EDP Sciences logo
Subscriber Authentication Point
Search
Menu
Home All issues Volume 689 (September 2024) A&A, 689 (2024) A89 Full HTML
Open Access
Issue
A&A
Volume 689, September 2024
Article Number A89
Number of page(s) 13
Section Stellar atmospheres
DOI https://doi.org/10.1051/0004-6361/202449290
Published online 03 September 2024

© The Authors 2024

Licence Creative CommonsOpen Access article, published by EDP Sciences, under the terms of the Creative Commons Attribution License (https://creativecommons.org/licenses/by/4.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

This article is published in open access under the Subscribe to Open model. Subscribe to A&A to support open access publication.

1 Introduction

Lithium (Li) is one of the few nuclei produced during the Big Bang nucleosynthesis. Because of their fragility, the Li nuclei are immediately destroyed in stellar layers exceeding temperatures of ~2.5 × 106 K. In a low-mass star, the surface Li abundance A(Li) changes with time, and its changes reflect the main episodes of mixing occurring during the evolution of the star. When the convective envelope deepens into the stellar interior reaching stellar layers hot enough to burn Li, Li-free material is brought to the surface reducing A(Li). The evolution of A(Li) with the luminos- ity (or the gravity) can be described as two phases with constant A(Li), namely the Spite Plateau (Spite & Spite 1982; Bonifacio & Molaro 1997; Aoki et al. 2009) and the lower red giant branch (RGB) plateau (Mucciarelli et al. 2012, 2022), both followed by an abrupt drop corresponding to the first dredge-up (FDU) and the RGB-bump (RGBb) mixing episode, respectively. The A(Li) depletion at the FDU is a natural consequence of the standard stellar evolution (Iben 1967), while to explain the second drop at the RGBb, non-canonical mixing processes have to be included, the most popular one being the thermohaline mixing (see e.g. Charbonnel et al. 2020).

Within this framework, the so-called Li-rich stars are a class of peculiar and rare objects whose A(Li) significantly exceeds that measured in stars of a similar evolutionary stage. In some cases (see e.g. Koch et al. 2011; Kowkabany et al. 2022), metal-poor stars have an A(Li) higher than the abundance formed during the Big Bang and inferred by the baryon density obtained from the Wilkinson Microwave Anisotropy Probe (WMAP) and Planck satellites (A(Li)=+2.72±0.04 dex, Coc & Vangioni 2017), while some metal-rich stars have an A(Li) higher than that of the interstellar medium (A(Li)~+3.3 dex; Asplund et al. 2009). These very high A(Li) values suggest that Li is not preserved, and other production mechanisms must be at work. Their fre- quency is ~1% or less of all the stars (Kirby et al. 2016; Casey et al. 2016; Gao et al. 2019; Deepak et al. 2020). They have been observed at all metallicity (Ruchti et al. 2011; Martell & Shetrone 2013; Gao et al. 2019; Sitnova et al. 2023), at all evolutionary stages (Kirby et al. 2016; Li et al. 2018), in the Milky Way halo (Ruchti et al. 2011; Li et al. 2018), the thick disc (Monaco et al. 2011), the thin disc (Casey et al. 2016; Deepak et al. 2020), the bulge (Gonzalez et al. 2009), dwarf spheroidal galaxies (Kirby et al. 2012), globular clusters (GCs, Smith et al. 1999; Ruchti et al. 2011; Koch et al. 2011; Monaco et al. 2012; D’Orazi et al. 2015; Kirby et al. 2016 Mucciarelli et al. 2019; Sanna et al. 2020), and in open clusters (Monaco et al. 2014).

Li-rich stars could be explained as invoking different classes of processes, namely the internal production or external origin of the extra Li. Three main groups of processes have been proposed to explain the existence of these stars, namely the Cameron- Fowler mechanism, mass transfer in binary systems, and the engulfment of small bodies. These are listed in the following paragraphs.

  • (a)

    The Cameron-Fowler mechanism - Cameron (1955) and Cameron & Fowler (1971) proposed a mechanism to produce fresh Li in asymptotic giant branch (AGB) stars experiencing the hot bottom burning. 7Li is produced after the decay of 7Be. However, the temperatures needed to produce 7Be from α-capture on3 He are one order of magnitude higher than the temperature of Li burning. Therefore, new Li is immediately destroyed, but if 7Be is fast transported towards cooler regions, Li can survive. The Cameron-Fowler mechanism, originally pro- posed for intermediate-mass AGB stars (indicatively in the mass range 4–8 M⊙, see e.g. Sackmann & Boothroyd 1992; Ventura & D’Antona 2011), can work whenever a mechanism carries 7Be to cooler stellar regions. In the case of AGB stars, this process is driven by the convection, being the bottom of the convective envelope hot enough to produce 7Be. The existence of Li-rich stars at different metallicity, mass and evolutionary stage can be explained with the Cameron-Fowler mechanism only invoking extra mixing, i.e. thermohaline mixing (Charbonnel & Primas 2005), magneto-thermohaline (Denissenkov et al. 2009), mass- loss mechanisms occurring in RGB (de La Reza et al. 1996). An additional mechanism able to trigger the Cameron-Fowler mechanism is the ingestion of sub-stellar companions, like rocky planets or brown dwarfs causing an increase of angular momen- tum and a rotationally induced mixing (Denissenkov & Herwig 2004).

  • (b)

    Mass transfer processes – the accretion of matter from a companion with an enhancement of Li; for instance, any star able to trigger the Cameron-Fowler mechanism. Intermediate- mass AGB stars are the natural candidates, assuming that the low-mass Li-rich star belongs to a binary system with a massive companion (now evolved as a white dwarf). Possible additional signatures of this process could be an enhancement of those ele- ments produced in the interiors of AGB stars (i.e. carbon and slow neutron-capture elements) and radial-velocity variability.

  • (c)

    Planets engulfment – the engulfment of sub-stellar com- panions, such as rocky planets, hot Jupiters, and brown dwarfs, which have high Li enhancement (Siess & Livio 1999). This process should be coupled with infrared excess, possible strong magnetic fields, and X-ray activities.

In this paper, we discuss the metal-poor giant star HE 0057- 5959 (Gaia EDR3 4903905598859396480), whose A(Li) has previously been derived by Jacobson et al. (2015) and interpreted as a normal giant star caught during its ongoing FDU.

2 MIKE Observations and chemical analysis

2.1 Observations and radial velocity

The target star HE 0057-5959 was observed with the Magellan Ianmori Kyocera Echelle (MIKE) spectrograph (Bernstein et al. 2003) mounted at the Magellan II Telescope at Las Campanas Observatory, under the programme CN2017B-54 (PI: Monaco) during the night of 27 October 2017. We adopted a 0.7″×5.0″ slit corresponding to a spectral resolution of 53 000 and 42 000, in the blue and the red arm, respectively, and covering the spectral range between ~3400 and ~9400 Å. A total exposure time of three by 3000 s was adopted, providing a signal-to-noise ratio of 80 per pixel around the Li line at 6708 Å. The spectral reduc- tion, including bias-subtraction, flat-fielding, spectral extraction, and wavelength calibration was performed with the dedicated CarPy pipeline (Kelson 2003). The heliocentric radial velocity (RV) was measured through a cross-correlation against a syn- thetic spectrum as template, obtaining +376.2±0.2 km s−1, which is in agreement with the previous estimates by Norris et al. (2013, +375.3 km s−1), Jacobson et al. (2015, +376.7 km s−1), and Arentsen et al. (2019, +377.90+3.98 and +378.23+1.47 km s−1). No RV values are provided by Gaia.

2.2 Atmospheric parameters

The effective temperature (Teff) and the surface gravity (log −g) were estimated by the photometry in order to avoid sig- nificant biases affecting the spectroscopic determinations of these parameters in metal-poor giant stars (see Mucciarelli & Bonifacio 2020). We adopted magnitudes from the early third data release of the ESA/Gaia mission (Gaia Collaboration 2016, Gaia Collaboration 2020) and a colour excess of E(B–V)= 0.017 mag by Schlafly & Finkbeiner (2011). The extinction coefficients of the three Gaia bands were derived by adopting the iterative procedure described in Lombardo et al. (2021). Effective temperature was estimated using the (GBP – GRF)0-Teff transformation by Mucciarelli et al. (2021a), deriving Teff= 5420+80 K, where the uncertainty includes those arising from the photometric colour, the colour excess, and the adopted colour-Teff transformation. We checked that Teff derived from the other two Gaia colours, (GBP – G)0 and (G – GRP)0, are in excellent agreement with those of (GBP – GRP)0, Teff=5477, and 5426 K, respectively.

Stellar luminosity (log(L/L⊙) = -0.4 · (Mbol – M⊙)) was cal- culated adopting the photogeometric distance posterior from the Gaia parallax provided by Bailer-Jones et al. (2021) and the bolometric correction calculated from a new grid of syn- thetic fluxes (Mucciarelli et al. in prep.) computed with the code ATLAS9 (Kurucz 2005). The surface gravity (log −g= −10.32 + log(M) +4.Teff- log(L)) was obtained adopting the photometric Teff and the stellar luminosity described above and a stellar mass equal to 0.76 M⊙, which is a reasonable value for RGB stars with old ages and low metallicities according to the theoreti- cal isochrones of Pietrinferni et al. (2021). Ages in the range of ~ 11–13 Gyr, which is reasonable for a very metal-poor star such as HE 0057-5959, provide very similar stellar masses along the RGB that are lower than ~0.82 M⊙ with a negligible impact (less than 0.04 dex) on log −g. In the quoted log −g uncertainty we accounted for uncertainties in the Teff, luminosity, and adopted stellar mass.

