How well defined are the elemental shells in evolved stars?

Question

Most textbooks and popular science show highly evolved massive stars with an onion-like structure with layers or shells of different elements fusing. This is clearly approximate (as noted in the comments here), but how approximate is it?

It would seem to me that even in a non-convective environment there would be significant diffusion over stellar timescales that would mix the layers, perhaps into a soft distribution with higher probability of finding heavy elements deeper. But what does actual models and simulations say?

Answer 1

Have a look at this figure form Hirschi, Meynet, Maeder (2004)

They are simulating the internal structure of the star and studying its evolution in the pre-supernova phase. This figure refers to a non rotating star of $M=20M_\odot$ at the end of the Silicon burning stage. It shows the relative abundance of the elements as a function of the lagrangian mass coordinate.

In this figure, the shells are not as clear-cut as in the cartoon in the question you cite, but it is easy to identify a sequence of shells nonetheless.

The sequence I see here is Ni - Si/S - O - O/Ne - O/C - He - H/He

If you go and look at the paper itself, there are many more figures like this one, referring to rotating stars and to different burning stages.