Lithium as a probe of stellar and galactic physics

Abstract

Lithium plays a unique role in astrophysics, as it is a powerful diagnostic for the physics and evolution of low-mass stars, Galactic archaeology, and cosmology. We review the Li observations in stars at different phases of their evolution, the strengths and the limitations of the current theoretical stellar models to explain the Li abundance data, our understanding of the Li sources and of the evolution of Li through- out the Galactic history. Key takeaways from the current state of the research in the field are: 1) Stellar evolution models accounting for fundamental transport processes of chemical species and angular momentum hold the promise of providing a common stellar Li depletion explanation to the Li abundance patterns observed in all Galactic stellar populations, including the dip and the plateau(s). 2) Novae are most probably the main source of Li in the Galaxy, on observational (but not yet theoretically established) grounds. 3) Radial migration of stars in the Galactic disk holds the key to understand many aspects of the Li evolution in the Milky Way.

Figures (10)

• Figure 1: Left : Half-times (time of survival of half the original amount) of the light isotopes $^6$Li, $^7$Li, $^9$Be, $^{10}$B and $^{11}$B as a function of temperature in an environment of solar composition and density $\rho$=1 gr cm$^{-3}$. Timescales for 2-body reactions are inversely proportional to $\rho$. The bottom of the convective zone of the Sun (BCZ$\odot$) has a density of $\rho$=0.2 gr cm$^{-3}$, implying a Li burning timescale larger than a Hubble time. Middle: Li and Be preservation zones in a classical stellar model of a 1.2 M$_{\odot}$ . The "step-like" abundance profile results from the temperature gradient in the external layers and the sensitivity to temperature of the proton captures on $^7$Li and $^9$Be 2000IAUS..198...61D. Right: Predictions of the classical theory, with only pre-main sequence Li depletion in cooler stars 2000IAUS..198...61D.
• Figure 2: Li abundances in OC and halo dwarf stars as a function of effective temperature. The stellar types and the main Li abundance patterns discussed in the text are indicated. The OC data are: for Pleiades from Boesgaard2003a and Bouvier2018, for M67 from Pace2012, for Praesepe from 2017AJ....153..128C, for Coma Ber from 2000AA...354..216B and for M48 from 2023ApJ...952...71S. The Li abundances of Pop II halo dwarfs are from the Norris_2023 sample, as cleaned in Borisov2024. The grey line indicates the meteoritic Li abundance. The temperature scales vary among the studies and have not been homogeneised here.
• Figure 3: Abundances of Li ( top), Be ( second from top), B ( third from top) and $\upsilon sini$ ( bottom) versus T$_{\rm eff}$ for dwarf stars in the Pleiades (green dots), Hyades (blue squares) and NGC752 (red pentagons), with upper limits indicated by open downturned triangles. Horizontal grey lines indicate proto-solar values. Figure adapted from 2023ApJ...943...40B and references therein.
• Figure 4: Li abundance (3D, NLTE) as a function of [$\alpha$/Fe] for the Norris_2023 and GALAH DR3 sample stars (squares and circles respectively) with colour-coded age. Solid lines show the theoretical Li upper and lower envelopes (left and right panels, respectively) at different ages (in the legend), i.e., the maximum and minimum Li values at a given [Fe/H] and age expected from Type II models with 5800 K$<$ T$_{\rm eff}$(ZAMS) $<$6500 K. The dashed line indicates the initial Li abundance assumed in the models. In the left panel, the four-pointed stars show predictions for 1 M$_{\odot}$ star from models of Dumont2021a with age similarly colour-coded. Figure from Borisov2024
• Figure 5: Top: Stellar astrophysical sites of thermonuclear production of $^7$Be with corresponding temperature ranges: core H-burning (horizontally extended from H-ignition on the left to H-exhaustion on the right), in Intermediate Mass Stars (IMS) and Massive Stars (MS), shell H-burning in Red Giant Branch stars (RGB, red) and Hot Bottom Burning in Asymptotic Giant Branch stars (AGB, orange); range of peak temperatures in He-layers of CCSN of 15 to 30 M$_{\odot}$, where $\nu$-induced nucleosynthesis occurs (violet box), novae (blue box) and SBBN (purple). Bottom: Overproduction factors $\rm f_0$ of $^7$Be (with respect to the meteoritic value of $^7$Li). The brown solid curve shows results for quiescent H-burning of solar mixture material at constant temperatures. Li is overpoduced, albeit for short timescales: a few 10$^3$ yr for T=25 MK to about a week for T=120 MK. The yellow shaded area indicates observations of Li-richest giant stars. The upper dotted horizontal line indicates the maximum possible value $\rm f_0$, with all solar He$^3$ turned into $^7$Be; some nova observations indicate higher overproduction (see Fig. \ref{['fig_Li_BBN_nova']}, right). Blue asterisks indicate maximum $\rm f_0$ values obtained in nova simulations for indicated peak temperatures (above 180 MK) from Starrfield2024, while the cyan-coloured shaded area indicates the range from nova observations Molaro2023. The vertical violet segment joins peak overproduction f$_{\rm 0,PEAK}$ of $^7$Li in the He-layers from $\nu$-induced nucleosynthesis and corresponding value in the total ejecta f$_{\rm 0,EJEC}$ in a 27 M$_{\odot}$ CCSN model Sieverding_2019. Adopted values (in horizontal lines) are: A(Li)=3.39 for Proto-solar Lodders_2025, 2.7 for SBBN Pitrou_2018 and 2.2 for halo plateau Bonifacio_2025.