-
Targeting the partition function of chemically disordered materials with a generative approach based on inverse variational autoencoders
Authors:
Maciej J. Karcz,
Luca Messina,
Eiji Kawasaki,
Emeric Bourasseau
Abstract:
Computing atomic-scale properties of chemically disordered materials requires an efficient exploration of their vast configuration space. Traditional approaches such as Monte Carlo or Special Quasirandom Structures either entail sampling an excessive amount of configurations or do not ensure that the configuration space has been properly covered. In this work, we propose a novel approach where gen…
▽ More
Computing atomic-scale properties of chemically disordered materials requires an efficient exploration of their vast configuration space. Traditional approaches such as Monte Carlo or Special Quasirandom Structures either entail sampling an excessive amount of configurations or do not ensure that the configuration space has been properly covered. In this work, we propose a novel approach where generative machine learning is used to yield a representative set of configurations for accurate property evaluation and provide accurate estimations of atomic-scale properties with minimal computational cost. Our method employs a specific type of variational autoencoder with inverse roles for the encoder and decoder, enabling the application of an unsupervised active learning scheme that does not require any initial training database. The model iteratively generates configuration batches, whose properties are computed with conventional atomic-scale methods. These results are then fed back into the model to estimate the partition function, repeating the process until convergence. We illustrate our approach by computing point-defect formation energies and concentrations in (U, Pu)O2 mixed-oxide fuels. In addition, the ML model provides valuable insights into the physical factors influencing the target property. Our method is generally applicable to explore other properties, such as atomic-scale diffusion coefficients, in ideally or non-ideally disordered materials like high-entropy alloys.
△ Less
Submitted 10 September, 2024; v1 submitted 27 August, 2024;
originally announced August 2024.
-
Semi-supervised generative approach to point-defect formation in chemically disordered compounds: application to uranium-plutonium mixed oxides
Authors:
Maciej J. Karcz,
Luca Messina,
Eiji Kawasaki,
Serenah Rajaonson,
Didier Bathellier,
Emeric Bourasseau
Abstract:
Machine-learning methods are nowadays of common use in the field of material science. For example, they can aid in optimizing the physicochemical properties of new materials, or help in the characterization of highly complex chemical compounds. An especially challenging problem arises in the modeling of chemically disordered solid solutions, for which some properties depend on the distribution of…
▽ More
Machine-learning methods are nowadays of common use in the field of material science. For example, they can aid in optimizing the physicochemical properties of new materials, or help in the characterization of highly complex chemical compounds. An especially challenging problem arises in the modeling of chemically disordered solid solutions, for which some properties depend on the distribution of chemical species in the crystal lattice. This is the case of defect properties of uranium-plutonium mixed oxides nuclear fuels. The number of possible configurations is so high that the problem becomes intractable if treated with direct sampling. We thus propose a machine learning approach, based on generative modeling, to optimize the exploration of this large configuration space. A probabilistic, semi-supervised approach using Mixture Density Network is applied to estimate the concentration of thermal defects in (U, Pu)O2. We show that this method, based on the prediction of the density of states of formation energy of a defect, is computationally much more cost-efficient compared to other approaches available in the literature.
△ Less
Submitted 28 November, 2022; v1 submitted 22 November, 2022;
originally announced November 2022.
-
Effect of cationic chemical disorder on defect formation energies in uranium-plutonium mixed oxides
Authors:
Didier Bathellier,
Luca Messina,
Michel Freyss,
Marjorie Bertolus,
Thomas Schuler,
Maylise Nastar,
Pär Olsson,
Emeric Bourasseau
Abstract:
At the atomic scale, uranium-plutonium mixed oxides (U,Pu)O_2 are characterized by cationic chemical disorder, which entails that U and Pu cations are randomly distributed on the cation sublattice. In the present work, we study the impact of disorder on point-defect formation energies in (U,Pu)O_2 using interatomic-potential and Density Functional Theory (DFT+U) calculations. We focus on bound Sch…
▽ More
At the atomic scale, uranium-plutonium mixed oxides (U,Pu)O_2 are characterized by cationic chemical disorder, which entails that U and Pu cations are randomly distributed on the cation sublattice. In the present work, we study the impact of disorder on point-defect formation energies in (U,Pu)O_2 using interatomic-potential and Density Functional Theory (DFT+U) calculations. We focus on bound Schottky defects (BSD) that are among the most stable defects in these oxides. As a first step, we estimate the distance R_D around the BSD up to which the local chemical environment significantly affects their formation energy. To this end, we propose an original procedure in which the formation energy is computed for several supercells at varying levels of disorder. We conclude that the first three cation shells around the BSD have a non-negligible influence on their formation energy (R_{D} = 7.0 \{AA}). We apply then a systematic approach to compute the BSD formation energies for all the possible cation configurations on the first and second nearest neighbor shells around the BSD. We show that the formation energy can range in an interval of 0.97 eV, depending on the relative amount of U and Pu neighboring cations. Based on these results, we propose an interaction model that describes the effect of nominal and local composition on the BSD formation energy. Finally, the DFT+U benchmark calculations show a satisfactory agreement for configurations characterized by a U-rich local environment, and a larger mismatch in the case of a Pu-rich one. In summary, this work provides valuable insights on the properties of BSD defects in (U,Pu)O_2, and can represent a valid strategy to study point defect properties in disordered compounds.
