Why do some r-process-enhanced stars have extra thorium and uranium?
In plain words
In old stars enhanced in r-process elements the medium-heavy elements follow the Sun's pattern closely, yet a sizable minority show extra thorium and uranium, hinting that the conditions or sites differ.
Precise statement
In r-process-enhanced metal-poor stars ($[\mathrm{Eu}/\mathrm{Fe}]>+0.3$) the abundance pattern from Ba to Hf closely matches the scaled solar r-process, while $\log(\mathrm{Th}/\mathrm{Eu})$ varies, with a sizable minority (estimates near one third) showing an actinide boost. Determine whether the variation reflects different electron-fraction ($Y_e$) distributions of merger dynamical ejecta versus disk winds, different sites, or fission-physics uncertainties. Answer: a site and $Y_e$ distribution reproducing the observed $\mathrm{Th}/\mathrm{Eu}$ distribution and its correlation with other abundance ratios.
What would settle it
Th, U and Eu abundances for a large sample of r-process-enhanced stars compared with yields from merger and collapsar simulations with consistent nuclear inputs.