First stars in Damped Lyman Alpha systems

First stars in Damped Lyman Alpha systems


In order to characterize Damped Lyman Alpha systems (DLAs) potentially hosting first stars, we present a novel approach to investigate DLAs in the context of Milky Way (MW) formation, along with their connection with the most metal-poor stars and local dwarf galaxies. The merger tree method previously developed is extended to include inhomogeneous reionization and metal mixing, and it is validated by matching both the Metallicity Distribution Function of Galactic halo stars and the Fe-Luminosity relation of dSph galaxies. The model explains the observed -Fe relation of DLAs along with the chemical abundances of [Fe/H] systems. In this picture, the recently discovered C-enhanced DLA (Cooke et al. 2011a), pertains to a new class of absorbers hosting first stars along with second-generation long-living low-mass stars. These ”PopIII DLAs” are the descendants of H-cooling minihalos with , that virialize at in neutral, primordial regions of the MW environment and passively evolve after a short initial period of star formation. The gas in these systems is warm  K, and strongly C-enriched by long-living, extremely metal-poor stars of total mass .

galaxies: abundances, evolution, stellar content - cosmology : theory - stars: Population II

1 Motivation

Damped Ly absorption systems (DLAs) are high-column density neutral gas reservoirs, cm, observed at intermediate redshifts, , in the spectra of distant quasars. Although their nature is still unclear (Pettini 2004), the key role of DLAs to understand galaxy formation (Wolfe, Gawiser & Prochaska 2005) is widely recognized. So far, more than DLAs have been observed and the iron abundance measured in systems (Prochaska et al. 2007). Among these, very metal-poor (VMP) DLAs with [Fe/H], can be used to study the initial phases of heavy element enrichment of the interstellar medium (ISM) of early galaxies. Indeed, if VMP stars observed today in the Galactic halo and in nearby dwarf spheroidal galaxies (dSphs) are the living fossils of the first stellar generations, VMP DLAs may well constitute the gas reservoir out of which such pristine stellar populations formed. Following the medium resolution study of VMP DLAs by Penprase et al. (2010), Cooke et al. (2011b) have recently presented a high spectral resolution sample, including VMP systems. In these DLAs [C/O] and [Fe/O] independently of [Fe/H] and with little scatter, in agreement with measurements in VMP Galactic halo stars (Fabbian et al. 2009). So far, the only exception to this general trend is represented by a DLA with [Fe/H] and cm observed in the spectrum of the QSO J0035-0918 at (Cooke et al. 2011a). This system has [C/Fe], i.e. times larger than any other DLA. Moreover, its abundance pattern shows a clear ’odd-even’ effect and is consistent with the predictions for the yields of faint supernovae (SN) with (Kobayashi et al. 2011). Are we observing for the first time a DLA whose gas retains the imprint left by the first stars?

To characterize DLAs potentially hosting the first stars or their ashes, we propose a novel approach that simultaneously follows the evolution and chemical properties of DLAs and their connection with the most metal-poor stars and galaxies observed in the Local Universe based on the results obtained using the merger-tree code GAMETE (Salvadori, Schneider & Ferrara 2007, SSF07; Salvadori, Ferrara & Schneider 2008; Salvadori, & Ferrara 2009, SF09).

2 Model summary and validation

Figure 1: Left panel: observed (points with error-bars) and simulated (histogram) metallicity distribution function of Galactic halo stars. The data points are the sample by Beers & Christlieb (2006) with the inclusion of the three hyper-iron poor stars (Christlieb et al. 2002, 2006; Frebel et al. 2005). The histogram is the average MDF value over 50 merger history of the MW, re-normalized to the total number of observed stars with [Fe/H]. The shaded area represents errors. Right panel: observed (points with error-bars) and simulated (contours) iron-luminosity relation for the MW dSph galaxies. The data points are by Kirby et al. (2008) with the inclusion of the two faintest dSphs (Willman et al. 2005, Geha et al. 2009). The colored shaded areas correspond to regions that include the of the total number of possible dSph candidates selected in 50 merger histories of the MW.

The model key points can be summarized as follows (see SSF07 and SF09 for details):

  1. Hierarchical merger histories of a MW-sized dark matter (DM) halo are reconstructed from via a Monte Carlo algorithm based on the Extended Press-Schechter theory (SSF07).