Microturbulent velocity was estimated spectroscopically by minimising any trend between the abundances from Fe I lines and their reduced equivalent widths. The final parameters for HE 0057-5959 are Teff = 5420+80 K, log −g = 3.05+0.10, log(L/L⊙) = 1.18+0.14, and vt= 1.5+0.2 km s−1 (see Table 1). These values are in good agreement with those estimated by Jacobson et al. (2015): Teff = 5413 K, log −g = 3.41, and vt =1.4 km s−1. On the other hand, the temperature provided by Norris et al. (2013) is cooler (5257 K) and derived as the aver- age of temperatures from spectrophotometric flux and Balmer lines. For the chemical analysis of this star, Yong et al. (2013) adopted the Teff of Norris et al. (2013), while a new value of log −g (2.65 dex) was derived based on the stellar temperature and a theoretical isochrone with an appropriate metallicity and age of 10 Gyr. Their lower log -g reflects mainly the difference in Teff between our and their analysis.

Table 1

Stellar parameters for target star HE 0057-5959.

2.3 Chemical analysis

Abundances of Mg, Al, Si, Ca, Ti, and Fe were obtained from measured equivalent widths (EWs) using the code GALA (Mucciarelli et al. 2013a), while for atomic transitions affected by blending, isotopic and/or hyperfine splittings (Li, Na, Sr, and Ba) and molecular features (G band for C), the abundances were derived through an χ2-minimisation between the observed lines and grids of synthetic spectra calculated with SYNTHE (Kurucz 2005). For other species, spectral lines are too weak and undetectable because of the very low metallicity of the tar- get star, coupled with its relatively high Teff , also preventing useful upper limits. We also checked the possibility of measur- ing N and 12C/13C, but we cannot identify 13C transitions or CN molecular bands at 3870-3885 Å. Despite the importance of N and 12C/13C, the derived upper limits are not meaningful ([N/Fe]<+3.5 dex, 12C/13C>5) and they do not provide useful insights for the interpretation of the target star. In this analysis, we adopted an ATLAS9 model atmosphere (Kurucz 1993, 2005) calculated with a new opacity distribution function (Mucciarelli et al. in prep.) with [Fe/H]=−4.0 dex and [α/Fe]=+0.4 dex.

Abundance ratios scaled on the solar abundances by Lodders (2010) and Caffau et al. (2011) are listed in Table 2, together with the number of measured lines and the total uncertainty. The latter was computed by adding the internal error and that arising from the adopted atmospheric parameters in quadrature (uncertainties in the derived parameters are discussed in Sect. 2.2). The internal error is estimated as the standard error of the mean (σ/(Nlines )$\sigma /\sqrt {\left( {{{\rm{N}}_{{\rm{lines }}}}} \right)}$) when at least two lines are measured. For elements with one only transition, we consider the uncertainty on the abundance as obtained from the EW error or from Monte Carlo simula- tions; this is for lines measured from EWs and spectral synthesis, respectively (see details in Mucciarelli et al. 2013b). In Table 2 we report both LTE and NLTE abundances, in the latter case adopting the corrections of Wang et al. (2021) for Li, Lind et al. (2011) for Na, Nordlander & Lind (2017) for Al, Bergemann et al. (2013) for Si, and Mashonkina et al. (2023) for the other ele- ments. The atomic data of the measured atomic lines are listed in Appendix B.

We derived [Fe/H]=−3.95+0.11 dex. Departures from LTE increase the Fe abundance of this star by ~+0.22 dex (Mashonkina et al. 2023). In the following discussion, we refer to the LTE [Fe/H] for consistency with the [Fe/H] values derived for all the other Li-rich metal-poor stars known to date, but the use of NLTE [Fe/H] does not change our conclusion on the nature of HE 0057-5959. The star is enhanced in [α/Fe] (Mg, Si, Ca, and Ti) and characterised by sub-solar [Sr/Fe] and [Ba/Fe]. Also, the star is enhanced in [C/Fe] (+1.07 dex), indicating that it can be labelled as CEMP-no (enhanced in C but not in s-process elements), as already suggested by Jacobson et al. (2015). The comparison with the abundances measured by Yong et al. (2013) and Jacobson et al. (2015) is discussed in Appendix C.

The Li abundance was derived from the resonance line at 6708 Å, while the subordinated Li line at 6103 Å is not detected. The 3D-NLTE correction by Wang et al. (2021) was applied to the Li abundance, leading to a final abundance of A(L1)nlte=+2.09+0.07, which is consistent with the value quoted by Jacobson et al. (2015): A(Li)NLTE=+1.97 dex. The NLTE correction for this star is of +0.04 dex.

The sodium abundance was derived from the Na D lines at 5890 and 5896 Å, which are the only Na transitions visible in the spectrum because of the very low metallicity of the star (Yong et al. 2013 and Jacobson et al. 2015 also derived the Na abundance of HE 0057-5959 measuring these two lines only). We checked that the Na D lines are not contaminated by Na inter- stellar lines because of the high RV of the star (see Fig. D.1). The NLTE corrections for Na abundance by Lind et al. (2011) were applied, leading to [Na/Fe]=+1.92+0.08 dex, in agreement with the values listed by Yong et al. (2013) and Jacobson et al. (2015). The Na doublet at 5682-88 Å is too weak to be observed, and we can only derive an upper limit of [Na/Fe]<2.1 dex. Figure 1 shows the best-fit synthetic spectra obtained for Li and Na lines as examples.

Table 2

LTE and NLTE chemical abundances measured in HE 0057-5959, together with the number of used lines and the total uncertainty.

3 The evolutionary stage of HE0057-5959

Jacobson et al. (2015) derived A(Li)NLTE for a sample of 24 metal-poor giant stars, including HE 0057-5959. All these stars are fainter than that of the RGBb (with a Teff between ~4700 and ~5400 K), and therefore they are not affected by the extra- mixing episode associated with the RGBb. The sample stars have a very similar A(Li)NLTE of around +1.0 dex, which matches the values observed by Mucciarelli et al. (2022) for stars in the same evolutionary stage well. In the Jacobson et al. (2015) sam- ple, HE 0057-5959 is a clear outlier, with a significantly higher Li abundance (A(Li)NLTE=+1.97 dex), despite it having only a slightly higher Teff and lower log(L/L⊙) than the other sample stars. Jacobson et al. (2015) quoted Teff = 5413 K for this star, while the hottest star in the remaining sample has Teff = 5260 K and A(Li)nlte = 1.21 dex. They do not recognise HE 0057-5959 as an Li-rich star, claiming that this star has an A(Li) depletion level appropriate for its Teff. In other words, they interpret this star as an Li-normal star with a high A(Li)NLTE due to an ongoing FDU, the only mechanism able to justify the measured A(Li)NLTE without invoking an anomalous enhancement of Li.

Stellar parameters and A(Li)NLTE from our new analysis agree with those by Norris et al. (2013) and Jacobson et al. (2015), and we confirm that HE 0057-5959 belongs to the first ascent of the RGB (see left panel of Fig. 2). The luminosity of the target is fainter than the RGBb (occurring at log(L/L⊙)~2.35 for the metallicity of the star; see left panel of Fig. 2.), con- firming that HE 0057-5959 is not affected by the extra-mixing episode associated with the RGBb. In order to establish whether its A(Li)NLTE is compatible with an ongoing FDU, we compared the A(Li)nlte of HE 0057-5959 with that predicted by a theoreti- cal model (calculated with the same code and input physics as in Pietrinferni et al. 2021) of a star with 0.76 M⊙ and Z=3.3×10−6 (corresponding to (Fe/H]=−4.0 dex and [α/Fe]=+0.4 dex) and without atomic diffusion.

As is clearly visible in the left panel of Fig. 2, at this metallicity the drop of A(Li) due to the FDU occurs at the end of the sub-giant branch. It is confined to a narrow region in Teff of 5500–5800 K and log(L/L⊙), 0.8–1.0. The stellar parameters of HE 0057-5959, within their uncertainties, are not compatible with the position of the FDU, and they indicate that the star has already finished the FDU. The same conclu- sion is reached whether or not the stellar parameters derived by Norris et al. (2013) and Jacobson et al. (2015) are adopted (see Sect. 2.2). If we assume that this star before the FDU had an A(Li)NLTE similar to that of the Spite Plateau stars (~2.2–2.3 dex), the measured abundance (A(Li)NLTE =+2.09 dex) should be observed when the star starts its FDU, at a Teff hotter by ~300–400 K and a log(L/L⊙) fainter by 0.4 dex with respect to those of HE 0057-5959. Therefore, the measured A(Li)NLTE of HE 0057-5959 is not compatible with what we expect for a star in that evolutionary stage starting from the Spite Plateau. Also, A(Li)NLTE of HE 0057-5959 is clearly incompatible with the abundances of lower RGB stars that share a very similar A(Li) close to +1.0 dex (see right panel of Fig. 2). Hence, we revise our conclusion on the nature of HE 0057-5959 and claim that it is a genuine Li-rich star with a Li abundance significantly higher than those measured in stars at the same evolutionary stage and not compatible with an ongoing FDU. This is the most metal-poor, Li-rich star discovered to date, followed by the giant star LAMOST J070542.30+255226.6, which has [Fe/H]=−3.12 dex (Li et al. 2018).

thumbnail Fig. 1

Spectral regions of MIKE spectrum (blue points) around the Li resonance line and a Na D line (upper and lower panel, respectively), with superimposed synthetic spectra calculated with the best-fit abun- dance (central red curve) and ±0.1 dex from the best-fit abundance.

