Solute diffusion calculation in Fe-Si and Fe-Cr-Si multicomponent alloys
2024
Read Solute diffusion calculation in Fe-Si and Fe-Cr-Si multicomponent alloys (2024) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2411-02053.txt.
Description
Abstract Authors \text{{\bf Miryashkin Т. N.}}^{1*} , \text{{Novoselov I. I.}}^{2} , \text{{Yanilkin A. V.}}^{2,1} 1 - Moscow Institute of Physics and Technology, Institutskiy Pereulok 9, Dolgoprudny, Moscow Region 141700, Russia 2 - Dukhov Automatics Research Institute (VNIIA), Sushchevskaya 22, Moscow 127055, Russia *[email protected]; Keywords Solute diffusion, bond potential, low-rank potential, correlation factor.
Catalog record
| YEAR | 2024 |
|---|---|
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 27 KB |
| VERSION | 1.0 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2411-02053
Browse the full library · About the Tower of Babel
Full text
The opening of the work, from the library's hosted text:
Solute diffusion calculation in Fe-Si and Fe-Cr-Si multicomponent alloys Abstract Authors \text{{\bf Miryashkin Т. N.}}^{1*} , \text{{Novoselov I. I.}}^{2} , \text{{Yanilkin A. V.}}^{2,1} 1 - Moscow Institute of Physics and Technology, Institutskiy Pereulok 9, Dolgoprudny, Moscow Region 141700, Russia 2 - Dukhov Automatics Research Institute (VNIIA), Sushchevskaya 22, Moscow 127055, Russia *[email protected]; Keywords Solute diffusion, bond potential, low-rank potential, correlation factor. Abstract Diffusion plays a key role in microstructure evolution at multicomponent alloys: diffusion controls the kinetics of phase transformations and alloy homogenization. This study aims at developing computationally efficient approaches to estimate the solute diffusion coefficients in two-component systems. We consider silicon as the solute example because it is highly used in industrial steels. We demonstrate that the silicon jump frequency may be calculated with the bond potential instead of the more computationally expensive machine learning potential in Fe-Si and Fe-Cr-Si alloys. We show that the silicon jump frequency can be estimated from thermodynamic simulations for the bond potential without kinetic simulations. The silicon correlation factor slightly depends on silicon concentration and can be approximately estimated by the analytical nine-frequency model. 1 Introduction Industry extensively uses steel as a construction material. Steel consists of an iron matrix and various alloying elements; alloying elements give specific properties to steel: hardness, durability, corrosion resistance. Diffusion of the alloying elements plays a crucial role in the kinetics of many microstructural changes: new phases nucleation, phase transformations, alloys homogenization [ 1 , 2 , 3 ] . The experimental studies of solute diffusion in materials are resource-intensive and time-consuming. The development of computational methods and analytical models that estimate diffusion helps to efficiently design steel with specific properties for various industrial needs. The analytical models can be used to estimate the solute diffusion in infinitely diluted binary alloys [ 4 , 5 , 6 ] . These models require binding energy and migration energy, which can be estimated from the ab-initio simulations [ 7 , 8 , 9 , 10 ] or from the experiments [ 11 , 12 ] . In real work steels, the typical concentrations of the alloying element reach up to 1 - 1.5 at. % [ 13 ] ; thus, we are required to account for the corresponding effects that interactions of the solute atoms produce [ 6 ] . The Darken and Manning theories describe how the solute diffusion changes when the solute concentration is increased [ 14 , 15 ] and account for the contributions from thermodynamic and kinetic factors. These theories were thoroughly analyzed and compared to atomic kinetic Monte Carlo simulations with pairwise interaction potential in [ 16 ] . The main drawback is that these theories do not consider changes in the hopping barriers due to the increasing concentration of solute atoms, which lead to exponential changes in diffusion constants and cannot be neglected in the full analysis. The computational approaches estimate the diffusion coefficients more accurately than the approximate analytical models. The major atomistic modeling methods that are used to study diffusion include the ab initio molecular dynamics, classical molecular dynamics, Monte Carlo