  2. Star formation (SF) is followed along the tree in halos exceeding a mass threshold, , whose evolution (Fig. 2) is governed by: (a) the photo-dissociating Lyman-Werner (LW) background, quenching H-formation in  K minihalos; (b) the gas temperature in ionized regions of the Galactic Medium (GM), preventing gas-infall in halos with lower than a threshold value, . We assume that SF is active in  K minihalos at (Dijkstra et al. 2004). At lower , (and hence ) is assumed to linearly increase up to the value set by the end of reionization  K (Kitayama et al. 2000) for .

  3. Inhomogeneous reionization is modeled by random sampling the reionization history implied by to switch off (on) gas accretion in minihalos that form in ionized (neutral) regions.

  4. The SF rate is taken to be proportional to the mass of cold gas, , where is the SF efficiency and the halo free-fall time. In minihalos is reduced as K due to the ineffective cooling by H molecules (SF09). Low-mass, Population II (PopII) stars form according to a Larson IMF when the gas metallicity exceeds (Schneider et al. 2002). At lower metallicity, PopIII stars form with a reference mass value and explosion energy  erg consistent with faint SNe.

  5. The abundance evolution of different chemical elements3 (from C to Zn) is traced in both the ISM and in the GM by taking into account mass- and metallicity-dependent stellar evolutionary timescales (Raiteri, Villata & Navarro 1996) and SN feedback (Salvadori, Ferrara & Schneider 2008).

  6. To account for the incomplete mixing of SN ejecta within the ISM of gas-poor galaxies, gas outflows have a metallicity , where is the mass of newly formed metals, is the mass of gas ejected out of the halo, and is a function of the gas mass that varies in the range .

  7. The probability for newly formed halos to reside in a GM metal enriched region is , where is the filling factor of metal bubbles within the MW physical volume4, and is the volume filling factor of fluctuations with overdensity above the critical threshold, , for the linear collapse (Miralda-Escudé, Haehnelt & Rees 1999). The latter quantity describes the abundance of high-density regions, in which metals first penetrate (Tornatore, Ferrara & Schneider 2007). Objects in enriched (primordial) regions are assigned an initial metallicity ().

The model is calibrated by best-fitting the SF and feedback efficiencies to reproduce the global properties of the MW (stellar/gas mass and metallicity). In Fig. 1 we show that the average metallicity distribution of [Fe/H] stars over 50 possible MW merger histories matches the observed Galactic halo MDF. In the same Figure, the theoretical Fe-Luminosity relation for dSph galaxies has been obtained by evolving in isolation star-forming halos with , candidates to remain MW satellites (Diemand, Madau & Moore 2001; SF09 for details). In Fig. 2 is compared with that obtained by using data-constrained reionization models. To this end we compute the dissociating LW background intensity associated with an early/late reionization history (Gallerani, Choudhury & Ferrara 2006) by using the results provided in Fig. 6 by Ahn et al. (2009); we then convert the flux into a critical mass following Machacek, Brian & Abel (2001) and extrapolate to higher masses. At each redshift our is within the range allowed by the two reionization histories (gray shaded area).

3 The dwarf galaxy zoo

Figure 2: Dark matter content of MW satellites () as a function of their formation redshift. The two lines show the evolution of: the minimum mass of star-forming haloes, (solid), the halo mass corresponding to density peaks (dashed). The dotted shaded area identifies the region populated by H-cooling minihalos with  K. The number labels and colors specify different populations of satellites (see the text). The gray shaded area delimits the evolution of predicted by assuming an early (upper limit) and a late (lower limit) reionization history (Gallerani et al. 2007).

We now apply our model to DLAs starting from the origin of the C-enhanced, [Fe/H] DLA at . Two possibilities must be considered: (a) a recently virialized halo, later merging into the MW, or (b) a satellite galaxy which assembled at earlier epochs and evolved in isolation. The first possibility can be excluded on the basis of a metallicity argument: metal-free regions of the MW environment are expected to disappear at when . Afterward, newly virializing halos form from metal enriched GM gas, whose initial abundances [X/H][X/H]. Since at , and [Fe/H], proto-galaxies with an initial iron abundance [Fe/H] cannot form anymore.