4 Metal-poor, Li-rich stars in the literature

In order to properly compare the properties of HE 0057-5959 with those of other Li-rich stars, we collected a database of all the metal-poor ([Fe/H]<−1.0 dex) Li-rich stars discovered in our Galaxy to date, making an effort to homogenise their stellar parameters and A(Li)NLTE. Adopting for each star the literature stellar parameters and LTE A(Li) derived from the 6708 Å line, we calculated the corresponding synthetic profile and its equivalent width by integration. After that, we calcu- lated the curve of growth for the Li line assuming the new stellar parameters, and we derived the new A(Li) based on the previ- ous equivalent width. The average difference between the LTE A(Li) values obtained with the new and literature parameters are +0.01±0.03 dex (σ=0.18 dex), with only three stars having an absolute discrepancy larger than 0.3 dex. Excluding these three stars, the average difference is +0.04±0.02 dex (σ=0.12 dex). For all the stars, we applied the 3D-NLTE corrections of Wang et al. (2021). Tables E.1 and E.2 list the new Teff, log −g, log(L/L⊙), A(Li)NLTE, and the literature values of [Fe/H] and [Na/Fe] for all the targets. The positions of all the metal-poor, Li-rich stars in the Hertzsprung-Russell diagram are shown in Fig. 3, where we highlight the mean loci of the FDU and RGBb. We also show two theoretical isochrones with ages of 13 Gyr and [Fe/H]=−3.2 and −1.2 dex (Pietrinferni et al. 2021) as a reference to identify the evolutionary stage of the Li-rich stars.

4.1 Galactic-field Li-rich stars

Aside from HE 0057-5959, 34 metal-poor, Li-rich stars (all of them with A(Li)NLTE higher than the average A(Li) measured in stars at the same evolutionary stage) have been discovered in the Milky Way field, which covers a metallicity range between [Fe/H]~−3.1 and ~−1.4 dex. All these stars were recovered in the Gaia eDR3 catalogue, and their parameters (Teff, log −g and log(L/L⊙)) were derived using the approach described in Sect. 2. We adopted the colour excess from the latest 3D dust extinction maps by Lallement et al. (2022) for the closest targets, and val- ues from Schlafly & Finkbeiner (2011) for targets whose distance is outside the maps of Lallement et al. (2022). The uncertainties in Teff are dominated by the precision of the adopted colour-Teff transformation (~80 K; see Mucciarelli et al. 2021a), with a neg- ligible contribution arising from photometric error and redden- ing. The uncertainties in log(L/L⊙) are under 0.1 dex and dom- inated by the distance error. Only for four stars (namely SDSS J093627.44+293535.7, 2MASS J10122548-2030068, 2MASS J16070923+0447126, and SDSS J143207.14+081406.1) errors in log(L/L⊙) are higher than 0.2 dex, reflecting their large uncer- tainty in the Gaia parallaxes. For seven stars for which log −g is under 1.3 (the boundary of the grid by Wang et al. 2021), we adopted log −g=1.3 to calculate the 3D-NLTE correction.

The new parameters are in good agreement with the litera- ture ones, with average differences (this study - literature) of + 19±23 K (σ=137 K) for Teff and +0.02±0.03 (σ=0.20) for log −g. The largest differences are in Teff for the stars LAM-OST J055408.54+523559.0, LAMOST J075816.39+470343.3 (Li et al. 2018), 2MASS J05241392-0336543 (Kowkabany et al. 2022), and UCAC4 212-183136 (Susmitha et al. 2024), with differences of −330, −254, −300 K, and +228 K, respectively.

thumbnail Fig. 2

The evolutionary stage and the surface A(Li) of HE 0057-5959 . Left panel: position of target star HE 0057-5959 in the Hertzsprung-Russell diagram (red square) in comparison with the theoretical track of a star with 0.76 M⊙, [Fe/H]=−1.0 dex and [α/Fe]=+0.4 dex, coloured according to the predicted A(Li). The colour-scale is shown on the right side. The model is re-scaled in order to match the average A(Li)nlte of the RGB stars by Mucciarelli et al. (2022). Right panel: A(Li)NLTE as function of log(L/L⊙) for HE 0057-5959 (red square) in comparison with the RGB star sample by Mucciarelli et al. (2022, cyan circles) and the stars of the metal-poor ([Fe/H]~−2.0 dex) globular cluster NGC 6397 (Lind et al. 2009, grey circles). The luminosity level of the RGBb for the metallicity of HE 0057-5959 is marked with an arrow.
thumbnail Fig. 3

Position in Hertzsprung-Russell diagram of the target star HE 0057-5959 (red square) and of the other metal-poor ([Fe/H]<−1 dex) Li-rich stars discovered so far (see details in Sect. 4): green circles are Galactic-field stars and blue circles are globular-cluster stars. Two theoretical isochrones with [Fe/H]=−3.2 and −1.2 dex, an α-enhanced chemical mixture, and an age of 13 Gyr (Pietrinferni et al. 2021) are shown for reference (black curves). The two grey shaded areas indicate the loci where the drops of A(Li)nlte due to the FDU and further extra-mixing episodes, respectively, occur.

4.2 Galactic globular clusters’ Li-rich stars

Sixteen Li-rich stars have been discovered so far in Galactic GCs. We recovered 15 of them in the Gaia catalogue, exclud- ing the cepheid star V42 in M5 (Carney et al. 1998) because its variability leads to large uncertainties in the stellar parameters. Three Li-rich stars have a double identification: Stet-M68-S232 in M68 (Ruchti et al. 2011; Kirby et al. 2016), M3-IV 101 in M3 (Kraft et al. 1999; Ruchti et al. 2011), and 132 in M30 (Kirby et al. 2016; Gruyters et al. 2016). Their stellar parameters were derived adopting [Fe/H], and the distance and E(B-V) of Harris (1996, 2010 edition), except for the two Li-rich stars in ω Centauri (Mucciarelli et al. 2019) for which we adopted their proper [Fe/H] values. For the Li-rich stars in NGC 1261 and NGC 6397, we assumed A(Li)=4.0 dex to avoid extrapolation in the 3D-NLTE grids. For the other three stars (namely M3-IV in M3, Stet-M68-S232 in M68, and V2 in NGC 362) we assumed log −g= 1.3 as explained above for the field Li-rich stars.

The Gaia Teff are in good agreement with the values quoted in the literature, with an average difference of +11±30 K (σ=116 K) and no relevant outliers aside from the cold star V2 in NGC 362 (Smith et al. 1999), which we found to be 250 K cooler. Concerning log −g, the average difference is +0.06±0.04 (σ=0.17); most of the targets have differences in log −g within ± 0.1, with three main exceptions. These are V2 in NGC 362, which has a log −g that is lower than the Smith et al. (1999) value by 0.4 dex; and 97 812 in NGC 3201 (Aguilera-Gómez et al. 2022) and M3-IV in M3 (Ruchti et al. 2011), which have log −g values that are higher than those in the literature by 0.4 dex.

5 Discussion

5.1 The Li-rich star HE 0057-5959

The star HE 0057-5959 discussed in this work is the most metal- poor, Li-rich star identified to date. We investigated the origin of its Li enhancement. We considered the main mechanisms previously proposed to explain Li-rich stars.

Concerning the scenario of internal Li production, the main issue for this star is the lack of an internal mixing process able to activate the Cameron-Fowler mechanism. Due to the low metallicity of the star, its shallow convective envelope cannot reach the layers where 7Be is produced through α captures on 3He, and this region is not in contact with the surface, preventing any internal mixing. Possible extra mixing mechanisms can occur in low- mass stars, but this is at the RGBb (Charbonnel & Balachandran 2000; Palacios et al. 2001) or close to the He flash (Silva Aguirre et al. 2014), while this star is located at the base of the RGB.

Mass-transfer processes from a more massive, now evolved, companion star able to produce fresh Li via the Cameron-Fowler mechanism is a valuable and simple route to explain Li-rich stars. The measured RV from our MIKE spectrum nicely matches the previous estimates indicating no large RV variations. The RUWE value of the star is 0.98, indicating a well-behaved single- star astrometric solution. However, wide binary systems may have periods that are too long to be identified with the Gaia observations. We cannot totally rule out the star belonging to a long-period or highly inclined binary. In this case, the enhance- ment of Li could be explained as the result of a mass-transfer process from a more massive companion star during its AGB stage. We note that none of the Li-rich stars for which multi- ple epochs are available has shown evidence of variability, with the only exception being 25 664 in ω Centauri (Mucciarelli et al. 2021b).

The Li overabundance in a giant star can be the result of engulfment of a planet as the star evolves on the RGB, thus increasing its radius. This seems to be the most likely expla- nation for the solar metallicity of the Li-rich giant BD+48 740 (Adamow et al. 2012). However, the extremely low metallicity of HE 0057-5959 makes it unlikely for it to host, or have hosted, planets. In spite of early claims that the frequency of planets around giant stars is not correlated with metallicity (Pasquini et al. 2007), subsequent investigations found that the higher the metallicity, the higher the probability of hosting a planet (Wolthoff et al. 2022, and references therein), similar to what was found for dwarf stars, see Adibekyan (2019) and references therein. To our knowledge, the two planet-host stars with the lowest metallicity are BD+20 2457 (Maldonado et al. 2013) and 24 Boo (Takarada et al. 2018), both of which have metallicities around −0.8 dex. Clearly, if we were to extrapolate this metallic- ity dependence of hosting planets down to −4.0 dex, we would find a very low number (practically zero). One should keep in mind, however, that the stars with metallicities below −3.0 dex are not represented in any of the planet search surveys; thus, such an extrapolation cannot be supported by any data. We neverthe- less believe that, based on our current knowledge, the possibility of the high Li abundance in HE 0057-5959 being due to planet engulfment can be discarded.

