We present detailed evolutionary calculations focused on the evolution of intermediate mass stars with 3 M _ { \odot } \leq M \leq 9 M _ { \odot } of metallicity typical of the Large Magellanic Cloud ( LMC ) , i.e . Z = 0.008 .
We compare carefully the models calculated by adopting a diffusive scheme for chemical mixing , in which nuclear burning and mixing are self-consistently coupled , while the eddy velocities beyond the formal convective core boundary are treated to decay exponentially , and those calculated with the traditional instantaneous mixing approximation .
We find that : i ) the physical and chemical behaviour of the models during the H-burning phase is independent of the scheme used for the treatment of mixing inside the CNO burning core ; ii ) the duration of the He-burning phase relative to the MS phase is systematically longer in the diffusive models , due to a slower redistribution of helium to the core from the outer layers ; iii ) the fraction of time spent in the blue part of the clump , compared to the stay in the red , is larger in the diffusive models .
The differences described in points ii ) and iii ) tend to vanish for M > 6 M _ { \odot } .
In terms of the theoretical interpretation of an open cluster stellar population , the differences introduced by the use of a self-consistent scheme for mixing in the core with adjacent exponential decay are relevant for ages in the range 80 Myr < t < 200 Myr .
These results are robust , since they are insensitive to the choice of the free-parameters regulating the extension of the extra-mixing region .