In the second hypothesis, the dark matter (DM) halo hosting the DLA must be associated to density fluctuations at its final assembling epoch in order to evolve in isolation (no further merger or accretion) and become a satellite. By inspecting Fig. 2 we can distinguish among different populations of objects that satisfy this condition: (1) Ly-cooling halos that assembled after the end of reionization; Fornax and LeoI, the most luminous among the observed dSphs, are predicted to belong to this domain; (2) star-forming Ly-cooling halos that assembled before the end of reionization; classical dSphs, , such as Sculptor (SF09) are members of this population. (3) H-cooling, inefficiently star-forming minihalos that appeared at ; this is the domain of ultra-faint dSphs (SF09). (4) ”sterile” halos that formed before the end of reionization and unable to trigger SF (). We selected the satellite candidates in 50 possible merger histories of the MW, and evolve them in isolation down to redshift , when we compute their H column density, , assuming that the gas is fully neutral with a molecular weight :


the gas radius is written as , with for star-forming galaxies, that have presumably developed a disk (Ferrara, Pettini & Shchekinov 2000), and otherwise. DLAs are canonically defined as systems with cm (Wolfe et al. 2005). Interestingly, we find that systems which populate Zone (2), i.e. the progenitors of classical dSphs, have already exhausted most of their gas by , i.e. . On the other hand, many objects among those formed in Zones (1), (3) and (4), have H column densities compatible with DLAs. We will then focus on these candidates.

Figure 3: Gas abundances of DLAs with respect to the solar values (Asplund et al. 2009). Points are the observed values in VMP DLAs at (triangles, Cooke+2011b) and in the C-enhanced DLA at (square, Cooke+2011a). The color shaded areas show the results of the model for the three different populations of DLAs at : PopIII (blue), very metal-poor (violet), and metal-poor DLAs (orange). For each population the intensity of the colors correspond to regions containing the ()% of DLAs.

4 DLAs and first stars

A comparison between the properties of candidate and observed DLAs at is given in Fig. 3 and Fig. 4. Objects formed in different Zones result in distinct DLA populations, differing in their chemical abundances and . Systems in Zone (3) are segregated in the C-rich, metal-poor, low- regions of the plots, matching the properties of the C-enhanced DLA. What is the origin of these systems?

According to our model, C-enhanced DLAs are hosted by H-cooling minihalos, which virialized before the end of reionization out of metal-free, neutral regions of the MW environment. Because of their low DM content the star formation history of such a primordial proto-galaxies is extremely short. Some tens of Myrs after the onset of star-formation the intensity of LW background becomes high enough () to suppress further SF activity, thus turning them into ”sterile” minihalos. Metal-free stars are hosted by this population of DLAs, to whom we will hereafter refer as PopIII DLAs.

The metal-enrichment by first stars gently proceeds in these objects, mainly because of ineffective cooling by H molecules and their low gas-mass content, , favoring metals and gas loss (). As soon as , however, ”normal” PopII stars can form and contribute to enrichment. Given the yields by faint SNe, the PopIII-to-PopII transition occurs when [Fe/H], implying that most of the iron observed in these systems originates from these second generations of stars. The mass of relic stars in PopIII DLAs is expected to vary between , the most star-rich systems being the most enriched ones. The rise of the C abundance at increasing metallicity (Fig. 3) reflects the gradual contribution by low-metallicity AGB stars, mainly producing C, some O, and limited amounts of N (for systems with [Fe/H] we find [C/H], [N/H] and [O/H]). The associated decrease (Fig. 4) is caused by gas consumption in the most star-rich systems, due to both astration and gas loss. Note that our model does not predict any [C/O] increase towards low [O/H] values, as reported by medium resolution observations (Penprase et al. 2010).

The properties of VMP DLAs are well reproduced (Fig. 3) by [Fe/H] starless systems formed in Zone (4), passively evolving since their assembly epoch, . They form through merging of primordial and metal-enriched progenitors, virialized out of neutral regions of the MW environment. The metallicity spread of these systems depends on the gas enrichment in progenitor minihalos. Abundance ratios, instead, closely reflect those of the GM at the time of formation. Although the overall GM metallicity increases with time, abundances ratios get locked to the dominant stellar population which contribute to the enrichment, i.e. type II SNe. As a consequence, [C/O] and [C/Fe] ratios show little dispersion and are independent of , but also of redshifts, because of the passive evolution of these starless DLAs. These findings are in perfect agreement with the new observational results by Becker et al. (2011).

Figure 4: distribution vs. iron abundances of DLAs. Points with error-bars are observed values for metal-poor DLAs at (circles, Prochaska et al. 2007), VMP DLAs at (triangles, Cooke+2011b) and for the C-enhanced DLA at (square, Cooke+2011a). The color shaded areas show the results of the model for the three different populations of DLAs predicted to exist at (see Fig. 3).