In conclusion, the mass-transfer scenario remains the most promising one, lacking a well established mechanism capable of inducing a Cameron-Fowler mechanism in a low-mass, very metal-poor stars such as HE 0057-5959. Whatever mechanism is capable of generating an Li-rich star, that mechanism must also occur down to [Fe/H]~−4 dex.

5.2 An overview of the metal-poor, Li-rich stars

We discuss the properties of HE 0057-5959 in comparison with the other metal-poor ([Fe/H]<−1 dex) Li-rich stars (see Sect. 4).

The vast majority of the low-mass dwarf stars in the metal- licity range between [Fe/H]~−3 and ~−1 dex share a similar abundance: A(Li)NLTE ~2.2–2.3 dex. At higher metallicities, two effects occur: destruction of the plateau and the significant increase of the star-to-star scatter. The first one is the pres- ence of more massive convective envelopes in these stars, which leads to more efficient surface A(Li) depletion (Meléndez et al. 2014). The second effect is the occurrence of novae that pro- duce fresh Li and contribute to the chemical enrichment of the Galaxy for [Fe/H]>−1 dex (see e.g. Izzo et al. 2015; Romano et al. 2021; Izzo et al. 2023). In fact, following the detection in RS Oph (Molaro et al. 2023), we know that recurrent novae also contribute to the Li production. In this way, the interpretation of metal-poor, Li-rich stars has the advantage of removing the effects of the extra dilution due to the massive convective enve- lope and the effect of novae producing additional Li from the discussion.

In principle, the A(Li) of an Li-rich star (regardless of its ori- gin) should follow the same evolutionary path of an Li-normal star, with a significant reduction at the FDU and the RGBb. For this reason, Li-rich stars should be discussed considering their evolutionary stage and the possible occurrence of the mixing episodes.

The upper left panel of Fig. 4 shows the run of A(Li)NLTE as a function of log(L/L⊙) for HE 0057-5959 and all the other metal-poor, Li-rich stars discovered to date. Despite a significant star-to-star scatter, it is possible to recognise some sequences where A(Li) decreases with increasing log(L/L⊙). Some of the field Li-rich stars draw a clear sequence starting from A(Li)~3.2 dex at log(L/L⊙) ~+0.5 down to A(Li)~2.2 dex at log(L/L⊙) ~+2. The GC Li-rich stars in the same luminosity range seem to draw a parallel sequence, but it is shifted by 0.4 dex towards lower A(Li). Finally, some field and GC stars define a super Li- rich sequence, with values higher than the other Li-rich stars of similar log(L/L⊙) by 1 dex. Two sequences of A(Li) as a function of log(L/L⊙) for the GC Li-rich stars have already been proposed by Sanna et al. (2020). We note that the Li-rich stars show a decrease of A(Li) that is less steep than that expected by the FDU.

The other panels of Fig. 4 show the behaviour of A(Li)NLTE as a function of [Fe/H] for the stars grouped according to their evolutionary stages. In particular, we consider stars located before the FDU (corresponding to the Spite Plateau for Li-normal stars), after the FDU and before the RGBb (corre- sponding to the lower RGB plateau for Li-normal stars), and after the RGBb (stars experienced the extra mixing episode at the luminosity level of the RGBb). Only one Li-rich star, namely 132 in the GC M30, is clearly located during the FDU (see Fig. 2 and Table E.2) and excluded from this discussion. Also, two field stars (namely GSC 03797-00204 and 2MASS J19524490-6008132) are close the RGBb, and we propose an attempt at classification (they are marked as empty symbols in Fig. 4; see Table E.1). Finally, for two stars (namely SDSS J143207.14+081406.1 and 2MASS J04315411-0632100) the attri- bution is too uncertain due to their anomalous position in the Hertzsprung-Russell diagram (see Fig. 3).

Seven Li-rich stars (two of them members of GCs) are located before the occurrence of the FDU. Six of them have A(Li)NLTE exceeding the primordial value obtained from the standard Big Bang nucleosynthesis model and the Planck/WMAP measurements of the baryon density. Among them, four stars have A(Li)nLte~3–3.2 dex (about 1 dex higher than the Spite Plateau), and the other two stars have A(Li)nLte~4 dex (about 2 dex higher than the Spite Plateau).

Only the Li-rich star in M 4 (Monaco et al. 2012) has a value compatible with the primordial value.

Twenty-one stars are located between the completion of the FDU and before the RGBb. They show a large A(Li) scatter and a typical value around +2.4 dex, which is 1.4 dex higher than the abundances measured along the lower RGB plateau. However, their average value is lower than that measured in the previous group, suggesting that a dilution due to the FDU occurred.

Eighteen stars (4 of them members of GCs) are located after the RGBb. In this group, we found the Li-rich stars with the low- est (Smith et al. 1999) and the highest (Kowkabany et al. 2022) A(Li)NLTE of the entire sample. The average value of these stars matches that of the stars between the FDU and the RGBb well.

For the stars between FDU and the RGBb, we corrected the measured A(Li)NLTE to take into account the effect of the Li dilution due to the FDU, following the method described in Mucciarelli et al. (2012). In particular, for each star in this evolu- tionary stage, we consider the stellar model with the appropriate metallicity that provides the amount of A(Li) dilution as a func- tion of the stellar luminosity. The A(Li) dilution to be added to the measured A(Li)NLTE value is computed according to the log(L/L⊙) value of each star. This approach is restricted to stars experiencing the FDU, but fainter than the RGBb, because the standard stellar-evolution models that we adopted do not account for non-canonical mixing processes.

Figure 5 shows the run of the corrected A(Li) with [Fe/H] for these stars, together with the (uncorrected) abundances for stars before the FDU. No evident trend between the initial A(Li) and [Fe/H] is found. The derived distribution exhibits a large star-to-star scatter, and almost all the stars (assuming that they formed with a high A(Li)) have abundances higher than the cos- mological value. For the target star, HE 0057-5959, the predicted initial A(Li) should be +3.05 dex, which is ~0.3 dex higher than the cosmological value. Only a few stars could have an ini- tial A(Li) compatible with the cosmological value (which could be explained by invoking some preservation of the pristine Li), while for most of the Li-rich stars, processes able to produce or enhance the surface A(Li) should occur.

thumbnail Fig. 4

Surface A(Li) of the Li-rich stars discov- ered so far. Upper left panel: behaviour of A(Li)NLTE as function of log(L/L⊙) for all the metal-poor ([Fe/H]<−1 dex) Li-rich stars discovered so far. Green circles are Milky Way field stars, blue cir- cles are globular-cluster stars, and the red squares mark HE 0057-5959. The stars of the metal-poor ([Fe/H]~−2.0 dex) globular cluster NGC 6397 (grey circles, Lind et al. 2009) are shown as a reference (and to identify the luminosity of the FDU and RGBb). Upper right panel: run of A(Li)NLTE as func- tion of [Fe/H] for the Li-rich stars located before the FDU. Red dashed line indicates the WMAP/SBNN A(Li) (Coc & Vangioni 2017). The cyan circles are Li-normal Milky Way field stars (Ryan et al. 1999; Lucatello et al. 2003; Ivans et al. 2005; Asplund et al. 2006; Sivarani et al. 2006; Thompson et al. 2008; Sbordone et al. 2010; Masseron et al. 2012; Ito et al. 2013; Hansen et al. 2014; Bonifacio et al. 2015; Li et al. 2015; Placco et al. 2016; Matsuno et al. 2017; Aguado et al. 2018; Bonifacio et al. 2018). Lower-left panel: run of A(Li)NLTE as function of [Fe/H] for the Li-rich stars located after the com- pletion of FDU and before the RGBb. The open cir- cle indicates the star 2MASS J19524490-6008132 with an uncertain attribution to this group. The cyan circles are Li-normal Milky Way field stars (Mucciarelli et al. 2022). Lower right panel: run of A(Li)NLTE as function of [Fe/H] for the Li-rich stars located after the RGBb. Open circle indicates the star GSC 03797-00204 with an uncertain attribution to this group.
thumbnail Fig. 5

Behaviour of initial A(Li), accounting for the dilution effect due to the FDU, for HE 0057-5959 and other metal-poor, Li-rich stars fainter than the RGBb (same symbols as Fig. 3).

5.3 HE 0057-5959: An Na-rich Li-rich metal-poor star

Previous works on metal-poor, Li-rich stars highlighted signif- icant over-abundances of [Na/Fe] in some of them (see e.g. Kowkabany et al. 2022; Sitnova et al. 2023) but this chemical signature has not been properly discussed. Na abundances are available for 23 field stars, as shown in Fig. 6 in comparison with the metal-poor Milky Way field stars (Andrievsky et al. 2007; Lombardo et al. 2022). We also show the six GC Li-rich stars with [Na/Fe] abundances even if the discussion of these stars is complicated by the self-enrichment processes occurring in the early epochs of the GC life, and these stars are likely able to form new stars with excesses of [Na/Fe] (see e.g. Bastian & Lardo 2018). In the following, we only refer to the field Li-rich stars.

