The measurement of the lithium isotope ratio in stars is extremely challenging, both from an observational and from a theoretical perspective. Observers need the highest quality data that modern telescopes and spectrographs can provide to disentangle the weak signature of the lighter isotope from the observational noise – in practice, the 6Li detection limit in metal-poor stellar spectra is only about 2% of the total lithium content. Moreover, many agents can influence the shape of spectral line profiles and hence affect the isotopic ratio, and they have to be modelled correctly.

First, thermal and convective gas motions cause Doppler shifts in the spectral lines, which call for realistic 3D radiation-hydrodynamic simulations of the atmospheres of stars, such as the Stagger models developed at MPA. Second, to correctly model the distribution of atoms in different states of excitation and ionisation one needs to account for the strong departures from local thermodynamic equilibrium (LTE). Such complex line-formation calculations are expensive and they must run for weeks on powerful multi-core machines. This is why until now, not all effects could be considered at the same time and simplifying assumptions have been made.

For the first time, a combined 3D, non-LTE technique has been be applied to model the lithium, sodium and calcium spectral lines in four very metal-poor stars. The aim was to constrain the lithium isotope ratios, while the other neutral elements serve as calibrators of the unknown projected rotational velocity of the stars. Surprisingly, the astronomers found that none of the stars displayed a significant presence of the lighter 6Li, contrary to evidence put forward in several studies over the past two decades.

The refined models show in particular that the assumptions of LTE leads to systematic overestimations, even false detections, of 6Li over a wide parameter space. The most interesting example is the metal-poor turn-off star HD84937, for which an undisputed detection has been demonstrated in at least three competing studies, while the new model shows no significant signs of 6Li.

The findings for all four stars are in gratifying agreement with the standard Big Bang nucleosynthesis model, which forges only insignificant amounts of the lighter isotope. What remains to be properly explained is why the element abundances of 7Li instead fall short of the primordial prediction. Our 3D, NLTE modelling strengthens the exciting claim that stars can act as sinks of both lithium isotopes, slowly draining their atmospheres of these and heavier elements over time. This slow diffusion process has been theoretically postulated and has the great potential to explain why the observed stellar abundances of heavy lithium are lower than expected. Thereby, both cosmological lithium problems, which have haunted particle physicists and astrophysicists since the launch of the WMAP satellite, can find their solution in improved physics of the stellar atmospheres.

Karin Lind (MPA), Jorge Melendez (Department of Astronomy, University of Sao Paulo, Brasil), Martin Asplund (Mount Stromlo Observatory, Australian National University, Australia), Remo Collet (Mount Stromlo Observatory, Australian National University, Australia), Zazralt Magic (MPA)

Reference

Lind K., Melendez J., Asplund M., Collet R. & Magic Z., "The lithium isotopic ratio in very metal-poor stars", (submitted to A&A)

Further reading

Asplund, M., Lambert, D. L., Nissen, P. E., Primas, F., & Smith, V.V., "Lithium Isotopic Abundances in Metal-poor Halo Stars", 2006, ApJ, 644, 229

Cayrel, R., Steffen, M., Chand, H., et al., "Line shift, line asymmetry, and the 6Li/7Li isotopic ratio determination", 2007, A&A, 473, L37