Hydrogen Damage - Environmental Degradation of Metals - page 459

where

σ applied stress

P hydrogen pressure in voids

n number of dislocations comprising cracks

b Burger's vector

γ s and γ p surface energy and plastic work term, respectively.

Since γ s , any reduction in the latter through adsorp-

tion would not change the energy requirement for crack propagation significantly.
Moreover, the discontinuous crack growth that has been observed for hydrogen
cracking is not explained by this mechanism.

γ p is considerably greater than

8.4.3 Decohesion Theory
The theory has been originally proposed by Troiano [12] and subsequently devel-
oped by Oriani [13] and others. The theory envisages that the effect of hydrogen
is to lower the cohesive strength of the lattice. Hydrogen distributed uniformly
throughout a metal lattice is nondamaging because its concentration is so small.
Under the influence of an applied stress, hydrogen would diffuse to the regions
of high triaxial stress provided by a crack tip or a network of internal voids. It
has been suggested that the electrons from the hydrogen atoms would enter the
d bands of the metallic cores and the increase of the electron concentration of
these bands produces an increase in repulsive forces between metallic cores, i.e.,
a decrease in the cohesive strength of the lattice. Since a crack would propagate
in a brittle manner if the ratio of the largest tensile fracture stress,

σ , at the crack

tip to the largest shear stress, , on slip phases intersecting crack tips was greater

than the ratio of the ideal cleavage stress,

τ

σ max , to the ideal shear stress, τ max , the

brittle fracture would be preferred by low value of τ max .

The mechanism envisages the fracture as a perfect cleavage. However, by the
Tresca criterion the group of atoms in a small element should slip apart, rather
than separate in a tensile mode, because the stress required for shear is one-half
that required for tensile fracture. The same considerations hold good at the tip
of a growing crack and slip should occur at a lower stress than tensile or cleavage
separation. Thus, some form of plastically based separation ought to occur at the
tip of a growing cleavage crack. The fractographic evidence of plastic flow even
in high-strength steels strengthens the postulate of shear separation. It has been
suggested that in such cases the decohesion is facilitated by hydrogen and there
is plastic tearing of ligaments unaffected by hydrogen between decohered regions.
There have been suggestions [14,15] that plastic flow around crack tips or the
stress concentrations produced by dislocation pile-ups near crack tip can indi-
rectly influence the decohesion process.

σ max /