The [Na/Fe] distribution of the metal-poor, Li-rich stars does not match that of the Milky Way field stars. For [Fe/H]<− 2.0 dex, the field stars have a constant value of [Na/Fe]~− 0.2 dex, while the Li-rich stars exhibit a significant star- to-star scatter in [Na/Fe], reaching very high values up to [Na/Fe]~+1.6 dex. In particular, the three Li-rich stars with [Fe/H]<−3.0 dex have [Na/Fe]>+1.3 dex and are the most Na-rich Li-rich stars: HE0057-5959, with [Na/Fe]=+1.37 dex; Gaia EDR3 883042050539140992, with [Na/Fe]=+1.37 dex (Li et al. 2018); and Gaia EDR3 2604066644687553792, with [Na/Fe]=+1.58 dex (Roederer et al. 2014). In this comparison and in Fig. 6 we used [Na/Fe]=+1.37 for HE 0057-5959, which was obtained by adopting NLTE Na abundance and LTE Fe abun- dance, similarly to the analyses of other Li-rich stars, where only the Na abundances are corrected for NLTE effects.

Even though only three Li-rich stars with [Fe/H]<−3 dex have been discovered to date, their extremely high [Na/Fe] abun- dance ratios could be a new characteristic feature of this class of rare objects thus far unexplored. Sodium is produced in both massive stars during the hydrostatic C and Ne burning and in AGB stars during the hot bottom-burning phase. In particular, super-AGB stars with initial masses larger than ~6–7 M⊙ should be able to produce large amounts of both Li and Na (at least for [Fe/H]>−2.5 dex; e.g. Ventura & D’Antona 2011; D’Antona et al. 2012; Doherty et al. 2014). In these stars, Li is produced through the Cameron-Fowler mechanism, and Na is produced through the Ne-Na cycle. The evidence that all three Li-rich stars with [Fe/H]<−3 dex have an excess of [Na/Fe] could be an important hint supporting the scenario of a mass-transfer process occurring in binary systems where the companion was a massive star able to produce Li and Na simultaneously. Unfortunately, theoretical models for AGB stars at [Fe/H]=−4 dex are not yet available.

thumbnail Fig. 6

Behaviour of [Na/Fe] as function of [Fe/H] of the metal- poor, Li-rich stars (same symbols of Fig. 3) in comparison with the NLTE abundances for metal-poor Milky Way stars (grey circles) from Andrievsky et al. (2007) and Lombardo et al. (2022). The [Na/Fe] abun- dance ratio of HE 0057-5959 plotted here accounts for NLTE effects for Na abundance only. The position of the Li-rich star by Kowkabany et al. (2022), the most Li-rich discovered so far, is marked.

6 Conclusions

We revised the nature of the metal-poor star HE 0057-5959, demonstrating that it is a genuine Li-rich star and that it belongs to the limited class of the metal-poor, Li-rich stars. Its very low metallicity demonstrates that at least one of the proposed mechanisms able to produce Li-rich stars should work down to [Fe/H]~−4 dex. However, we are not yet able to identify the process capable of producing the excess Li observed in this star and generally in other metal-poor stars with absolute cer- tainty. This is due to our still only partial understanding of the properties of these stars and to the lack of a sound mechanism capable of triggering the Cameron-Fowler mechanism in many of these stars (especially those fainter than the RGBb). For the star HE 0057-5959, we can only speculate that the excess of Li may be attributable to a mass-transfer process in a binary system, despite not having strong evidence of this, except per- haps for the very high [Na/Fe] supporting this hypothesis. The extremely high Li and Na abundances could be compatible with a mass transfer from a companion in the stellar range of 6–8 M⊙, while the lack of RV variations from the three available epochs and the RUWE Gaia parameter close to the unity do not sup- port the binary nature of the star (but neither do they rule out a long-period binary system). On the other hand, HE 0057-5959 does not exhibit any enhancement of neutron-capture elements (Sr and Ba) usually associated with the mass-transfer process from AGB stars. However, it is important to bear in mind that AGB stars of 2–4 M⊙ produce a large amount of neutron-capture elements, while stars with 6–8 M⊙ are much less efficient in pro- ducing these elements (see e.g. Fishlock et al. 2014; Shingles et al. 2015). Therefore, all the chemical evidence collected so far for HE 0057-5959 is in agreement with this scenario.

Detailed chemical abundances of the main groups of ele- ments are limited to a few stars; in particular, elements that are tracers of mass transfer from AGB stars (i.e. CNO, 12C/13C, and neutron-capture processes) are simultaneously available for only nine Li-rich stars. Another missing piece of evidence includes dedicated surveys of RV in order to monitor possible variabil- ity and establish the binary nature of some of these stars. In the same way, systematic studies of other diagnostics of interac- tions (chromospheric activity, stellar rotation…) are still lacking. Metal-poor, Li-rich stars therefore belong to an unexplored field of research that deserves deeper investigation.

Acknowledgements

We thanks the anonymous referee for the useful and con- structive suggestions. A.M. acknowledges support from the project “LEGO - Reconstructing the building blocks of the Galaxy by chemical tagging” (PI: A. Mucciarelli). granted by the Italian MUR through contract PRIN 2022LLP8TK_001. M.M. acknowledges support from the ERC Consolidator Grant funding scheme (project ASTEROCHRONOMETRY, https://www.asterochronometry.eu, G.A. n. 772293). This work has made use of data from the European Space Agency (ESA) mission Gaia (https://www.cosmos.esa.int/gaia), processed by the Gaia Data Processing and Anal- ysis Consortium (DPAC, https://www.cosmos.esa.int/web/gaia/dpac/consortium). Funding for the DPAC has been provided by national institutions, in particular the institutions participating in the Gaia Multilateral Agreement. This research has made use of the SIMBAD database, operated at CDS, Strasbourg, France.

Appendix A Asteroseismic data

Here we discuss asteroseismic observations of metal-poor Li- rich stars (Table E.1). Our sample of stars has been observed by TESS (Transiting Exoplanet Survey Satellite; Ricker et al. 2015), but only 16 stars have available light curves in the Mikulski archive for space telescopes (MAST1). We apply a Lomb-Scargle transform (Lomb 1976; Scargle 1982) to these lightcurves by means of the Python package LIGHTKURVE V2.4.12. We do not find evidence of solar-like oscillations in any of these stars even with lightcurves based on the MIT quick- look pipeline (QLP; Huang et al. 2020), or the TESS data for asteroseismology light curves pipeline (TASOC; Handberg et al. 2021; Lund et al. 2021). Fig. A.1 shows as example of the power spectral density for HE 0057-5959 based on TESS obser- vations. Indeed, we could not find any of these stars in catalogues of solar-like oscillators observed with TESS (Hon et al. 2021; Mackereth et al. 2021; Hatt et al. 2023), and a similar result is shown in Kowkabany et al. (2022) for the star Gaia DR3 3210839729979320064. Finally, these results suggest that more than 130 days (corresponding to longest observations made in our sample, that is those for Gaia DR3 3360259919923274496) are needed to observe solar-like oscillations in such stars.

thumbnail Fig. A.1

Power spectral density for HE 0057-5959 based on TESS observations during Sector 1 and 2. The light curve from which we obtain the power spectral density is provided by the TESS Asteroseis- mic Science Operations Center (TASOC; Handberg et al. 2021; Lund et al. 2021). We find no evidence of solar-like oscillations.

Appendix B Information about the measured atomic lines

Table B.1 lists the main atomic data (wavelength, oscillator strength, excitation potential and ion) for all the measured transitions used in the analysis.

Table B.1

Main atomic data for the measured transitions used in the analysis.

Appendix C Comparison with previous chemical analyses

The chemical composition of HE0057-5959 has been already investigated by Yong et al. (2013) and Jacobson et al. (2015) both analysing MIKE spectra. Fig. C.1 shows the comparison between our analysis and those by Yong et al. (2013) and Jacobson et al. (2015) for the elements in common among the three studies. We consider our LTE abundances but for Li and Na because these two studies provided NLTE abundances only for these two elements. Our analysis well agrees with the previous ones, with differences that do not exceed +0.2 dex. The largest differences are for the [C/Fe] and [Ca/Fe] by Yong et al. (2013), 0.21 dex lower and higher than our values, respectively. The dif- ferences with respect to these two studies can be ascribable to several differences in the chemical analyses, in particular, both the works adopted solar abundances by Asplund et al. (2009), Jacobson et al. (2015) used MARCS model atmospheres, while

Yong et al. (2013) adopted ATLAS9 model atmospheres like our analysis, and Yong et al. (2013) adopted a Teff value ~200 Κ cooler than our one (see Sect. 2.2).

thumbnail Fig. C.1

Difference of our abundances with respect to those by Yong et al. (2013) and Jacobson et al. (2015), red and blue points respectively, as a function of the atomic number. We consider our LTE abundances but for Li and Na that are provided by Yong et al. (2013) and Jacobson et al. (2015) corrected for NLTE effects.

Appendix D Contamination of the Na D lines by interstellar features

The photospheric Na D lines at 5889.9 and 5895.9 Å can be contaminated by the same transitions arising from the interstel- lar medium along the line of sight. We checked that the Na D lines in the spectrum of HE 0057-5959 are not contaminated by interstellar features (see Fig. D.1) because of the large RV of the star. Also the interstellar lines are very weak because of the low colour excess of HE 0057-5959.

thumbnail Fig. D.1

Spectral region of the MIKE spectrum of HE 0057-5959 with marked the photospheric and interstellar Na D lines (blue and red arrows, respectively).

Appendix E Information about the metal-poor Li-rich stars

Tables E.1 and E.2 summarise the main information about the metal-poor Li-rich stars discovered so far in the Milky Way field and in Galactic globular clusters.

Table E.1

Main information about the Li-rich stars discovered so far in the Milky Way field.

