Stress corrosion cracking studied with grain mapping by diffraction contrast tomography

08-09-2008

A new grain mapping technique developed by researchers at the ESRF has been used to study stress corrosion cracking in stainless steels.

  • Share

Intergranular stress corrosion cracking is a failure mechanism that can affect stainless steels.  It is characterised by the nucleation and growth of cracks by localised corrosion at sensitised grain boundaries in the presence of an applied stress.  Researchers at the ESRF used diffraction contrast tomography to map the grain structure of a stainless steel sample in 3D, and then observed stress corrosion cracking in situ using microtomography.  This revealed the crack path between the grains, and showed which special grain boundaries are resistant to cracking.

Stress corrosion cracking can be a critical failure mechanism in some stainless steel components of power stations [1] when there is a combination of a susceptible material, an aggressive environment, and a mechanical driving force.  Steels can become sensitised by heat treatments such as post weld stress relief, or by fast neutron irradiation in nuclear plants.  In both cases, sensitisation decreases the local resistance of boundaries to corrosion due to chromium depletion at the boundaries, although for irradiation damage other factors may also be involved [2-4].  Sensitisation resistance is affected by the grain boundary structure, related to the crystallographic orientation of the adjoining grains and the plane of the boundary.  Certain special grain boundaries are resistant to sensitisation, and previous in situ observations made at the ESRF have shown that these grain boundaries can form bridges across the growing crack, which are thought to retard its progress [5,6].

To understand the behaviour of the growing crack requires knowledge of the grain boundary properties in 3D throughout the bulk of the sample. The newly developed 3D grain orientation mapping techniques, such as diffraction contrast tomography (DCT) at beamline ID19 [7], or the ID11 3D X-ray microscope project [8] can be used to study such a system.  In this case DCT, developed at the ESRF by researchers from Manchester University and INSA de Lyon, was used to map the grain structure of a stainless steel wire, 0.4 mm in diameter.  The grain map consisted of some 360 grains, with a network of 1600 fully-characterised boundaries between them.

Renderings showing four steps during the propagation of a stress corrosion crack. Bridges across the crack are seen as holes in the crack surface.

Figure 1. Renderings showing four steps during the propagation of a stress corrosion crack.  Bridges across the crack are seen as "holes" in the crack surface.

After mapping the grain structure, the sample was placed in a corrosive environment and loaded to initiate stress corrosion cracking.  At intervals during the growth of the crack, the load was removed and the sample scanned using microtomography to generate 3D images of the growing crack, revealing the crack path along the grain boundaries.  By studying these 3D images it is possible to identify those grain boundaries that resist cracking, and instead form bridging ligaments (Figure 1).  The geometry of  these boundaries is known from the grain map (Figure 2), revealing the types of boundary that resist sensitisation.  It was found that not all of the special boundaries were the expected low Σ coincident site lattice boundaries, and furthermore they were not the Σn (n=1,2,3) twin variant boundaries that grain boundary engineering usually seeks to increase [9].

Rendering of the grain map (randomly coloured) and crack (red). The crack can be seen to propagate along the grain boundaries.

Figure 2. Rendering of the grain map (randomly coloured) and crack (red). The crack can be seen to propagate along the grain boundaries.

This type of 3D characterisation and in situ observation allows a more complete understanding of the mechanisms of stress corrosion cracking, and the grain boundaries structures that resist cracking.  A better understanding of the process will allow improved modelling of cracking in existing materials, and will also provide input to the design of new materials that will better resist this form of degradation.

References
[1] P.M. Scott, Corrosion 56, 771 (2000).
[2] S.M. Bruemmer, G.S. Was, J. Nuclear Materials 216, 348 (1994).
[3] P. Scott, J. Nuclear Materials 211, 101 (1994).
[4] J.T. Busby, G.S. Was, and E.A. Kenik, J. of Nuclear Materials 302, 20 (2002).
[5] L. Babout, T.J. Marrow, D. Engelberg, P.J. Withers, Materials Science and Technology 22, 1068 (2006).
[6] A.P. Jivkov, N.P.C. Stevens, T.J. Marrow, Computational Materials Science 38, 442 (2006).
[7] G. Johnson, A. King, M. Gonzalves Honnicke, T.J. Marrow and W. Ludwig, J. Appl. Cryst. 41, 310 (2008).
[8] H.F. Poulsen, Three-dimensional X-ray diffraction microscopy. Mapping polycrystals and their dynamics. Springer Tracts in Modern Physics, Springer, Berlin (2004).
[9] V. Randle, Acta Mater. 46, 1459 (1998).

Principal publication and authors
A. King (a,b), G. Johnson (a,b), D. Engelberg (a), W. Ludwig (b,c) and J. Marrow (a), Science 321, 382 (2008).
(a) School of Materials, University of Manchester (UK)
(b) ESRF
(c) Institut National des Sciences Appliquées de Lyon, Villeurbanne (France)