The extended main-sequence turn-off cluster NGC 1856: rotational evolution in a coeval stellar ensemble

Multiple or extended turn-offs in young clusters in the Magellanic Clouds have recently received large attention. A number of studies have shown that they may be interpreted as the result of a significant age spread (several 10(8) yr in clusters aged 1-2 Gyr), while others attribute them to a spread in stellar rotation. We focus on the cluster NGC 1856, showing a splitting in the upper part of the main sequence, well visible in the colour m(F336W) - m(F555W), and a very wide turn-off region. Using population synthesis available from the Geneva stellar models, we show that the cluster data can be interpreted as superposition of two main populations having the same age (similar to 350 Myr), composed for 2/3 of very rapidly rotating stars, defining the upper turn-off region and the redder main sequence, and for 1/3 of slowly/non-rotating stars. Since rapid rotation is a common property of the B-A type stars, the main question raised by this model concerns the origin of the slowly/non-rotating component. Binary synchronization is a possible process behind the slowly/non-rotating population; in this case, many slowly/nonrotating stars should still be part of binary systems with orbital periods in the range from 4 to 500 d. For these orbital periods, Roche lobe overflow occurs during the evolution of the primary off the main sequence, so most primaries may not be able to ignite core helium burning, consistently why the lack of a red clump progeny of the slowly rotating population.