The stars are in order of the Gaia EDR3 identification number (we report also the alternative identification number or name used in the literature). Stellar parameters and A(Li)NLTE are those described in Sect. 2. [Fe/H] and [Na/Fe] are those listed in the corresponding papers. The evolutionary stage of each target is obtained as described in Sect. 4: (1) pre-FDU, (2) FDU, (3) post-FDU /pre-RGBb, (4) RGBb. References: (1) Li et al. (2018), (2) Martell & Shetrone (2013), (3) Sitnova et al. (2023), (4) Roederer et al. (2014), (5) Ruchti et al. (2011), (6) Kowkabany et al. (2022), (7) Roederer et al. (2008), (8) this study, (9) Susmitha et al. (2024).

Table E.2

Main information about the Li-rich stars discovered so far in Galactic globular clusters. The clusters are in order of right ascension. We list for each target the Gaia ED3 identification number and the alternative identification number or name used in the literature. Stellar parameters and A(Li)nlte are those described in Sect. 2. [Fe/H] and [Na/Fe] are those listed in the corresponding papers. The evolutionary stage of each target is obtained as described in Sect. 4: (1) pre-FDU, (2) FDU, (3) post-FDU / pre-RGBb, (4) RGBb. References: (1) Smith et al. (1999), (2) D’Orazi et al. (2015), (3) Sanna et al. (2020), (4) Aguilera-Gomez et al. (2022), (5) Ruchti et al. (2011), (6) Kirby et al. (2016), (7) Mucciarelli et al. (2019), (8) Kraft et al. (1999), (9) Monaco et al. (2012), (10) Koch et al. (2011), (11) Gruyters et al. (2016).

