ... ... 14.9 ... ... ... ... ... ... ...5341.024 Fe I ... 27.7 11.9 9.2 28.0 18.7 ... 5.9 10.2 19.7 10.9 11.65393.167 Fe I ... ... ... ... 11.3 ... ... ... ... ... ... ...5397.128 Fe I 25.4 ... 19.1 23.5 ... 41.5 32.7 22.7 20.5 22.4 29.45405.775 Fe I 28.8 ... 21.7 31.2 ... 45.7 35.2 24.5 23.1 ... 26.1 31.65429.696 Fe I 29.8 ... 22.3 26.8 ... 46.9 ... 25.3 26.9 ... 26.1 32.85434.524 Fe I 17.4 ... 12.4 22. 36.5 30.8 24.2 12.3 12.5 30.8 15.7 20.05455.609 Fe I 19.1 ... 15.2 18.6 ... 34.6 28.6 17.1 16.1 35.2 16.3 25.45497.516 Fe I ... ... ... ... ... ... ... ... ... ... ... ...5501.465 Fe I ... ... ... ... ... ... ... ... ... ... ... ...5569.618 Fe I ... ... ... ... 12.0 ... ... ... ... ... ... ...5572.841 Fe I 5.8 20.6 5.1 6.2 20.7 15.3 11.8 ... 7.0 13.8 6.6 10.75576.090 Fe I ... ... ... ... 8.3 ... ... ... ... ... ... ... 41 –Table 6a—Continued λ (˚A) Ele. 1 2 3 4 5 6 7 8 9 10 11 125586.756 Fe I ... 29.9 8.6 12.2 30.9 21.5 17.6 10.8 9.8 17.9 12.0 14.75658.816 Fe I ... ... ... ... 8.3 ... ... ... ... ... ... ...4508.289 Fe II 6.1 17.2 5.3 7.7 9.6 10.8 9.5 6.3 8.6 12.1 6.5 10.54515.339 Fe II 5.5 14.8 5.7 5.5 8.2 10.4 8.4 4.9 5.4 9.5 5.9 9.64522.634 Fe II 10.3 24.6 7.1 11.2 13.9 14.9 13.3 13.4 10.1 16.7 9.5 15.14576.339 Fe II ... 6.5 ... ... ... ... ... ... ... ... ... 5.04583.837 Fe II 17.6 ... 9.6 21.4 25.1 27.8 23.4 21.0 18.5 28.4 20.7 26.84629.339 Fe II 7.4 18.6 6.4 6.9 9.1 10.8 8.9 7.5 7.0 12.0 6.6 9.85197.576 Fe II 4.7 13.5 ... 4.2 6.1 6.7 6.5 4.8 5.7 6.6 4.8 6.15234.630 Fe II 3.6 15.3 ... 5.8 7.7 9.2 7.5 4.0 5.8 9.2 5.3 8.75276.002 Fe II 4.5 18.6 ... 8.3 8.2 10.7 8.3 8.5 7.0 11.9 7.0 9.24518.023 Ti I ... 5.1 ... ... ... 5.0 ... ... ... ... ... ...4527.305 Ti I ... ... ... ... ... ... ... ... ... ... ... ...4533.239 Ti I 9.7 19.5 5.3 8.8 16.3 13.0 10.4 8.4 9.3 16.6 9.3 11.94534.778 Ti I ... 16.5 ... 5.9 12.0 13.6 6.5 6.4 4.5 9.7 10.4 10.64535.570 Ti I ... 12.6 ... 4.7 9.0 8.7 4.9 5.1 ... 5.3 7.7 9.54681.908 Ti I ... 5.9 ... ... 5.6 ... 3.6 ... ... 5.9 ... ...4981.732 Ti I 12.9 23.8 5.6 13.6 18.0 17.9 12.4 8.5 8.5 16.9 11.8 14.24991.067 Ti I 13.2 23.5 10.9 7.1 17.9 13.8 15.6 5.7 11.2 20.1 13.9 11.94999.504 Ti I 7.8 15.9 ... ... 12.1 13.1 8.6 5.9 5.7 11.0 7.2 9.95016.162 Ti I ... 3.6 ... ... ... ... ... ... ... ... ... ...5020.028 Ti I ... 7.5 ... ... 5.3 ... ... ... ... 