Diversity of planetary origins and compositions traced by stellar abundance differences among co-natal star systems

So called co-natal stars are wide binaries that form from the same molecular cloud and are expected to share the same stellar composition. However, observations indicate stellar abundance differences in a small fraction of these co-natal systems. These stellar abundance differences have been attributed to the ingestion of planets by their host stars. In this study, we analyze a previous sample of co-natal stars to determine the composition of the planets that had been ingested by their host stars. The abundance differences in these co-natal pairs come in different flavours, where systems can show a positive abundance difference of the refractory elements (Fe, Mg, Si) paired with a positive, zero, or negative abundance difference of volatiles (O, C, S). Positive trends of refractory and volatile abundance differences can be explained (at minimum) by the ingestion of a single planet by one of the stars. However, positive trends of refractory abundance differences paired with zero or negative abundance differences of oxygen and carbon imply that both stars are required to ingest planets formed at different locations (i.e. interior or exterior to the water and/or carbon grain evaporation front). Furthermore, some systems show a sulfur abundance difference that is a factor of a few lower than the abundance difference of Fe, Si, and Mg, even though sulfur is mostly accreted via the moderately volatile compound FeS. We show that this can be explained by the migration of one of the planets across the FeS evaporation front, where the planet first accretes large amounts of sulfur with the solids and then no sulfur, while still accreting the other refractory elements before being ingested by its host star. Our study shows that the planetary compositions derived from the detailed stellar abundances are sufficient to explain the measured abundance differences of co-natal stars, giving us a handle on the composition of exoplanets. At the same time it implies very different planetary compositions caused by the different formation location of these planets, in agreement with planet formation models dedicated to the formation of inner super-Earths and sub-Neptunes.