References

  1. Adamów, M., Niedzielski, A., Villaver, E., et al. 2012, ApJ, 754, L15 [CrossRef] [Google Scholar]
  2. Adibekyan, V. 2019, Geosciences, 9, 105 [Google Scholar]
  3. Aguado, D. S., Allende Prieto, C., González Hernández, J. I., et al. 2018, ApJ, 854, L34 [NASA ADS] [CrossRef] [Google Scholar]
  4. Aguilera-Gómez, C., Monaco, L., Mucciarelli, A., et al. 2022, A&A, 657, A33 [NASA ADS] [CrossRef] [EDP Sciences] [Google Scholar]
  5. Andrievsky, S. M., Spite, M., Korotin, S. A., et al. 2007, A&A, 464, 1081 [NASA ADS] [CrossRef] [EDP Sciences] [Google Scholar]
  6. Arentsen, A., Starkenburg, E., Shetrone, M. D., et al. 2019, A&A, 621, A108 [NASA ADS] [CrossRef] [EDP Sciences] [Google Scholar]
  7. Asplund, M., Lambert, D. L., Nissen, P. E., et al. 2006, ApJ, 644, 229 [Google Scholar]
  8. Asplund, M., Grevesse, N., Sauval, A. J., et al. 2009, ARA&A, 47, 481 [Google Scholar]
  9. Aoki, W., Beers, T. C., Sivarani, T., et al. 2008, ApJ, 678, 1351 [Google Scholar]
  10. Aoki, W., Barklem, P. S., Beers, T. C., et al. 2009, ApJ, 698, 1803 [NASA ADS] [CrossRef] [Google Scholar]
  11. Bailer-Jones, C. A. L., Rybizki, J., Fouesneau, M., et al. 2021, AJ, 161, 147 [NASA ADS] [CrossRef] [Google Scholar]
  12. Bastian, N., & Lardo, C. 2018, ARA&A, 56, 83 [Google Scholar]
  13. Bergemann, M., Kudritzki, R.-P., Würl, M., et al. 2013, ApJ, 764, 115 [NASA ADS] [CrossRef] [Google Scholar]
  14. Bernstein, R., Shectman, S. A., Gunnels, S. M., et al. 2003, Proc. SPIE, 4841, 1694 [NASA ADS] [CrossRef] [Google Scholar]
  15. Bonifacio, P., & Molaro, P. 1997, MNRAS, 285, 847 [NASA ADS] [CrossRef] [Google Scholar]
  16. Bonifacio, P., Caffau, E., Spite, M., et al. 2015, A&A, 579, A28 [NASA ADS] [CrossRef] [EDP Sciences] [Google Scholar]
  17. Bonifacio, P., Caffau, E., Spite, M., et al. 2018, A&A, 612, A65 [NASA ADS] [CrossRef] [EDP Sciences] [Google Scholar]
  18. Caffau, E., Ludwig, H.-G., Steffen, M., et al. 2011, Sol. Phys., 268, 255 [NASA ADS] [CrossRef] [Google Scholar]
  19. Cameron, A. G. W. 1955, ApJ, 121, 144 [NASA ADS] [CrossRef] [Google Scholar]
  20. Cameron, A. G. W., & Fowler, W. A. 1971, ApJ, 164, 111 [NASA ADS] [CrossRef] [Google Scholar]
  21. Carney, B. W., Fry, A. M., & Gonzalez, G. 1998, AJ, 116, 2984 [NASA ADS] [CrossRef] [Google Scholar]
  22. Casey, A. R., Ruchti, G., Masseron, T., et al. 2016, MNRAS, 461, 3336 [NASA ADS] [CrossRef] [Google Scholar]
  23. Charbonnel, C., & Balachandran, S. C. 2000, A&A, 359, 563 [Google Scholar]
  24. Charbonnel, C., & Primas, F. 2005, A&A, 442, 961 [NASA ADS] [CrossRef] [EDP Sciences] [Google Scholar]
  25. Charbonnel, C., Lagarde, N., Jasniewicz, G., et al. 2020, A&A, 633, A34 [NASA ADS] [CrossRef] [EDP Sciences] [Google Scholar]
  26. Coc, A., & Vangioni, E. 2017, Int. J. Mod. Phys. E, 26, 1741002 [CrossRef] [Google Scholar]
  27. D’Antona, F., D’Ercole, A., Carini, R., et al. 2012, MNRAS, 426, 1710 [CrossRef] [Google Scholar]
  28. Deepak, Lambert, D. L., & Reddy, B. E. 2020, MNRAS, 494, 1348 [Google Scholar]
  29. de La Reza, R., Drake, N. A., & da Silva, L. 1996, ApJ, 456, L115 [NASA ADS] [CrossRef] [Google Scholar]
  30. Denissenkov, P. A., & Herwig, F. 2004, ApJ, 612, 1081 [NASA ADS] [CrossRef] [Google Scholar]
  31. Denissenkov, P. A., Pinsonneault, M., & MacGregor, K. B. 2009, ApJ, 696, 1823 [NASA ADS] [CrossRef] [Google Scholar]
  32. Doherty, C. L., Gil-Pons, P., Lau, H. H. B., et al. 2014, MNRAS, 441, 582 [NASA ADS] [CrossRef] [Google Scholar]
  33. Domínguez, I., Abia, C., Straniero, O., et al. 2004, A&A, 422, 1045 [NASA ADS] [CrossRef] [EDP Sciences] [Google Scholar]
  34. D’Orazi, V., Gratton, R. G., Angelou, G. C., et al. 2015, ApJ, 801, L32 [CrossRef] [Google Scholar]
  35. Fischer, D. A., & Valenti, J. 2005, ApJ, 622, 1102 [NASA ADS] [CrossRef] [Google Scholar]
  36. Fishlock, C. K., Karakas, A. I., Lugaro, M., et al. 2014, ApJ, 797, 44 [NASA ADS] [CrossRef] [Google Scholar]
  37. Gaia Collaboration (Prusti, T., et al.) 2016, A&A, 595, A1 [NASA ADS] [CrossRef] [EDP Sciences] [Google Scholar]
  38. Gaia Collaboration (Brown, A. G. A., et al.) 2021, A&A, 649, A1 [NASA ADS] [CrossRef] [EDP Sciences] [Google Scholar]
  39. Gao, Q., Shi, J.-R., Yan, H.-L., et al. 2019, ApJS, 245, 33 [NASA ADS] [CrossRef] [Google Scholar]
  40. Gonzalez, O. A., Zoccali, M., Monaco, L., et al. 2009, A&A, 508, 289 [NASA ADS] [CrossRef] [EDP Sciences] [Google Scholar]
  41. Gruyters, P., Lind, K., Richard, O., et al. 2016, A&A, 589, A61 [NASA ADS] [CrossRef] [EDP Sciences] [Google Scholar]
  42. Handberg, R., Lund, M. N., White, T. R., et al. 2021, AJ, 162, 170 [NASA ADS] [CrossRef] [Google Scholar]
  43. Hansen, T., Hansen, C. J., Christlieb, N., et al. 2014, ApJ, 787, 162 [NASA ADS] [CrossRef] [Google Scholar]
  44. Harris, W. E. 1996, AJ, 112, 1487 [Google Scholar]
  45. Hatt, E., Nielsen, M. B., Chaplin, W. J., et al. 2023, A&A, 669, A67 [NASA ADS] [CrossRef] [EDP Sciences] [Google Scholar]
  46. Hon, M., Huber, D., Kuszlewicz, J. S., et al. 2021, ApJ, 919, 131 [NASA ADS] [CrossRef] [Google Scholar]
  47. Huang, C. X., Vanderburg, A., Pál, A., et al. 2020, Res. Notes Am. Astron. Soc., 4, 204 [Google Scholar]
  48. Iben, I. 1967, ApJ, 147, 624 [NASA ADS] [CrossRef] [Google Scholar]
  49. Ito, H., Aoki, W., Beers, T. C., et al. 2013, ApJ, 773, 33 [Google Scholar]
  50. Ivans, I.I., Sneden, C., Gallino, R., et al. 2005, ApJ, 627, L145 [NASA ADS] [CrossRef] [Google Scholar]
  51. Izzo, L., Della Valle, M., Mason, E., et al. 2015, ApJ, 808, L14 [NASA ADS] [CrossRef] [Google Scholar]
  52. Izzo, L., Molaro, P., Bonifacio, P., et al. 2023, Exp. Astron., 55, 191 [NASA ADS] [CrossRef] [Google Scholar]
  53. Jacobson, H. R., Keller, S., Frebel, A., et al. 2015, ApJ, 807, 171 [NASA ADS] [CrossRef] [Google Scholar]
  54. Johnson, J. A., Aller, K. M., Howard, A. W., et al. 2010, PASP, 122, 905 [Google Scholar]
  55. Kelson, D. D. 2003, PASP, 115, 688 [NASA ADS] [CrossRef] [Google Scholar]
  56. Kirby, E. N., Fu, X., Guhathakurta, P., et al. 2012, ApJ, 752, L16 [NASA ADS] [CrossRef] [Google Scholar]
  57. Kirby, E. N., Guhathakurta, P., Zhang, A. J., et al. 2016, ApJ, 819, 135 [NASA ADS] [CrossRef] [Google Scholar]
  58. Koch, A., Lind, K., & Rich, R. M. 2011, ApJ, 738, L29 [NASA ADS] [CrossRef] [Google Scholar]
  59. Kowkabany, J., Ezzeddine, R., Charbonnel, C., et al. 2022, arXiv e-prints [arXiv: 2209.02184] [Google Scholar]
  60. Kraft, R. P., Peterson, R. C., Guhathakurta, P., et al. 1999, ApJ, 518, L53 [NASA ADS] [CrossRef] [Google Scholar]
  61. Kurucz, R. L. 1993, VizieR Online Data Catalog: VI/39 [Google Scholar]
  62. Kurucz, R. L. 2005, Mem. Soc. Astron. It. Suppl., 8, 14 [Google Scholar]
  63. Lallement, R., Vergely, J. L., Babusiaux, C., et al. 2022, A&A, 661, A147 [NASA ADS] [CrossRef] [EDP Sciences] [Google Scholar]
  64. Li, H., Aoki, W., Zhao, G., et al. 2015, PASJ, 67, 84 [NASA ADS] [CrossRef] [Google Scholar]
  65. Li, H., Aoki, W., Matsuno, T., et al. 2018, ApJ, 852, L31 [NASA ADS] [CrossRef] [Google Scholar]
  66. Lightkurve Collaboration (de Cardoso, J. V. M., et al.) 2018, Astrophysics Source Code Library [record ascl:1812.013] [Google Scholar]
  67. Lind, K., Primas, F., Charbonnel, C., et al. 2009, A&A, 503, 545 [NASA ADS] [CrossRef] [EDP Sciences] [Google Scholar]
  68. Lind, K., Asplund, M., Barklem, P. S., et al. 2011, A&A, 528, A103 [NASA ADS] [CrossRef] [EDP Sciences] [Google Scholar]
  69. Lodders, K. 2010, Astrophys. Space Sci. Proc., 16, 379 [NASA ADS] [CrossRef] [Google Scholar]
  70. Lomb, N. R. 1976, Ap & SS, 39, 447 [NASA ADS] [CrossRef] [Google Scholar]
  71. Lombardo, L., François, P., Bonifacio, P., et al. 2021, A&A, 656, A155 [NASA ADS] [CrossRef] [EDP Sciences] [Google Scholar]
  72. Lombardo, L., Bonifacio, P., François, P., et al. 2022, A&A, 665, A10 [NASA ADS] [CrossRef] [EDP Sciences] [Google Scholar]
  73. Lucatello, S., Gratton, R., Cohen, J. G., et al. 2003, AJ, 125, 875 [Google Scholar]
  74. Lund, M. N., Handberg, R., Buzasi, D. L., et al. 2021, ApJS, 257, 53 [NASA ADS] [CrossRef] [Google Scholar]
  75. Mackereth, J. T., Miglio, A., Elsworth, Y., et al. 2021, MNRAS, 502, 1947 [NASA ADS] [CrossRef] [Google Scholar]
  76. Maldonado, J., Villaver, E., & Eiroa, C. 2013, A&A, 554, A84 [NASA ADS] [CrossRef] [EDP Sciences] [Google Scholar]
  77. Martell, S. L., & Shetrone, M. D. 2013, MNRAS, 430, 611 [NASA ADS] [CrossRef] [Google Scholar]
  78. Mashonkina, L., Pakhomov, Y., Sitnova, T., et al. 2023, MNRAS, 524, 3526 [Google Scholar]
  79. Masseron, T., Johnson, J. A., Lucatello, S., et al. 2012, ApJ, 751, 14 [NASA ADS] [CrossRef] [Google Scholar]
  80. Matsuno, T., Aoki, W., Beers, T. C., et al. 2017, AJ, 154, 52 [NASA ADS] [CrossRef] [Google Scholar]
  81. Meléndez, J., Schirbel, L., Monroe, T. R., et al. 2014, A&A, 567, L3 [NASA ADS] [CrossRef] [EDP Sciences] [Google Scholar]
  82. Molaro, P., Izzo, L., Selvelli, P., et al. 2023, MNRAS, 518, 2614 [Google Scholar]
  83. Monaco, L., Villanova, S., Moni Bidin, C., et al. 2011, A&A, 529, A90 [NASA ADS] [CrossRef] [EDP Sciences] [Google Scholar]
  84. Monaco, L., Villanova, S., Bonifacio, P., et al. 2012, A&A, 539, A157 [NASA ADS] [CrossRef] [EDP Sciences] [Google Scholar]
  85. Monaco, L., Boffin, H. M. J., Bonifacio, P., et al. 2014, A&A, 564, L6 [NASA ADS] [CrossRef] [EDP Sciences] [Google Scholar]
  86. Mucciarelli, A., & Bonifacio, P. 2020, A&A, 640, A87 [NASA ADS] [CrossRef] [EDP Sciences] [Google Scholar]
  87. Mucciarelli, A., Salaris, M., & Bonifacio, P. 2012, MNRAS, 419, 2195 [NASA ADS] [CrossRef] [Google Scholar]
  88. Mucciarelli, A., Pancino, E., Lovisi, L., et al. 2013a, ApJ, 766, 78 [NASA ADS] [CrossRef] [Google Scholar]
  89. Mucciarelli, A., Bellazzini, M., Catelan, M., et al. 2013b, MNRAS, 435, 3667 [NASA ADS] [CrossRef] [Google Scholar]
  90. Mucciarelli, A., Monaco, L., Bonifacio, P., et al. 2019, A&A, 623, A55 [NASA ADS] [CrossRef] [EDP Sciences] [Google Scholar]
  91. Mucciarelli, A., Bellazzini, M., & Massari, D. 2021a, A&A, 653, A90 [NASA ADS] [CrossRef] [EDP Sciences] [Google Scholar]
  92. Mucciarelli, A., Monaco, L., Bonifacio, P., et al. 2021b, A&A, 652, A139 [NASA ADS] [CrossRef] [EDP Sciences] [Google Scholar]
  93. Mucciarelli, A., Monaco, L., Bonifacio, P., et al. 2022, A&A, 661, A153 [NASA ADS] [CrossRef] [EDP Sciences] [Google Scholar]
  94. Nordlander, T., & Lind, K. 2017, A&A, 607, A75 [NASA ADS] [CrossRef] [EDP Sciences] [Google Scholar]
  95. Norris, J. E., Bessell, M. S., Yong, D., et al. 2013, ApJ, 762, 25 [NASA ADS] [CrossRef] [Google Scholar]
  96. Palacios, A., Charbonnel, C., & Forestini, M. 2001, A&A, 375, L9 [NASA ADS] [CrossRef] [EDP Sciences] [Google Scholar]
  97. Pasquini, L., Döllinger, M. P., Weiss, A., et al. 2007, A&A, 473, 979 [NASA ADS] [CrossRef] [EDP Sciences] [Google Scholar]
  98. Pietrinferni, A., Hidalgo, S., Cassisi, S., et al. 2021, ApJ, 908, 102 [NASA ADS] [CrossRef] [Google Scholar]
  99. Placco, V. M., Beers, T. C., Reggiani, H., et al. 2016, ApJ, 829, L24 [NASA ADS] [CrossRef] [Google Scholar]
  100. Ricker, G. R., Winn, J. N., Vanderspek, R., et al. 2015, J. Astron. Telesc. Instrum. Syst., 1, 014003 [Google Scholar]
  101. Roederer, I. U., Frebel, A., Shetrone, M. D., et al. 2008, ApJ, 679, 1549 [NASA ADS] [CrossRef] [Google Scholar]
  102. Roederer, I. U., Preston, G. W., Thompson, I. B., et al. 2014, AJ, 147, 136 [Google Scholar]
  103. Romano, D., Magrini, L., Randich, S., et al. 2021, A&A, 653, A72 [NASA ADS] [CrossRef] [EDP Sciences] [Google Scholar]
  104. Ruchti, G. R., Fulbright, J. P., Wyse, R. F. G., et al. 2011, ApJ, 743, 107 [Google Scholar]
  105. Ryan, S. G., Norris, J. E., & Beers, T. C. 1999, ApJ, 523, 654 [CrossRef] [Google Scholar]
  106. Sackmann, I.-J., & Boothroyd, A. I. 1992, ApJ, 392, L71 [NASA ADS] [CrossRef] [Google Scholar]
  107. Sanna, N., Franciosini, E., Pancino, E., et al. 2020, A&A, 639, L2 [NASA ADS] [CrossRef] [EDP Sciences] [Google Scholar]
  108. Sbordone, L., Bonifacio, P., Caffau, E., et al. 2010, A&A, 522, A26 [NASA ADS] [CrossRef] [EDP Sciences] [Google Scholar]
  109. Scargle, J. D. 1982, ApJ, 263, 835 [Google Scholar]
  110. Schlafly, E. F., & Finkbeiner, D. P. 2011, ApJ, 737, 103 [Google Scholar]
  111. Shingles, L. J., Doherty, C. L., Karakas, A. I., et al. 2015, MNRAS, 452, 2804 [NASA ADS] [CrossRef] [Google Scholar]
  112. Siess, L., & Livio, M. 1999, MNRAS, 308, 1133 [Google Scholar]
  113. Silva Aguirre, V., Ruchti, G. R., Hekker, S., et al. 2014, ApJ, 784, L16 [NASA ADS] [CrossRef] [Google Scholar]
  114. Sitnova, T. M., Matsuno, T., Yuan, Z., et al. 2023, MNRAS, 526, 5976 [Google Scholar]
  115. Sivarani, T., Beers, T. C., Bonifacio, P., et al. 2006, A&A, 459, 125 [NASA ADS] [CrossRef] [EDP Sciences] [Google Scholar]
  116. Smith, V. V., Shetrone, M. D., & Keane, M. J. 1999, ApJ, 516, L73 [NASA ADS] [CrossRef] [Google Scholar]
  117. Spite, M., & Spite, F. 1982, Nature, 297, 483 [NASA ADS] [CrossRef] [Google Scholar]
  118. Susmitha, A., Mallick, A., & Reddy, B. E. 2024, ApJ, 966, 109 [NASA ADS] [CrossRef] [Google Scholar]
  119. Takarada, T., Sato, B., Omiya, M., et al. 2018, PASJ, 70, 59 [NASA ADS] [CrossRef] [Google Scholar]
  120. Thompson, I. B., Ivans, I.I., Bisterzo, S., et al. 2008, ApJ, 677, 556 [NASA ADS] [CrossRef] [Google Scholar]
  121. Ventura, P., & D’Antona, F. 2011, MNRAS, 410, 2760 [Google Scholar]
  122. Yong, D., Norris, J. E., Bessell, M. S., et al. 2013, ApJ, 762, 26 [Google Scholar]
  123. Wang, E. X., Nordlander, T., Asplund, M., et al. 2021, MNRAS, 500, 2159 [Google Scholar]
  124. Wolthoff, V., Reffert, S., Quirrenbach, A., et al. 2022, A&A, 661, A63 [NASA ADS] [CrossRef] [EDP Sciences] [Google Scholar]

