Experimental and theoretical evidence for carbon-vacancy binding in austenite

Abstract

Our research and results from the literature all consistently suggest a binding energy of nearest-neighbor carbon-vacancy (C-V) pairs of the order 35 to 40 kJ/mole in austenitic alloys. Results examined include point-defect anelasticity, self-diffusion, high-temperature creep, strain aging, strain-age hardening, radiation damage, and point-defect structure modeling. Increases in the height of carbon-based anelastic peaks by quenching, cold work, and electron irradiation are consistent with a substantial nonexclusive contribution of C-V complexes. Increased carbon content in austenite increases the iron self-diffusivity and the high-temperature creep rate of fcc Fe, implying a C-V binding energy of ϳ40 kJ/mol. Dynamic strain aging of carbon-containing austenites occurs in temperature ranges too low to involve interstitial solute mobility and requires an interpretation of large C-V binding wherein the vacancy is the more mobile component. Strengthening in heavily deformed austenitic stainless steels associated with strain aging or long-term aging near room temperature implies solution hardening by tetragonal-like C-V complexes formed at these temperatures. Results on radiation damage of austenitic steels show effects of carbon on irradiation susceptibility. Finally, we have performed first-principles gradient-corrected density functional calculations to determine the binding energy of nearest-neighbor C-V pairs in fcc iron; a value of ϳ35 kJ/mol is obtained.

First-principles study of helium, carbon, and nitrogen in austenite, dilute austenitic iron alloys, and nickel

“…This level of binding agrees well with previous experimental and theoretical work in austenite and austenitic alloys. 47 We find that V-N binding is significantly stronger than for C with binding energies in the range from 0.3 to 0.6 eV at 1 nn and around 0.1 eV at 2 nn. For both C and N, the substitutional configuration is strongly disfavored.…”

“…40 Calculations in austenite are, however, limited primarily to solute dissolution, diffusion, and their influence on the electronic structure, local environment, and stacking fault energies, 34,[41][42][43][44][45][46] although calculations of vacancy-C binding have been performed. 47 In this work we present a detailed study of the energetics, kinetics, and interactions of He, C, and N solutes in model austenite and austenitic systems using DFT. A full treatment of paramagnetic austenite and FeCrNi austenitic alloys would naturally take into account the magnetic and composition dependence of the variables under study, and while ab initio techniques are now becoming available to model the paramagnetic state [48][49][50][51] and calculations in concentrated alloys are certainly achievable, 52 their complexity precludes a broad study of all the necessary variables relevant for radiation damage modeling.…”

Diffusion of carbon in bcc Fe in the presence of Si

“…[11][12][13][14][15][16] Recently, it has become possible to compute solute-interaction energies, diffusion activation energies, and attempt frequencies from ab initio calculations. 1,[17][18][19][20] This provides a verification of earlier intuitive atomistic models of diffusion and soluteinteraction models, and it also provides an opportunity for a deeper understanding of the atomistic processes. Here, our aim is to gain a deeper understanding of the effect of Si on the diffusion of C in ferromagnetic bcc Fe.…”

“…Of particular importance is the diffusion of carbon in steel, both in the metallic matrix [1][2][3] and within precipitated carbide phases 4 as recent computer simulations and experiments indicate. It is known that diffusion of carbon, and other interstitial species, is strongly affected by the presence of other alloying elements.…”

Ab initio моделирование энергии растворения и активности углерода в ГЦК-Fe

“…Энергия этого процесса была вычислена экспериментально в работах [12][13][14][15][16], на основе анализа которых можно сделать вывод, что ее величина лежит в интервале от 0.36 до 0.48 eV. Некоторые авторы ис-пользовали НМ-состояния для моделирования парамаг-нитного состояния, поскольку считали, что в парамаг-нитном состоянии магнитный момент равняется нулю и его можно использовать для моделирования парамаг-нитного состояния [17,18]. Ферромагнитное состояние разделяется на два магнитных состояния с большим и маленьким спином: ФМВС и ФМНС.…”