1. S-Process in C-Rich EMPS: predictions versus observations Sara Bisterzo (1) Roberto Gallino (1) Oscar Straniero (2) I. I. Ivans (3, 4) and Wako Aoki, Sean Ryan, Timoty C. Beers (1) Dipartimento di Fisica Generale, Università di Torino, 10125 (To) Italy (2) Osservatorio Astronomico di Collurania – Teramo, 64100 (3) The Observatories of the Carnegie Institution of Washington, Pasadena, CA, (USA) (3) The Observatories of the Carnegie Institution of Washington, Pasadena, CA, (USA) (4) Princeton University Observatory, Princeton, NJ (USA)

2. TP Convective envelope He-intershell 22 Ne(a,n) 25 Mg During the TDU (third dredge-up)  p ingestion in the top of He- intershell (few protons). At H-shell ignition  13 C-pocket formation via 12 C + p  13 N + , and 13 N(  ) 13 C At T~ 10 8 K  13 C(a,n) 16 O in radiative conditions  s-process. 13 C(a,n) 16 O Neutron source: 12 C(p,  ) 13 N(  + ) 13 C( ,n). Type: primary When: interpulse T 6 >90. Where: He-intershell Density: 10 6 -10 7 ( n/cm 3) Straniero et al. 1995, Gallino et al. 1998 The AGB engine

3. The two neutron sources in AGB stars 13 C( ,n) 16 O 22 Ne( ,n) 25 Mg Needs 13 C Major neutron source 13 C-pocket Primary source! T 8 = 0.9-1 Interpulse phase (1- 0.4) 10 5 yr Radiative conditions N n = 10 7 cm -3 Abundant 22 Ne Minor neutron source Neutron burst Secondary (primary) source T 8 = 3 (low 22 Ne efficiency) Thermal pulse 6 yr Convective conditions N n (peak) = 10 10 cm -3

4. AGB models at very low [Fe/H] M = 1.5 Msun 1.2 M sun < M < 3 M sun 13 C-pocket: ST*2 …. ST/100 Constant pulse by pulse ( ST: 4. 10 -6 M sun, [Fe//] = -0.3, Reproduction of Solar Main Component ) 1.2 Msun  3 pulses 1.3 Msun  6 pulses 1.4 Msun  8 pulses 1.5 Msun  20 pulses 2 Msun  26 pulses 3 Msun  30 pulses Mass loss : from 10 -7 to 10 -4 M sun /yr  Reimers 1.2 Msun  η = 0.3 1.3 Msun  η = 0.3 1.4 Msun  η = 0.3 1.5 Msun  η = 0.3 2 Msun  η = 0.5 3 Msun  η = 1

5. At very low metallicity Today, Intrinsic AGB halo stars: typical mass is ~ 0.6 Msun (initial mass 0.8 – 0.9 Msun)   NO TDU  No C or s-process enrichment observable.

6. Then all CRUMPS are Extrinsic AGB stars: Binary systems  transfer of material C- and s-rich on the companion (through stellar wind, Roche Lobe …). The unevolved companion shows the tipical AGB composition, while the true AGB star is now a White Dwarf.

7. Extrinsic AGB models Diluition factor: used to simulate the mixing effect in the envelope of the extrinsic stars Note:  for main sequence stars dil ≈ 0  for giants dil may be important

8. AGB models: envelope abundances ls hs Pb M ≈ 1.5 Msun

9. To reproduce stars with both s+r enhancements Different choice of initial chemical abundances of Eu in the progenitor clouds [Eu/Fe] ini from 0.5 to 1.5 and 2.0

10. Effect of pre r-enrichment in s-enhanced stars Model with pre r-enrichment normalized to [Eu/Fe] ini = 2.0 in the parental cloud: the envelope abundances in these stars are predicted by mass transfer from the more massive AGB companion in a binary system which formed from a parental cloud already enriched in r elements. r-process rich AGB star model of M ≈ 1.3 M sun with [Fe/H] = - 2.60. NO r-process rich [Eu/Fe] ini = 2.0[Eu/Fe] ini = 0.0

11. Choice of initial abundances The choice of the initial r-rich isotope abundances normalised to Eu is made considering the r- process solar prediction from Arlandini et al.1999.

12. 1- Lead stars (C, s, Pb rich) 2 – C and s+r rich Lead stars

13. References 1. J. A. Johnson, M. Bolte, ApJ 579, L87 (2002) 2. W. Aoki, et al., ApJ 580, 1149 (2002) 3. T. Sivarani, et al., A&A 413, 1073 (2004) 4. J. A. Johnson, M. Bolte, ApJ 605, 462 (2004) 5. W. Aoki, et al., ApJ 561, 346 (2001) 6. S. Van Eck, S. Goriely, A. Jorissen, B. Plez, A&A 404, 291 (2003) 7. S. Lucatello, et al., AJ 125, 875 (2003) 1.8* 8. J. G. Cohen, N. Christlieb, Y. Z. Quian, G. J. Wasserburg, ApJ 588, 1082 (2003) 9. B. Barbuy, et al., A&A 429, 1031 (2005) 11. I. Ivans et al., ApJ accepted (2005) [Eu/Fe] measured; **sigma(dil) = ± 0.2 dex NOTE: Initial Mass are estimates dependent also on mass loss rates adopted