All Tables

Table 1

Stellar parameters for target star HE 0057-5959.

In the text
Table 2

LTE and NLTE chemical abundances measured in HE 0057-5959, together with the number of used lines and the total uncertainty.

In the text
Table B.1

Main atomic data for the measured transitions used in the analysis.

In the text
Table E.1

Main information about the Li-rich stars discovered so far in the Milky Way field.

In the text
Table E.2

Main information about the Li-rich stars discovered so far in Galactic globular clusters. The clusters are in order of right ascension. We list for each target the Gaia ED3 identification number and the alternative identification number or name used in the literature. Stellar parameters and A(Li)nlte are those described in Sect. 2. [Fe/H] and [Na/Fe] are those listed in the corresponding papers. The evolutionary stage of each target is obtained as described in Sect. 4: (1) pre-FDU, (2) FDU, (3) post-FDU / pre-RGBb, (4) RGBb. References: (1) Smith et al. (1999), (2) D’Orazi et al. (2015), (3) Sanna et al. (2020), (4) Aguilera-Gomez et al. (2022), (5) Ruchti et al. (2011), (6) Kirby et al. (2016), (7) Mucciarelli et al. (2019), (8) Kraft et al. (1999), (9) Monaco et al. (2012), (10) Koch et al. (2011), (11) Gruyters et al. (2016).

In the text

All Figures

thumbnail Fig. 1

Spectral regions of MIKE spectrum (blue points) around the Li resonance line and a Na D line (upper and lower panel, respectively), with superimposed synthetic spectra calculated with the best-fit abun- dance (central red curve) and ±0.1 dex from the best-fit abundance.
In the text
thumbnail Fig. 2

The evolutionary stage and the surface A(Li) of HE 0057-5959 . Left panel: position of target star HE 0057-5959 in the Hertzsprung-Russell diagram (red square) in comparison with the theoretical track of a star with 0.76 M⊙, [Fe/H]=−1.0 dex and [α/Fe]=+0.4 dex, coloured according to the predicted A(Li). The colour-scale is shown on the right side. The model is re-scaled in order to match the average A(Li)nlte of the RGB stars by Mucciarelli et al. (2022). Right panel: A(Li)NLTE as function of log(L/L⊙) for HE 0057-5959 (red square) in comparison with the RGB star sample by Mucciarelli et al. (2022, cyan circles) and the stars of the metal-poor ([Fe/H]~−2.0 dex) globular cluster NGC 6397 (Lind et al. 2009, grey circles). The luminosity level of the RGBb for the metallicity of HE 0057-5959 is marked with an arrow.
In the text
thumbnail Fig. 3

Position in Hertzsprung-Russell diagram of the target star HE 0057-5959 (red square) and of the other metal-poor ([Fe/H]<−1 dex) Li-rich stars discovered so far (see details in Sect. 4): green circles are Galactic-field stars and blue circles are globular-cluster stars. Two theoretical isochrones with [Fe/H]=−3.2 and −1.2 dex, an α-enhanced chemical mixture, and an age of 13 Gyr (Pietrinferni et al. 2021) are shown for reference (black curves). The two grey shaded areas indicate the loci where the drops of A(Li)nlte due to the FDU and further extra-mixing episodes, respectively, occur.
In the text
thumbnail Fig. 4

Surface A(Li) of the Li-rich stars discov- ered so far. Upper left panel: behaviour of A(Li)NLTE as function of log(L/L⊙) for all the metal-poor ([Fe/H]<−1 dex) Li-rich stars discovered so far. Green circles are Milky Way field stars, blue cir- cles are globular-cluster stars, and the red squares mark HE 0057-5959. The stars of the metal-poor ([Fe/H]~−2.0 dex) globular cluster NGC 6397 (grey circles, Lind et al. 2009) are shown as a reference (and to identify the luminosity of the FDU and RGBb). Upper right panel: run of A(Li)NLTE as func- tion of [Fe/H] for the Li-rich stars located before the FDU. Red dashed line indicates the WMAP/SBNN A(Li) (Coc & Vangioni 2017). The cyan circles are Li-normal Milky Way field stars (Ryan et al. 1999; Lucatello et al. 2003; Ivans et al. 2005; Asplund et al. 2006; Sivarani et al. 2006; Thompson et al. 2008; Sbordone et al. 2010; Masseron et al. 2012; Ito et al. 2013; Hansen et al. 2014; Bonifacio et al. 2015; Li et al. 2015; Placco et al. 2016; Matsuno et al. 2017; Aguado et al. 2018; Bonifacio et al. 2018). Lower-left panel: run of A(Li)NLTE as function of [Fe/H] for the Li-rich stars located after the com- pletion of FDU and before the RGBb. The open cir- cle indicates the star 2MASS J19524490-6008132 with an uncertain attribution to this group. The cyan circles are Li-normal Milky Way field stars (Mucciarelli et al. 2022). Lower right panel: run of A(Li)NLTE as function of [Fe/H] for the Li-rich stars located after the RGBb. Open circle indicates the star GSC 03797-00204 with an uncertain attribution to this group.
In the text
thumbnail Fig. 5

Behaviour of initial A(Li), accounting for the dilution effect due to the FDU, for HE 0057-5959 and other metal-poor, Li-rich stars fainter than the RGBb (same symbols as Fig. 3).
In the text
thumbnail Fig. 6

Behaviour of [Na/Fe] as function of [Fe/H] of the metal- poor, Li-rich stars (same symbols of Fig. 3) in comparison with the NLTE abundances for metal-poor Milky Way stars (grey circles) from Andrievsky et al. (2007) and Lombardo et al. (2022). The [Na/Fe] abun- dance ratio of HE 0057-5959 plotted here accounts for NLTE effects for Na abundance only. The position of the Li-rich star by Kowkabany et al. (2022), the most Li-rich discovered so far, is marked.
In the text
thumbnail Fig. A.1

Power spectral density for HE 0057-5959 based on TESS observations during Sector 1 and 2. The light curve from which we obtain the power spectral density is provided by the TESS Asteroseis- mic Science Operations Center (TASOC; Handberg et al. 2021; Lund et al. 2021). We find no evidence of solar-like oscillations.
In the text
thumbnail Fig. C.1

Difference of our abundances with respect to those by Yong et al. (2013) and Jacobson et al. (2015), red and blue points respectively, as a function of the atomic number. We consider our LTE abundances but for Li and Na that are provided by Yong et al. (2013) and Jacobson et al. (2015) corrected for NLTE effects.
In the text
thumbnail Fig. D.1

Spectral region of the MIKE spectrum of HE 0057-5959 with marked the photospheric and interstellar Na D lines (blue and red arrows, respectively).
In the text

Current usage metrics show cumulative count of Article Views (full-text article views including HTML views, PDF and ePub downloads, according to the available data) and Abstracts Views on Vision4Press platform.

Data correspond to usage on the plateform after 2015. The current usage metrics is available 48-96 hours after online publication and is updated daily on week days.

Initial download of the metrics may take a while.