The first term on the right side of this expression is almost identical to the one
obtained by Mullins (1957, Eq.36) for a surface layer of infinite extent, using the
small slope approximation in addition to the assumption that the dihedral angles are in equilibrium configuration.
The В parameter, which is introduced by Mullins (1957) in his paper for scaling
purpose, is exactly equal to in our representation.
Mullins (1957) obtain a prefactor
compared to our value 4/5.