methods. Ab initio molecular dynamics accurately reproduces energy and forces in the system but is computationally expensive [ 17 , 18 ] . Classical molecular dynamics requires interatomic potentials. Semiempirical potentials cannot provide the desired accuracy for describing multicomponent systems, while machine learning interatomic potentials need a tremendous amount of configurations for training [ 19 , 20 , 21 , 22 ] . Moreover, the total simulation time for classical molecular dynamics is typically limited to several microseconds as the timesteps should be of \sim 10^{-15} s to accurately integrate the equations of motion. Interatomic potentials for the Monte Carlo simulations are constructed using sufficiently fewer configurations [ 23 ] . The kinetic Monte Carlo method may be used for several orders longer simulations compared to the molecular dynamics [ 24 , 25 ] . This paper aims to develop computationally efficient methods to estimate solute diffusion in two-component alloys on the example of the Fe-Si system. Firstly, the current work investigates the application of the bond potential instead of the more expensive machine learning potential. Secondly, we explore how to estimate the solute diffusion coefficient from thermodynamic modeling without requiring additional kinetic simulations. We use the kinetic Monte Carlo method to conduct kinetic simulations and to sample equilibrium configuration for the thermodynamics computations. We investigate Fe-Si (and Fe-Cr-Si) system at 800 K, where the iron is ferromagnetic; the Curie temperature for iron is 1043 К [ 26 ] . We select Si as an example of alloying element because industry uses Si to increase corrosion resistance and to strengthen the steel. The selected temperature is particularly important as steels in nuclear reactors operate under this temperature [ 27 ] . Understanding the physics at this temperature is crucial to guarantee the safety of power plants. We are unaware of experiments for silicon diffusion in Fe-Si (and Fe-Cr-Si) system for the ferromagnetic phase of Fe. The experimental works we found investigate the higher temperatures, where the iron is paramagnetic [ 28 , 29 , 30 ] . We do not extrapolate the results of these experiments to lower temperatures, as the phase transition of Fe arguably changes the physics of the diffusion. Future experiments and computational works for lower temperatures should clarify the physics of ferromagnetic iron. We organize the article in the following way. We describe theoretical methods and computational details in Section 2 . We present the results for the silicon jump frequency and correlation factor in Fe-Si alloy in Section 3 . We summarize our results in Section 4 . In the Appendix, we investigate silicon jump frequency and correlation factor in Fe-Cr-Si alloy with Cr concentration up to 12 at.%. 2 Computational Details 2.1 Solute Diffusion Solute diffusion coefficient ( D_{\rm sol} ) in cubic lattices is expressed by the formula: D(C_{\rm sol})=\frac{\Delta^{2}}{6}f({C_{\rm sol}})\,\nu({C_{\rm sol}})\,C_{V}({C_{\rm sol}}), (1) where \Delta is the distance between the nearest neighbors, f is the correlation factor, \nu is the solute jump frequency, C_{V} is the vacancy concentration, C_{\rm sol} is the solute concentration. We consider the vacancy diffusion mechanism as existing research shows it is a dominant mechanism for steels [ 31 , 6 ] . The present article considers the solute jump frequency and correlation factor. We describe how one can calculate the solute jump frequency from the thermodynamic simulations without requiring direct kinetic simulations for the bond potential (Section 2.3 ). In the Results (Section 3 ), we estimate the correlation factor from the analytical models [ 32 , 5 , 4 ] and show how it depends on solute concentration. Calculation of vacancy concentration from thermodynamic Monte Carlo simulations is established routine and can be found in the articles [ 33 , 34 ] . 2.2 Interatomic Potentials We select the low-rank potential [ 23 , 35 ] and bond potential [ 36 ] as the interaction models in our computations. Kinetic Monte Carlo simulations with the low-rank potential serve as reference simulations, as the authors in [ 37 ] demonstrate that the low-rank potential is reliable for diffusion calculations. We propose the bond potential instead of the low-rank potential, as the bond potential is more computationally