Carbon diffusion paths and segregation at high-angle tilt grain boundaries inα-Fe studied by using a kinetic activation-relation technique

Abstract

Carbon diffusion and segregation in iron is fundamental to steel production but is also associated with corrosion. Using the kinetic activation-relaxation technique (k-ART), a kinetic Monte Carlo (KMC) algorithm with an on-the-fly catalog that allows to obtain diffusion properties over large time scales taking into account long-range elastic effects coupled with an EAM force field, we study the motion of a carbon impurity in four Fe systems with high-angle grain boundaries (GB), focusing on the impact of these extended defects on the long-time diffusion of C. Short and long-time stability of the various GBs is first analyzed, which allows us to conclude that the 3(1 1 1)θ = 109.53 • 110 GB is unstable, with Fe migration barriers of ∼0.1 eV or less, and C acts as a pinning center. Focusing on three stable GBs, in all cases, these extended defects trap C in energy states lower than found in the crystal. Yet, contrary to general understanding, we show, through simulations extending to 0.1 s, that even tough C diffusion takes place predominantly in the GB, it is not necessarily faster than in the bulk and can even be slower by one to two orders of magnitude depending on the GB type. Analysis of the energy landscape provided by k-ART also shows that the free cavity volume around the impurity is not a strong predictor of diffusion barrier height. Overall, results show rather complex diffusion kinetics intimately dependent on the local environment.

Pressure effect on diffusion of carbon at the 85.91∘〈100〉 symmetric tilt grain boundary of α -iron

“…A negative value of segregation energy means that the grain boundary site is energetically more favorable than the bulk site and signals a general flux from the bulk to these sites at the GB. As demonstrated in our previous work, 9 the disorder associated with GB generates a variety of atomic environments, facilitating the formation of sites more energetically favorable to the presence of carbon than interstitial crystalline sites. Depending on the specific nature of the interface between the grains, these favorable sites could be connected to accelerate or to slow down carbon diffusion as compared to the bulk.…”

“…In the past, k-ART has been used in various systems to describe diffusion of point defects in metals and semiconductors and more complex systems with grain boundaries and amorphous silicon. 9,35,36 B. Handling flickering states KMC simulations can become inefficient in the presence of low energy barriers that dominate the kinetics, as each step is then associated with a very small clock increase.…”

“…In our earlier works we focused on carburisation by characterising reaction of iron with CO as well as the diffusion and segregation of carbon within the grain boundaries of the metal. [1][2][3][4][5][6][7][8][9][10] In this work we aim to turn our attention towards realistic scenarios that occur at various stages of dusting corrosion, drawing comparisons with experiments whenever possible, focusing more particularly on the effect of pressure on C diffusion in Fe, particularly at grain boundaries (GBs). The diffusion of carbon interstitial atoms around GBs in iron has long been an area of interest for researchers due to its impact on various mechanical properties of steel such as corrosion resistance, and embrittlement, but also at it serves as a rich system for modelling and simulation.…”

Pearlite in Multicomponent Steels: Phenomenological Steady-State Modeling

“…[15,40] Q B C is the activation energy for boundary diffusivity D Bk subtracted by the binding energy of C to austenite-pearlite interface. Hashiguchi and Kirkaldy made an attempt to separate the two parts, estimating the binding energy to be several tens of kJ mol À1 , [16] which means the activation energy of D Bk is even higher than that of kD Bk d À Á C However, recent atomistic simulation [41] shows that C diffusion along GB in pure Fe does not necessarily exhibit a lower activation energy than volume diffusion, the difference depending significantly on grain boundary structure.…”

Kinetic Monte Carlo Simulations of Irradiation Effects

“…a), b), c), d) and e) present the structure with one, two, three, four and five mono-self-defects for each case. From (Mahmoud 2018) Off-lattice KMC methods are even more important when strain and randomness play an important role, such as for characterizing the 100-interstitial loop formation in Fe (Xu et al 2013), following the relaxation of ion-bombarded silicon over long-time (more than 1 s) (Béland et al 2013) (Béland et al 2013) or understanding C diffusion at an Fe grain boundary, where pathways are strongly influenced by the local deformation (Restrepo et al 2018).…”