5.2 ... ...5022.871 Ti I ... ... ... ... ... ... ... ... ... ... ... ...5035.907 Ti I ... ... ... ... ... ... ... ... ... ... ... ...5036.468 Ti I ... 5.9 ... ... ... ... ... ... ... ... ... ...5039.959 Ti I ... 6.4 ... ... ... ... ... ... ... ... ... ...5064.654 Ti I ... 5.8 ... ... ... ... 3.2 ... ... 5.4 ... ...4501.272 Ti II ... ... 17.5 40.4 ... 47.9 ... 35.6 34.6 ... 43.1 ...4563.761 Ti II 29.9 ... 14.0 33.8 ... 40.7 ... 27.5 27.9 ... 33.9 ...4571.968 Ti II ... ... 17.2 40.8 ... 45.7 ... 34.8 34.9 ... 41.0 ... 42 –Table 6a—Continued λ (˚A) Ele. 1 2 3 4 5 6 7 8 9 10 11 124589.958 Ti II 6.8 17.4 ... 6.6 7.1 11.8 8.5 7.6 6.3 14.6 6.6 9.74657.203 Ti II ... 5.1 ... ... ... ... ... ... ... ... ... ...4779.985 Ti II ... 10.0 ... ... ... 5.5 ... ... ... 6.0 ... ...5129.152 Ti II ... 9.6 ... ... 7.4 6.3 5.0 ... ... 9.2 ... 9.35154.070 Ti II ... 8.7 ... ... 6.1 ... 5.6 ... ... 5.6 ... ...5188.680 Ti II 7.9 27.6 ... 13.2 15.6 15.7 16.5 9.0 12.0 22.2 14.8 13.35226.543 Ti II 7.3 21.3 ... 12.9 11.2 12.6 11.8 6.5 8.6 16.2 9.2 12.35336.781 Ti II ... 12.5 ... ... 7.5 8.3 5.6 ... 4.2 6.8 ... 7.65381.018 Ti II ... 9.7 ... ... ... ... ... ... ... ... ... ... 43 –Table 6b. Measured Equivalent Widths λ (˚A) Ele. 13 14 15 16 17 18 19 20 21 22 233100.304 Fe I ... ... ... ... 46.4 ... ... ... ... ... ...3100.665 Fe I ... ... ... ... ... ... ... ... ... ... ...3116.631 Fe I 20.1 ... 11.6 ... 14.7 16.6 ... 26.5 29.9 15.1 ...4494.563 Fe I 15.5 26.8 4.9 24.1 12.6 9.9 ... 15.3 18.8 9.1 5.64528.614 Fe I 26.1 ... 8.2 ... 23.5 21.2 ... 27.8 31.6 17.1 8.04531.148 Fe I 7.0 16.2 ... 15.2 8.4 7.7 25.1 8.6 11.3 4.4 ...4556.126 Fe I ... 7.6 ... ... ... ... ... ... ... ... ...4592.651 Fe I ... 10.6 ... ... ... ... ... ... ... ... ...4602.941 Fe I 8.6 15.4 ... 12.6 5.0 5.1 23.3 9.2 10.3 3.1 ...4647.434 Fe I ... 6.1 ... ... ... ... ... ... ... ... ...4678.846 Fe I ... 5.3 ... ... ... ... ... ... ... ... ...4691.411 Fe I ... ... ... ... ... ... ... ... ... ... ...4733.591 Fe I ... ... ... ... ... ... ... ... ... ... ...4736.773 Fe I 6.5 13.5 ... 12.2 ... ... 19.9 7.2 9.1 4.6 ...4871.318 Fe I 17.6 30.1 6.4 30.3 16.8 13.1 ... 20.7 22.0 17.0 ...4872.137 Fe I 12.2 21.8 ... 22.7 12.4 8.7 33.4 16.0 17.8 12.8 ...4890.750 Fe I 17.6 31.7 5.0 28.7 15.5 13.4 ... 19.7 22.4 10.9 5.64891.490 Fe I 26.8 ... 10.2 ... 24.9 23.0 ... 30.6 35.0 18.6 8.34918.994 Fe I 20.7 34.8 13.0 26.8 ... 12.8 ... 20.6 24.4 12.6 6.64920.503 Fe I ... ... 18.0 ... 