Finally, [Fe/H] DLAs are  K halos forming through merging of star-rich progenitors either before (Zone 4) or after (Zone 1) the end of reionization. Their systematically higher is a consequence of self-enrichment by a substantial () population of long-living stars. Similarly, the high [C/Fe] ratios reflect the strong contribution by AGB stars. The gas mass in these DLAs is , thus resulting in higher values, despite of the lower formation and larger (eq. 1). Note however that the derived for these DLAs must be interpreted as an upper limit. In fact, objects (Fig. 3) are actively star-forming at , with yr, and hence part of their gas is presumably ionized. Moreover, the [Fe/H] value ([C/Fe]) has to be interpreted as a lower (upper) limit, since the contribution by SN type Ia may be significant in star-rich DLAs (Calura, Matteucci & Vladilo 2002). The same interpretation has to be applied to [Fe/H] measurements, as Fe is depleted onto dust grains (Noterdaeme et al. 2008). Barred these limitations, included the unknown number of surviving satellites, we note that the apparent concentration of [Fe/H] DLAs in a small region of the plane (Fig. 4) implies that these absorbers are only partially representative of the progenitors of MW-like systems (Pontzen et al. 2008).

5 Discussion

The recently discovered C-enhanced DLA pertains to a new class of systems, dubbed PopIII DLAs, hosting the first stars along with second generations of long-living stars. These systems are associated to H-cooling minihalos that virialize at in neutral and primordial regions of the MW environment, and only experience a short period of SF before the increasing LW background burns out their H. Once ”sterilized” these minihalos passively evolve as an inert gas reservoir, , retaining the imprint of the stellar generations they hosted. Their gas temperature, , is regulated by the balance between molecular/metal radiative cooling and photo-heating by the external ionizing radiation (Black 1981):


where is the gas density, the gas cooling rate (Maio et al. 2007),  eV the mean UV background photon energy, and the optically-thick H photo-ionization rate (Abel & Mo 1998). cms with and . We find that [Fe/H] minihalos have , implying that these systems can likely survive photo-evaporation thanks to self-shielding and metal cooling. Their gas temperature increases with decreasing metallicity, in agreement with the results by Kanekar et al. (2009). The C-enhanced DLA has  K while in [Fe/H] () systems  K.

The peculiar abundance pattern observed in the DLA does not result from faint SNe, but rather from the enrichment by low-metallicity SNII and AGB stars, which may start to form as soon as . While SNII nucleosynthetic products are mostly lost in winds, AGB metals are retained in the ISM, causing a dramatic increase of [C/Fe]. The chemical evolution of [Fe/H] DLAs is independent on the assumed yields and IMF of PopIII stars, confirming the role of ordinary PopII stars in driving the enrichment of VMP systems (SSF07), recently emphasized by the detection of a star with a ”normal” chemical abundance pattern (Caffau et al. 2011). If , as suggested by the existence of such star, the PopIII-PopII transition would be even quicker. The mass of relic stars in PopIII DLAs is found to be , implying that they are the gas-rich counterpart of the faintest dSphs.

As stated, the C-enhanced DLA is a minihalo. However, we cannot exclude a different interpretation, in which such absorber might be a newly formed halo virializing at from a rare, metal-free region of the Inter Galactic Medium, and actively forming PopIII stars. Since the total mass of the star-forming DLA is , strong feedback is required to substantially remove the initial gas mass and match the observed . By determining the dark matter content and SF rate of the C-enhanced DLA, it would be possible to disentangle these two pictures. We finally note that our simple semi-analytical model, which holds similarities with that proposed by Abel & Mo (1998) for Lyman Limit Systems, prevent us from making specific predictions on the number of DLAs at . However, the relative contribution of C-enhanced DLAs to the total population is expected to be extremely low, , thus explaining the rarity of these systems.


We thank R. Cooke, P. Molaro, P. Petitjean, X. Prochaska & R. Schneider for enlightening comments on the draft version of the paper. S.S. acknowledges a NOVA fellowship granted by the Netherlands Research School for Astronomy.


  1. pagerange: First stars in Damped Lyman Alpha systemsReferences
  2. pubyear:
  3. For stars we use yields by van den Hoek & Groenewegen (1997) () and by Meynet & Maeder (2002) (); for more massive stars we use Woosley & Weaver (1995) with a systematic halving of the Fe yield (Timmes, Woosley & Weaver 1995); yields for faint SNe are from Kobayashi et al. (2011).
  4. We assume  Mpc.


