Desorption vs reflection (or sticking coefficients) in thermal physical vapor deposition (PVD) of gold

I’m having some issues squaring competing concepts behind the kinetics of adatom deposition in thermal PVD. Mattox (Handbook of Physical Vapor Deposition) repeatedly states Au atoms vaporized via thermal PVD have an approximate energy of 3 eV, of which only 0.3 eV is kinetic energy, therefore far too low to subsequently desorb when Au metallophilic bonding is around 13 ev. However, publications like Chandra and Scott (link) show that Au has a sticking coefficient of .9376, Bachmann, 0.89 at <10A (link).

How can both be true? Given the energies involved, how does an impinging adatom not simply adsorb? What other factors affect whether or not an adatom adsorbs?

Thanks in advance.