49.4 28.2 ... 35.9 44.3 24.8 14.74985.253 Fe I ... ... ... ... ... ... ... ... ... ... ...4985.547 Fe I ... ... ... ... ... ... ... ... ... ... ...4994.130 Fe I ... 13.9 ... ... ... ... ... ... ... ... ...5001.862 Fe I ... 11.8 ... ... ... ... ... ... ... ... ...5006.119 Fe I 13.0 24.8 ... 21.6 10.6 9.8 35.1 11.9 16.7 9.0 5.25012.068 Fe I 10.1 24.5 ... 20.0 11.0 9.5 32.0 12.4 18.3 7.1 3.65022.236 Fe I ... 6.5 ... ... ... ... ... ... ... ... ...5049.819 Fe I ... 16.3 ... ... ... ... ... ... ... ... ...5051.635 Fe I ... 15.4 ... ... ... ... ... ... ... ... ...5068.766 Fe I ... 8.4 ... ... ... ... ... ... ... ... ...5079.224 Fe I ... 5.4 ... ... ... ... ... ... ... ... ...5079.740 Fe I ... 6.0 ... ... ... ... ... ... ... ... ...5083.339 Fe I ... 5.4 ... ... ... ... ... ... ... ... ... 44 –Table 6b—Continued λ (˚A) Ele. 13 14 15 16 17 18 19 20 21 22 235098.697 Fe I ... ... ... ... ... ... ... ... ... ... ...5123.720 Fe I ... ... ... 5.8 ... ... 13.1 ... 6.5 ... ...5150.840 Fe I ... 7.5 ... ... ... ... ... ... ... ... ...5171.596 Fe I 17.4 31.6 5.5 27.2 13.3 12.5 ... 17.5 20.6 8.0 6.85192.344 Fe I 13.7 23.8 ... 23.3 11.0 8.6 33.2 13.3 18.7 6.4 6.85194.942 Fe I ... 17.8 ... ... ... ... ... ... ... ... ...5198.711 Fe I ... 5.3 ... ... ... ... ... ... ... ... ...5202.336 Fe I ... 10.3 ... ... ... ... ... ... ... ... ...5215.182 Fe I ... 6.4 ... ... ... ... ... ... ... ... ...5216.274 Fe I ... 13.0 ... ... ... ... ... ... ... ... ...5217.390 Fe I ... 6.5 ... ... ... ... ... ... ... ... ...5227.190 Fe I 36.6 ... 10.9 ... 30.2 31.0 ... 35.0 42.5 23.1 15.55232.940 Fe I 25.8 ... 7.8 ... 22.4 20.1 ... 28.8 34.7 16.0 8.15242.491 Fe I ... 5.6 ... ... ... ... ... ... ... ... ...5250.646 Fe I ... 5.5 ... ... ... ... ... ... ... ... ...5263.305 Fe I ... 6.9 ... ... ... ... ... ... ... ... ...5269.537 Fe I ... ... 30.2 ... 58.3 ... ... 66.4 73.9 ... 36.45328.039 Fe I ... ... 23.5 ... 51.3 ... ... 56.8 61.7 ... 28.95332.900 Fe I ... ... ... ... ... ... ... ... ... ... ...5339.930 Fe I ... 11.5 ... ... ... ... ... ... ... ... ...5341.024 Fe I 11.6 23.4 ... 23.1 ... ... 34.3 ... 15.4 ... ...5393.167 Fe I ... 10.8 ... ... ... ... ... ... ... ... ...5397.128 Fe I 29.4 ... 10.1 ... 27.6 24.6 ... 31.3 36.4 17.4 10.85405.775 Fe I 31.6 ... 11.5 ... 27.2 27.1 ... 32.2 40.6 19.7 13.15429.696 Fe I 32.8 ... 8.8 ... 29.3 28.4 ... 35.9 42.4 20.5 12.85434.524 Fe I 20.0 ... 