PMOS can be easier to control (no need for a precharged
bootstrap supply, no limitations on duty cycle to charge
same, etc.). PMOS can be driven from within the rails,
on the high side; NMOS needs a supply above the rail.
NMOS will also have a leak-back issue when Vout > Vin,
the drain-body diode would be forward biased. That
"leak" could be a gusher. NMOS is more appropriate to
a buck high side (with enough design constraint /
support) because Vout < Vin always.
But before you get all excited, just how significant is
the diode Vf drop, in a 60V system? Not very. A decent
Schottky probably puts the conduction loss at about 1%.
Lot of effort and die area in making a sync flying switch,
for a fractional-% improvement, and likely you throw a
bunch of the conduction loss onto switching losses
rather than simply eliminating it.