| 1. |
- Malerba, L., et al.
(author)
-
Ab initio calculations and interatomic potentials for iron and iron alloys : Achievements within the Perfect Project
- 2010
-
In: Journal of Nuclear Materials. - : Elsevier BV. - 0022-3115 .- 1873-4820. ; 406:1, s. 7-18
-
Journal article (peer-reviewed)abstract
- The objective of the FP6 Perfect Project was to develop a first example of integrated multiscale computational models, capable of describing the effects of irradiation in nuclear reactor components, namely vessel and internals. The use of ab initio techniques to study, in the most reliable way currently possible, atomic-level interactions between species and defects, and the transfer of this knowledge to interatomic potentials, of use for large scale dynamic simulations, lie at the core of this effort. The target materials of the Project were bainitic steels (vessel) and austenitic steels (internals), i.e. iron alloys. In this article, the advances made within the Project in the understanding of defect properties in Fe alloys, by means of ab initio calculations, and in the development of interatomic potentials for Fe and Fe alloys are overviewed, thereby providing a reference basis for further progress in the field. Emphasis is put in showing how the produced data have enhanced our level of understanding of microstructural processes occurring under irradiation in model alloys and steels used in existing nuclear power plants.
|
|
| 2. |
- Castin, N., et al.
(author)
-
Advanced atomistic models for radiation damage in Fe-based alloys : Contributions and future perspectives from artificial neural networks
- 2018
-
In: Computational materials science. - : Elsevier. - 0927-0256 .- 1879-0801. ; 148, s. 116-130
-
Journal article (peer-reviewed)abstract
- Machine learning, and more specifically artificial neural networks (ANN), are powerful and flexible numerical tools that can lead to significant improvements in many materials modelling techniques. This paper provides a review of the efforts made so far to describe the effects of irradiation in Fe-based and W-based alloys, in a multiscale modelling framework. ANN were successfully used as innovative parametrization tools in these models, thereby greatly enhancing their physical accuracy and capability to accomplish increasingly challenging goals. In the provided examples, the main goal of ANN is to predict how the chemical complexity of local atomic configurations, and/or specific strain fields, influence the activation energy of selected thermally-activated events. This is most often a more efficient approach with respect to previous computationally heavy methods. In a future perspective, similar schemes can be potentially used to calculate other quantities than activation energies. They can thus transfer atomic-scale properties to higher-scale simulations, providing a proper bridging across scales, and hence contributing to the achievement of accurate and reliable multiscale models.
|
|
| 3. |
- Castin, N., et al.
(author)
-
Improved atomistic Monte Carlo models based on ab-initio -trained neural networks : Application to FeCu and FeCr alloys
- 2017
-
In: Physical Review B. - : American Physical Society. - 2469-9950 .- 2469-9969. ; 95:21
-
Journal article (peer-reviewed)abstract
- We significantly improve the physical models underlying atomistic Monte Carlo (MC) simulations, through the use of ab initio fitted high-dimensional neural network potentials (NNPs). In this way, we can incorporate energetics derived from density functional theory (DFT) in MC, and avoid using empirical potentials that are very challenging to design for complex alloys. We take significant steps forward from a recent work where artificial neural networks (ANNs), exclusively trained on DFT vacancy migration energies, were used to perform kinetic MC simulations of Cu precipitation in Fe. Here, a more extensive transfer of knowledge from DFT to our cohesive model is achieved via the fitting of NNPs, aimed at accurately mimicking the most important aspects of the ab initio predictions. Rigid-lattice potentials are designed to monitor the evolution during the simulation of the system energy, thus taking care of the thermodynamic aspects of the model. In addition, other ANNs are designed to evaluate the activation energies associated with the MC events (migration towards first-nearest-neighbor positions of single point defects), thereby providing an accurate kinetic modeling. Because our methodology inherently requires the calculation of a substantial amount of reference data, we design as well lattice-free potentials, aimed at replacing the very costly DFT method with an approximate, yet accurate and considerably more computationally efficient, potential. The binary FeCu and FeCr alloys are taken as sample applications considering the extensive literature covering these systems.
|
|
| 4. |
- Terentyev, D., et al.
(author)
-
Further development of large-scale atomistic modelling techniques for Fe-Cr alloys
- 2011
-
In: Journal of Nuclear Materials. - : Elsevier BV. - 0022-3115 .- 1873-4820. ; 409:2, s. 167-175
-
Journal article (peer-reviewed)abstract
- In this paper we review the current status of our efforts to model the Fe-Cr system, which is a model alloy for high-Cr ferritic-martensitic steels, using large-scale atomistic methods. The core of such methods are semi-empirical interatomic potentials. Here we discuss their performance with respect to the features that are important for an accurate description of radiation effects in Fe-Cr alloys. We describe their most recent improvements regarding macroscopic thermodynamic properties as well as microscopic point-defect properties. Furthermore we describe a new type of large-scale atomistic kinetic Monte Carlo (AKMC) approach driven by an artificial neural network (ANN) regression method to generate the local migration barrier for a defect accounting for the local chemistry around it. The results of the thermal annealing of the Fe-20Cr alloy modelled using this AKMC approach, parameterized by our newly developed potential, were found to be in very good agreement with experimental data. Furthermore the interaction of a 1/2 (1 1 1) screw dislocation with Cr precipitates as obtained from the AKMC simulations was studied using the same potential. In summary, we critically discuss our current achievements, findings and outline issues to be addressed in the near future development.
|
|
| 5. |
- Bonny, G., et al.
(author)
-
Numerical prediction of thermodynamic properties of iron-chromium alloys using semi-empirical cohesive models : The state of the art
- 2009
-
In: Journal of Nuclear Materials. - : Elsevier BV. - 0022-3115 .- 1873-4820. ; 385:2, s. 268-277
-
Journal article (peer-reviewed)abstract
- In this work the capability of existing cohesive models to predict the thermodynamicproperties of Fe-Cr alloys are critically evaluated and compared. The two-band model and the concentration-dependent model, which are independently developed extensions of the embedded-atom method, are demonstrated to be equivalent and equally capable of reproducing the thermodynamic properties of Fe-Cr alloys. The existing potentials fitted with these formalisms are discussed and compared with an existing cluster expansionmodel. The phase diagram corresponding to these models is evaluated using different but complementary methods. The influence of mixing enthalpy, low-energy states and vibrational entropy on the phase diagram is examined for the different cohesive models.
|
|
| 6. |
- Djurabekova, F., et al.
(author)
-
Kinetics versus thermodynamics in materials modeling : The case of the di-vacancy in iron
- 2010
-
In: Philosophical Magazine. - : Informa UK Limited. - 1478-6435 .- 1478-6443. ; 90:19, s. 2585-2595
-
Journal article (peer-reviewed)abstract
- Monte Carlo models are widely used for the study of microstructural and microchemical evolution of materials under irradiation. However, they often link explicitly the relevant activation energies to the energy difference between local equilibrium states. We provide a simple example (di-vacancy migration in iron) in which a rigorous activation energy calculation, by means of both empirical interatomic potentials and density functional theory methods, clearly shows that such a link is not granted, revealing a migration mechanism that a thermodynamics-linked activation energy model cannot predict. Such a mechanism is, however, fully consistent with thermodynamics. This example emphasizes the importance of basing Monte Carlo methods on models where the activation energies are rigorously calculated, rather than deduced from widespread heuristic equations.
|
|
