Investigation of Local Hydrogen Uptake in Rescaled Model Occluded Sites Using Crevice Scaling Laws

Wednesday, 4 October 2017: 14:10
Camellia 2 (Gaylord National Resort and Convention Center)
J. R. Scully (University of Virginia), J. S. Lee (Naval Research Laboratory), J. Locke (The Ohio State University), M. A. Switzer (University of Virginia), P. Lu, and J. Saal (QuesTek Innovations, LLC)
A major challenge in the quantification of hydrogen (H) assisted cracking is the determination of local H contents in pits and cracks and subsequently the environment assisted cracking behavior as a function of H concentration for the given metallurgical condition. Moreover, the lack of insight on these issues makes it difficult to computational design improved alloys for resistance to H assisted cracking. One difficulty in understanding H uptake at cracks and pits stems from the occluded chemistry and the local current and potential distribution due to ohmic potential drop.

The effects of occluded site geometry and applied potential leading to depassivation/activation, and potential drop on H production and uptake in a martensitic precipitation/age hardened stainless steel PH 13-8Mo (Fe-13Cr-8Ni-2Mo-1Al) were explored. A potential (E)-pH diagram for PH 13-8Mo was calculated using the CALPHAD approach which illustrates the H/H+ equilibrium line and the oxide stability regions. On planar electrode surfaces, the total H concentration was found to increase exponentially with H overpotential in simulated acidified pit solutions. Comparison of CALPHAD and experimental findings indicated that the potential and pH range expected in acidified pit bottoms that H production would be thermodynamically possible on metal surfaces without a stable oxide film. This data could later be used to quantify H uptake in pits as a function of potential and pH.

The x2/gap scaling law, where x is the pit/crevice depth and “gap” is the pit/crevice width, was computationally determined and used to rescale model pits from micrometer to millimeter dimensions using E-log current density (i) data as a boundary condition for finite element analysis. Such rescaling enabled local H measurements by thermal desorption as a function of pit depth in rescaled pits. Two values of critical depth (xcrit) were identified. Significant local H uptake was observed at x > xcrit under conditions where external surfaces were in a passive state and above the H electrode potential (EH/H+). The local potential drops below EH/H+ at a depth, xcrit1= xHER and reaches the primary passivation potential (EPass) of the stainless steel at a depth, xcrit2 = xPass. Thus, at x>xcrit1, the stainless steel experiences H production and uptake at/near the pit bottom but remains passivated. At x>xcrit2, H absorbs on actively dissolving pit surfaces since the local potential drops below EPass. Concurrent metal dissolution and H uptake lead to significant amounts of local H absorption at simulated pit surfaces. The ability of this rescaled approach to provide insight on the development of crack tip conditions that negatively impact environment assisted fracture are further discussed.

Acknowledgments: This work was supported in part as part of the Center for Performance and Design of Nuclear Waste Forms and Containers, an Energy Frontier Research Center funded by the U.S. Department of Energy, Office of Science, Basic Energy Sciences under Award # DE-SC0016584. Finite element analysis, CALPHAD calculation and assessments of environment assisted cracking were conducted under the auspices of this grant while corrosion electrochemistry and hydrogen analysis was conducted under funding from NSF.