We collect the observational results that have been well established and must be taken into account by any global explanation of the Am phenomenon.
In Praesepe cluster, the Am stars are overabundant in Fe by a factor of 2 (+0.35 dex) compared with the F-G stars, which are not evolved and have the initial abundance. In the first paper, we similarly have found an overabundance of + 0.30 dex in a younger cluster, the Pleiades.
Just like for the Pleiades, our Li results in Praesepe challenge predictions from model envelopes coupling diffusion and evolution in non-rotating Population I A stars by Richer & Michaud (1993): at the age of Praesepe no lithium is expected to be observed in the slowly-rotating Am stars. Richer and co-workers (1997, 1998) improve radiative force calculations and show up the possible strong coupling that can exist between the radiative force of one element and the abundance of another one. For example the Li force, which is coupled with the He abundance, increases by a large factor when helium abundance decreases by gravitational settling. Richer and co-workers foresee to calculate self-consistent evolution models with the refined forces. Significant changes are expected in the evolution of the lithium distribution in the radiative stellar envelope, leading to Li better supported, i.e., smaller photospheric underabundances for stars in the range 9000-7200 K and a better agreement with our observational results.
In the range 8100-8000 K, the three stars, HD 73174, HD 73709, and HD 73730 with well-defined abundances lie at the same place inside the sequence turn-off of the cluster in Fig. 1. So is the place of the SB2 HD 73618 when assuming its binarity with both similar components. We are, thus, studying four stars of the same cluster; they presumably have the same age, mass, and initial chemical composition. They are turn-off stars at the same stage of evolution: the standard deviation of the mean of the V magnitude is 0.15. They have begun their evolution out of the Main Sequence: they are brighter than ZAMS stars of the same temperature by about one magnitude. These four Am stars have the same abundances, particularly in Si, S, and Fe for which the standard deviation of each mean is about 0.10 dex. We,however, note that the abundances of HD 73709 are the highest of each range. Li is even significantly higher and similar to that of cluster normal A stars when for the three other Am stars Li is similar to that of cluster Am stars (Burkhart & Coupry 1995). Many stellar parameters are the same. Another important parameter can make the difference between HD 73709 and the others: rotation or more exactly angular momentum history. During the whole evolution, this can affect the thickness of the mixing zone and the temperature at its bottom, and lead to "Li is affected when Fe and Si are not.".
A more extended refined abundance analysis of this "8100-8000 K" Am quartet will be fruitful. It is technically possible in consideration of the magnitudes and projected rotational velocities. We can imagine that the peculiar behavior of Li (and others elements to be found) may add enough constraints upon the models to test them.
The cooler well-studied star, HD 73045, at 7500K, is not distinguished from the quartet with respect to abundances of Li, Al, Si, Fe, and Ni, although that star is normally less evolved.
© European Southern Observatory (ESO) 1998
Online publication: September 17, 1998