5.8 33.5 18.5 16.3 ... 21.3 26.4 12.0 8.05455.609 Fe I 25.4 ... 5.9 36.7 22.3 19.2 ... 23.8 31.8 13.2 10.15497.516 Fe I ... 13.1 ... ... ... ... ... ... ... ... ...5501.465 Fe I ... 10.0 ... ... ... ... ... ... ... ... ...5569.618 Fe I ... 12.1 ... ... ... ... ... ... ... ... ...5572.841 Fe I 10.7 17.8 ... 16.0 9.0 6.3 26.6 9.0 14.3 ... ...5576.090 Fe I ... ... ... ... ... ... ... ... ... ... ...5586.756 Fe I 14.7 24.4 4.6 23.6 11.3 9.5 33.7 15.7 16.2 9.5 ... 45 –Table 6b—Continued λ (˚A) Ele. 13 14 15 16 17 18 19 20 21 22 235658.816 Fe I ... 6.6 ... ... ... ... ... ... ... ... ...4508.289 Fe II 10.5 13.8 3.8 11.7 8.9 5.4 26.3 9.5 14.3 5.8 ...4515.339 Fe II 9.6 9.4 ... 9.1 ... 3.2 20.3 7.9 10.4 7.0 ...4522.634 Fe II 15.1 15.7 7.1 16.5 9.3 6.2 ... 12.2 17.9 9.5 ...4576.339 Fe II 5.0 ... ... 5.4 ... ... 10.1 ... ... ... ...4583.837 Fe II 26.8 28.4 5.8 31.0 18.8 13.0 ... 21.0 29.3 16.0 8.54629.339 Fe II 9.8 11.7 ... 13.6 6.9 4.9 27.0 10.5 12.0 6.1 ...5197.576 Fe II 6.1 6.7 ... 10.7 ... 5.2 18.2 8.2 9.7 3.5 ...5234.630 Fe II 8.7 8.0 3.5 12.1 ... ... 22.7 8.4 12.2 5.2 ...5276.002 Fe II 9.2 9.3 ... 20.2 ... 5.2 28.2 10.6 15.0 6.2 ...4518.023 Ti I ... ... ... 5.3 ... ... 7.0 ... ... ... ...4527.305 Ti I ... ... ... ... ... ... 4.9 ... ... ... ...4533.239 Ti I ... ... 4.6 ... ... ... ... ... ... ... 7.04534.778 Ti I 11.9 15.3 ... 13.9 9.5 7.0 25.9 ... 12.4 7.0 5.54535.570 Ti I 9.5 6.2 ... 5.6 ... 3.7 17.2 7.7 6.7 4.1 ...4681.908 Ti I ... ... ... ... ... ... 8.1 ... ... ... ...4981.732 Ti I 14.2 16.0 4.3 14.0 9.4 8.4 ... 10.6 14.5 11.1 ...4991.067 Ti I 11.9 20.9 4.7 19.1 10.5 9.8 24.2 13.3 16.8 ... 5.74999.504 Ti I 9.9 10.6 ... 12.6 ... 4.2 21.7 7.7 11.5 4.6 ...5016.162 Ti I ... ... ... ... ... ... 5.3 ... ... ... ...5020.028 Ti I ... ... ... ... ... ... 5.0 ... ... ... ...5022.871 Ti I ... ... ... ... ... ... 9.7 ... ... ... ...5035.907 Ti I ... 5.5 ... ... ... ... 9.2 5.3 ... ... ...5036.468 Ti I ... ... ... 5.0 ... ... 7.0 ... ... ... ...5039.959 Ti I ... ... ... ... ... ... 7.9 ... ... ... ...5064.654 Ti I ... ... ... ... ... ... 8.0 ... ... ... ...4501.272 Ti II ... ... 13.6 ... 34.0 24.4 ... 38.1 51.5 29.9 13.74563.761 Ti II ... ... 11.7 ... 26.5 19.6 ... 30.8 42.7 24.6 13.04571.968 Ti II ... ... 15.3 ... 33.3 24.8 ... 36.1 50.6 30.7 13.14589.958 Ti II 9.7 11.1 ... 9.7 6.1 4.8 27.8 6.2 12.9 4.5 ... 46 –Table 6b—Continued λλ