Lithium depletion is a strong test of core-envelope recoupling
Rotational mixing is a prime candidate for explaining the gradual depletion of lithium from the photospheres of cool stars during the main sequence. However, previous mixing calculations have relied primarily on treatments of angular momentum transport in stellar interiors incompatible with solar and stellar data, in the sense that they overestimate internal differential rotation. Instead, recent studies suggest that stars are strongly differentially rotating at young ages, but approach solid body rotation during their lifetimes. We modify our rotating stellar evolution code to include an additional source of angular momentum transport, a necessary ingredient for explaining the open cluster rotation pattern, and examine the consequences for mixing. We confirm that core-envelope recoupling with a $\sim$20 Myr timescale is required to explain the evolution of the mean solar-mass rotation pattern along the main sequence, and demonstrate that it also provides a more accurate description of the Li depletion pattern seen in open clusters. Recoupling produces a characteristic pattern of efficient mixing at early ages and little mixing at late ages, thus predicting a flattening of Li depletion at a few Gyr, in agreement with the observed late-time evolution. Using Li abundances, we argue that the timescale for core-envelope recoupling during the main sequence decreases sharply with increasing mass. We discuss implications of this finding for stellar physics, including the viability of gravity waves and magnetic fields as agents of angular momentum transport. We also raise the possibility of intrinsic differences in initial conditions in star clusters, using M67 as an example.