Chemical abundances of fast-rotating stars in the LMC cluster

Chemical abundances of fast-rotating stars in the LMC cluster NGC 2004 NGC 2004 LMC (Richtler et al., 1998) M. F. Nieva1, S. Daflon1, K. Cunha1,2 , A. J. Korn3 1- Observatório Nacional (Brazil) 2- Department of Physics, University of Texas, El Paso (USA) 3- Uppsala Astronomical Observatory (Sweden) Introduction  Young clusters like NGC 2004 in LMC are ideal objects to study: • current state of chemical evolution in LMC • physics of hot stars  Why are the program stars important to be studied? high mass: 9-15 MSun rapid rotators: vseni > 100 kms-1 low metallicity: [M/H]= -0.5 • rotational mixing High mass stars: HHe in the core through CNO cycle Rapid rotation can lead to contamination of the atmospheres with CNO-cycled material from the core:  N enrichment  C and O depletion. This effect increases with lower metallicities, higher rotation rates and higher masses (Maeder & Meynet (2001), Meynet & Maeder (2002) ) .  Basically... Rotational mixing in hot stars:  Meridional circulation: plays a major role in the redistribution of internal angular momentum  Mixing increases with rotational rates  Main Sequence  Metallicity? Low metallicity  higher angular velocities  higher mixing effects How? Low metallicity  lower opacities  low mass loss by radiative driven winds  less angular momentum loss At lower metallicities, the star is able to maintain for longer time a high rotational velocity, and the effects of rotational mixing are higher There is evidence of surface enrichment induced by rotation in the Galaxy: C vsen(i): 450 kms-1 O9 III N/O N Solar Ab. Cunha & Lambert 1994 Daflon et al. 1999 N/O O vseni Daflon et al. (2001) Villamariz et al. (2002)  Goal of this work: Check the hypothesis of surface mixing induced by rotation in environments of low metallicity.  We combined our results with: Korn et al. (2000, 2002) => derived non-LTE abundances of slow rotators in LMC NGC 2004 (vsini lower than 70 kms-1).  How are the chemical abundances derived? Spectra from UVES on VLT LTE model atmospheres NLTE line formation Data The spectra were obtained using UVES on VLT. A sample spectrum in the region 4630-4670 of NGC 2004 D15 Object D15 C8 B24 B18 mV 15.20 15.10 15.10 14.8 Texp [h] 1.0 1.5 1.5 2.0 S/N 75 75 75 75 Date 12/00 12/00 12/00 12/00 Spectral Range: 3760-4980 Å Resolving power: R= 40000 Analysis Chemical abundances were determined from the fitting of synthetic non-LTE profiles to the observed spectra. Model atmospheres: LTE plane parallel models from ATLAS9 (Kurucz, 1993) (Main Sequence Stars) Spectral synthesis Non-LTE line formation was adopted in this analysis. Level populations were computed with program DETAIL and line profiles, with SURFACE (Giddings, 1981; Butler & Giddings, 1985): recent version. Atomic models O Becker & Butler (1988) N Becker & Butler (1989) Si Becker & Butler (1990) Mg Przybilla et al. (2001) Al Dufton et al. (1986) C Eber & Butler (1988): CII 4411 Å (Sigut (1996) ?, CII 4267 Å?)  Temperature Scale •Photometric Analysis (UBV, Q) •Spectroscopic Analysis (SiII/III,OII)  Abundance Results • Photometric Analysis (UBV, Q) Photometric data from different sources  Teff from literature: Caloi & Cassatella (1995)  UV fluxes Keller et al. (2002)  WFPC2 fotometry  Photometric calibrations: UVB photometry from Balona et al. (1993) Napiwotzki et al. (1993)  (B-V)0 Daflon et al. (1999)  Q* Lyubimkov et al. (2002)  Q * Q free-reddening parameter Q= (U-B) – X(B-V) X=E(U-B)/E(B-V) 1 Grid Teff-logg from Hg • Spectroscopic Analysis NLTE ionization equilibrium of Si II/III Simultaneous solution of : Teff, logg, e(Si), e(O), x (after some iterations) SiII 2 e(Si) vs. Teff for different x 3 e from OII and Si vs. x SiIII The Temperature Scale Comparison of Teff of the program stars and other of the same cluster in Korn et al. ’00,’02 from:  literature (photometric derivations)  different photometric calibrations  purely spectroscopic derivations different scales! Teff(Spec) - Teff(Phot) ~ 3500 K Check of abundance scale: D15 from Korn et al. ‘02 e(O) H II LMC1 Sun2 8.40 ± 0.2 8.66 ±0.05 e(N) 6.90 ±0.2 7.78 ± 0.06 e(C) 7.90 ±0.2 8.39 ±0.05 e(Si) 6.70 ±0.2 7.51 ±0.04 1- Garnett (1999), 2- Asplund et al. (2004) CII 4267 Very weak line CII 44411 The lower scale gives higher abundances for fixed values of x (Mg, Al: uncertain values). We found higher values of x for the lower Teff scale, and the O abundances were just relatively high in terms of the uncertainties. We decided to use the higher temperature scale in order to use the same abundance scale of the previous works and be able to compare our and their results. Stars with high vseni have several lines blended, and this feature causes 3 important effects: 1- the continnum gets lower and the difficulty to trace it rises with the number of blended lines 2- in a blend caused by a mixture of lines of different elements we have to consider both components fixing the abundance of one element by isolated lines before fitting the second element 3- in blends of lines of the same element we have to find the best simoultaneous fit for the two lines. Fits of some blended lines of the star B18 (Fixing 1st the abundance of one element, calculated by isolated lines, to fit the profile) Effect of vsini in metallic lines for the sample stars. O II and Si III lines vseni (km s-1) 50 109 124 134  Results (N,O,Mg,Al,Si, C??) NGC 2004 - LMC Galaxy Solid curves: evolution tracks of Schaerer et al. (1993): Z=0.004 N/O ratio vs. logg as an indicator of evolutionary state. N/O ratio vs. vseni shows no evidence of enhancement of N for fast rotators. Daflon et al. (2001) Conclusions  First results for fast-rotating MS stars in LMC  To form a conclusion about rotationally induced mixing one should wait for abundance results of more stars with high vseni. References Asplund. M, Grevesse, N., Sauval, J., astro-ph/410214v2 Balona, L. A.; Jerzykiewicz, M., 1993, MNRAS, 260, 782B. Becker, S. R.; Butler, K., 1988, A&A, 201, 232B. Becker, S. R.; Butler, K.,1989, A&A, 209, 244B. Becker, S. R.; Butler, K., 1990, A&A, 235, 326B. Butler, K., 1994, http://ccp7.dur.ac.uk/Docs/detail.ps Daflon, S., Cunha, K., Becker, S. R., 1999,ApJ, 522, 950D. Dufton, P. L.; Brown, P. J. F.; Lennon, D. J.; Lynas-Gray, A. E., 1986, MNRAS, 222, 713D. Eber, F.; Butler, K., 1988, A&A, 202, 153E. Garnett, D. R., 1999, in New Views of the Magellanic Clouds, ed. Y.-H. Chu, N. B., Suntzeff, J. E. Hesser, & D. A. Bohlender (Kluwer, Dordrecht), IAU Symp., Ser., 190, 266. Korn, A. J.; Becker, S. R.; Gummersbach, C. A.; Wolf, B., 2000, A&A, 353, 655K. Korn, A. J.; Keller, S. C.; Kaufer, A.; Langer, N.; Przybilla, N.; Stahl, O.; Wolf, B., 2002, A&A, 385, 143K. Kurucz, R., 1993, ATLAS( Stellar Atmosphere Programs and 2 km/s Grid CD-ROm 13, Smithsonian Astrophysical Observatory, Cambridge. Maeder, A., Meynet, G., 2001, A & A, 373, 555. Przybilla, N.; Butler, K.; Becker, S. R.; Kudritzki, R. P., 2001, A&A, 369, 1009P. Robertson, J. W., 1974, A&A, 15, 261R Schaerer, D.; Meynet, G.; Maeder, A.; Schaller, G., 1993, A&AS, 98, 523S. CNO cycle 1. 12C + 1H 13N +g 6. 15N + 1H 5. 15O 12C + 4He 2. 13N 13C + e+ + n e +g 15N + e+ + n e 4. 14N + 1H 15O 3. 13C + 1H +g 14N T > 2x106 K (cores of stars of M > 1.5 Msun) CNO catalyzed by 12C.