Page 420 - Failure Analysis Case Studies II

404

can not be removed completely once the galvanising treatment has been carried out [24]. The
baking operation may not have removed all the H, particularly the molecular H which caused the
pores in the microstructure. Therefore, the baking operation has to be carried out immediately
after pickling, before galvanising, in order to avoid entrapment of H by defects present in the
material. The profusion of pores observed during microstructural examination (Fig. 3) indicates
that molecular H has evolved in the material. The fracture surfaces showing dimples in tensile tests
should not be taken as a signature of ductile fracture. The pores are in fact responsible for the
presence of dimples on the fracture surface and reduction in the ductility of the material. It is
reported in the literature that H assisted cracking can also occur by microvoid coalescence alone
and H affects the rate of nucleation of voids and reduces plasticity at later stages of fracture [25].
The split wire surface, however, does not show dimples but shows major features of delamination/
decohesion type of fracture with interspersed hard particles (cementite) and longitudinal cracks.
These defects are effective during bending and twisting, but not during tensile testing, and lead to
a delamination or decohesion type of cracking. From the microstructural features (Fig. 4(b)), it
appears that the delamination has taken place either along the prior austenite grain boundaries,
Le., along the interface of the pearlite colonies or along ferrite/cementite interfaces, which have
been weakened by the presence of H. In the case of tensile tests these defects do not play a role as
they are parallel to the tensile axis of the test sample.
The reason for failures of the wire rods has thus been attributed to hydrogen induced cracking.
After the H has been trapped inside the material due to improper pickling and baking operations,
it has initiated the cracking in the presence of a high tensile residual stress at the axis of the wire
and led to a delamination type failure. During coiling or twisting, plastic bending is induced in a
plane which, along with residual stress, causes the wires to fail by splitting along the weakest plane
i.e., the direction of drawing.

6. Concluding remarks
Based on the investigations carried out, and the observations thereon, it appears that the cohesive
strength of the boundaries/interfaces was reduced by the presence of atomic H. This ultimately led
to abrupt splitting during subsequent twisting and bending because the symmetric residual stress
pattern developed during the drawing operation is relieved. It appears that a proper inhibitor
during the pickling process can prevent ingress of H into the wire rods. Further, the baking
operation after pickling can be optimised to avoid interaction of H with the materials leading to
H assisted cracking.

Acknowledgements

The authors thankfully acknowledge useful discussions with Dr A. Biswas (UMIL, Ranchi), Dr
R. N. Ghosh (NML, Jamshedpur) and Dr D. K. Bhattacharya (NML, Jamshedpur). Appreciation
is also due to Mr S. Das (NML, Jamshedpur) for his help in conducting the SEM examinations.
The authors are grateful to Prof. P. Ramachandra Rao, Director, NML, for his support and
encouragement to publish this work.