  1. Abel T. & H. J. Mo, 1998, ApJL, 494, 151
  2. Ahn K., Shapiro P., Iliev I. T., Mellema G. & Pen U. L., 2009, ApJ, 695, 1430
  3. Asplund M., Grevesse N., Sauval A. J., Scott P., 2009, ARA&A, 47, 481
  4. Becker G. D., Sargent W. L. W., Rauch M. & Carswell R. F., ApJ accepted
  5. Beers T. C., Christlieb N., 2006, private communication
  6. Black J. H., 1981, MNRAS, 197, 553
  7. Calura F., Matteucci F., Vladilo G., 2003, MNRAS, 340, 59
  8. Caffau E. et al., 2011, Nature, 477, 67
  9. Cooke R. et al. , 2011a, MNRAS, 412, 1047
  10. Cooke R. et al. , 2011b, MNRAS, 417, 1534
  11. Christlieb N. et al. , 2002, Nat, 904, 419
  12. Diemand J., Madau P., Moore B., 2005, MNRAS, 364, 367
  13. Dijkstra M., Haiman Z., Rees M. J., Weinberg D. H., 2004, ApJ, 601, 666
  14. Fabbian D., Nissen P. E., Asplund M., Pettini M. & Akerman C., 2009, A&A, 500, 1143
  15. Ferrara A., Pettini M. & Shchekinov Y., 2000, MNRAS, 319, 539
  16. Frebel A., Christlieb N., Norris J. E., Aoki W., Asplund M., 2005, Nat, 434, 871
  17. Kanekar N. et al., 2009, ApJ, 705, L40
  18. Kobayashi C., Tominaga N., Nomoto K., 2011, ApJ, 730, 14
  19. Kirby E. N., Simon J. D., Geha M., Guhathakurta P., Frebel A., 2008, ApJ, 685, 43
  20. Kitayama T., Tajiri Y., Umemura M., Susa H., Ikeuci S., 2000, MNRAS, 315, 1
  21. Gallerani S., Choudhury T. & Ferrara A., 2006, MNRAS, 360, 1401
  22. Geha M. et al., 2009, ApJ, 629, 1464
  23. Machacek M. E., Brian G. L. & Abel T., 2001, ApJ, 548, 509
  24. Meynet G. & Maeder A., 2002, A&A, 390, 561
  25. Maio U., Dolag K., Ciardi B. & Tornatore L., 2007, MNRAS, 379, 963
  26. Miralda-Escudé J., Haehnelt M. & Rees M. J., 2000, ApJ, 530, 1
  27. Noterdaeme et al. 2008, A&A, 481, 327
  28. Penprase et al. 2010, ApJ, 721, 1
  29. Pettini M., 2004, Cambridge University Press 257
  30. Pontzen A. et al., 2008, MNRAS, 390, 1349
  31. Prochaska J. X. et al., 2007, ApJs, 171, 29
  32. Raiteri C. M., Villata M. & Navarro J. F., 1996, A&A, 105, 315
  33. Salvadori S., Schneider R. & Ferrara A., 2007, MNRAS, 381, 647
  34. Salvadori S., Ferrara A. & Schneider R., 2008, MNRAS, 386, 348
  35. Salvadori S. & Ferrara A., 2009, MNRAS, 395, 6
  36. Schneider R., Ferrara A., Natarajan P., Omukai K., 2002, ApJ, 571, 30
  37. Timmes F. X., Woosley S. E. & Weaver T. A., 1995, ApJs, 98, 617
  38. Tornatore, A. Ferrara & Schneider R., 2007, MNRAS, 382, 945
  39. van den Hoek L. B., Groenewegen M. A. T., 1997, A&AS, 123, 305
  40. Willman B. et al. , 2005, ApJ, 129, 2692
  41. Wolfe A. M., Gawiser E., Prochaska J. X., 2005, ARA&A, 43, 861
  42. Woosley S. E., Weaver T. A., 1995, ApJ, 101, 181
Comments 0
Request Comment
You are adding the first comment!
How to quickly get a good reply:
  • Give credit where it’s due by listing out the positive aspects of a paper before getting into which changes should be made.
  • Be specific in your critique, and provide supporting evidence with appropriate references to substantiate general statements.
  • Your comment should inspire ideas to flow and help the author improves the paper.

The better we are at sharing our knowledge with each other, the faster we move forward.
The feedback must be of minimum 40 characters and the title a minimum of 5 characters
Add comment
Loading ...
This is a comment super asjknd jkasnjk adsnkj
The feedback must be of minumum 40 characters
The feedback must be of minumum 40 characters

You are asking your first question!
How to quickly get a good answer:
  • Keep your question short and to the point
  • Check for grammar or spelling errors.
  • Phrase it like a question
Test description