14. T eff = 5850 K

15. T eff = 6625 K

16. 2 – C and s+r rich Lead stars

17. 0.0 Without r-process enhancement  [Eu/Fe] ini = 0.0 With r-process enhancement  [Eu/Fe] ini = 2.0 HE2148-1247 Cohen et al. 2003 T eff = 6380 K

18. Prediction updated Extrinsic AGBs indicator CS29497-030 Ivans et al. 2005 With r-process enhancement  [Eu/Fe] ini = 2.0 Without r-process enhancement  [Eu/Fe] ini = 0.0 T eff = 7000 K

19. The s elements enhancement in low-metallicity stars interpreted by mass transfer in binary systems (extrinsic AGBs). For extrinsic AGBs [Zr/Nb] ~ 0. Instead, for intrinsic AGBs [Zr/Nb] ~ – 1. Zr over Nb: Intrinsic or Extrinsic AGBs M ≈ 1.3 M sun [Fe/H] = -2.60 Fig. 2 s-process path Case ST*2 [Eu/Fe] ini = 2.0

20. Without r-process enhancement  [Eu/Fe] ini = 0.0 With r-process enhancement  [Eu/Fe] ini = 1.5 CS29497-34 Barbuy et al. 2005 T eff = 4800 K

21. Without r-process enhancement  [Eu/Fe] ini = 0.0 With r-process enhancement  [Eu/Fe] ini = 1.8 CS31062-050 Aoki et al. 2002 T eff = 5600 K

22. With r-process enhancement  [Eu/Fe] ini = 1.8 T eff = 5500 K

23. Barklem et al. 2005: s-enhanced stars Star[Fe/H][C/Fe][Mg/Fe][Sr/Fe][Y/Fe][Zr/Fe][Ba/Fe][La/Fe][Ce/Fe][Nd/Fe][Sm/Fe][Eu/Fe] CS 22892-052-2.951.000.120.610.45-1.191.02-1.14-1.54 HE 0131-3953-2.712.450.300.46--2.201.941.931.76-1.62 HE 0202-2204-1.981.16-0.010.570.410.471.411.361.301.021.030.49 HE 0231-4016-2.081.360.220.670.72-1.471.221.531.35-- HE 0338-3945-2.412.070.390.73 -2.412.262.212.09-1.89 HE 0432-0923-3.190.240.340.470.510.880.72----1.25 HE 1105+0027-2.422.000.470.730.75-2.452.10-2.06-1.81 HE 1127-1143-2.730.540.220.240.22-0.63--0.86-1.08 HE 1135+0139-2.331.190.330.660.360.461.130.931.170.77-0.33 HE 1343-0640-1.900.770.370.680.510.980.70----- HE 1430-1123-2.711.840.350.240.50-1.82--1.72-- HE 2150-0825-1.981.350.360.660.850.971.701.411.481.42-- HE 2227-4044-2.321.670.300.41--1.381.28---- HE 2240-0412-2.201.350.280.24--1.37-----

24. 0

27. CONCLUSIONS: The spectroscopic abundances of low-metallicity s- and r-process enriched stars are interpreted using theoretical AGB models (FRANEC CODE), with an initial composition already enriched in r elements from the parental cloud from which the binary system was formed. [Zr/Nb] is an indicator of an extrinsic AGB in a binary system: [Zr/Nb] ~ 0 for an extrinsic AGB, [Zr/Nb] ~ – 1 for an intrinsic AGB. [Zr/Nb] is an indicator of an extrinsic AGB in a binary system: [Zr/Nb] ~ 0 for an extrinsic AGB, [Zr/Nb] ~ – 1 for an intrinsic AGB. Spectroscopic determination of [Na/Fe] and [Mg/Fe] permits an estimate of the initial AGB stellar mass. Spectroscopic determination of [Na/Fe] and [Mg/Fe] permits an estimate of the initial AGB stellar mass.

28. CONCLUSIONS: Open Problem: the strong discrepancy of C and N predictions with respect to observations may be reconciled: Open Problem: the strong discrepancy of C and N predictions with respect to observations may be reconciled: (1) by introducing the effect of cool bottom process (CBP) in the TP-AGB phase (*); (2) for N and [Fe/H] < -2.3, by the effect of Huge First TDU (see Gallino presentation). (3) Uncertainties in the spectroscopic abundances of C, N, O, Na, Mg  M. Asplund, ARAA 2005 (*) Nollett, K. M., Busso, M., Wasserburg, G. J., ApJ 582, 1036 (2003); Wasserburg, G. J., Busso, M., Gallino, R., Nollett, K. M., (2006), Nucl. Physics, in press.