△ Less
Submitted 27 June, 2022;
originally announced June 2022.
-
Multiscale modelling in nuclear ferritic steels: from nano-sized defects to embrittlement
Authors:
N. Castin,
G. Bonny,
M. J. Konstantinović,
A. Bakaev,
F. Bergner,
C. Courilleau,
C. Domain,
B. Gómez-Ferrer,
J. M. Hyde,
L. Messina,
G. Monnet,
M. I. Pascuet,
B. Radiguet,
M. Serrano,
L. Malerba
Abstract:
Radiation-induced embrittlement of nuclear steels is one of the main limiting factors for safe long-term operation of nuclear power plants. In support of accurate and safe reactor pressure vessel (RPV) lifetime assessments, we developed a physics-based model that predicts RPV steel hardening and subsequent embrittlement as a consequence of the formation of nano-sized clusters of minor alloying ele…
▽ More
Radiation-induced embrittlement of nuclear steels is one of the main limiting factors for safe long-term operation of nuclear power plants. In support of accurate and safe reactor pressure vessel (RPV) lifetime assessments, we developed a physics-based model that predicts RPV steel hardening and subsequent embrittlement as a consequence of the formation of nano-sized clusters of minor alloying elements. This model is shown to provide reliable assessments of embrittlement for a very wide range of materials, with higher accuracy than industrial correlations. The core of our model is a multiscale modelling tool that predicts the kinetics of solute clustering, given the steel chemical composition and its irradiation conditions. It is based on the observation that the formation of solute clusters ensues from atomic transport driven by radiation-induced mechanisms, differently from classical nucleation-and-growth theories. We then show that the predicted information about solute clustering can be translated into a reliable estimate for radiation-induced embrittlement, via standard hardening laws based on the dispersed barrier model. We demonstrate the validity of our approach by applying it to hundreds of nuclear reactors vessels from all over the world.
△ Less
Submitted 25 April, 2022;
originally announced April 2022.
-
First principles model for voids decorated by transmutation solutes: Short-range order effects and application to neutron irradiated tungsten
Authors:
Duc Nguyen-Manh,
Jan S. Wrobel,
Michael Klimenkov,
Matthew J. Lloyd,
Luca Messina,
Sergei L. Dudarev
Abstract:
Understanding how properties of materials change due to nuclear transmutations is a major challenge for the design of structural components for a fusion power plant. In this study, by combining a first-principles matrix Hamiltonian approach with thermodynamic integration we investigate quasi-steady state configurations of multi-component alloys, containing defects, over a broad range of temperatur…
▽ More
Understanding how properties of materials change due to nuclear transmutations is a major challenge for the design of structural components for a fusion power plant. In this study, by combining a first-principles matrix Hamiltonian approach with thermodynamic integration we investigate quasi-steady state configurations of multi-component alloys, containing defects, over a broad range of temperature and composition. The model enables simulating transmutation-induced segregation effects in materials, including tungsten where the phenomenon is strongly pronounced. Finite-temperature analysis shows that voids are decorated by Re and Os, but there is no decoration by tantalum (Ta). The difference between the elements is correlated with the sign of the short range order (SRO) parameter between impurity and vacancy species, in agreement with Atom Probe Tomography (APT) observations of irradiated W-Re, W-Os, W-Ta alloys in the solid solution limit. Statistical analyses of Re and Os impurities in vacancy-rich tungsten show that the SRO effects involving the two solutes are highly sensitive to the background concentration the species. In quaternary W-Re-Os-Vac alloys containing 1.5% Re and 0.1% Os, the SRO Re-Os parameter is negative at 1200K, driving the formation of concentrated Re and Os precipitates. Comparison with experimental Transmission Electron Microscopy (TEM) and APT data on W samples irradiated at the High Flux Reactor (HFR) shows that the model explains the origin of anomalous segregation of transmutation products (Re,Os) to vacancy clusters and voids in the high temperature limit pertinent to the operating conditions of a fusion power plant.
△ Less
Submitted 2 February, 2021;
originally announced February 2021.
-
Adsorption of a fabric conditioner on cellulose nanocrystals: Synergistic effects of surfactant vesicles and polysaccharides on softness properties
Authors:
Evdokia K. Oikonomou,
Grazia M. L. Messina,
Laurent Heux,
Giovanni Marletta,
Jean-François Berret
Abstract:
Concentrated fabric conditioners are water-based formulations containing 10 wt. % of cationic surfactants that are deposited on textile fibers during the rinse cycle in a washing machine to make them smoother and softer to touch. In topical formulations, the concentration of cationic surfactants is reduced by half, this reduction being compensated by the addition of environment-friendly polysaccha…
▽ More
Concentrated fabric conditioners are water-based formulations containing 10 wt. % of cationic surfactants that are deposited on textile fibers during the rinse cycle in a washing machine to make them smoother and softer to touch. In topical formulations, the concentration of cationic surfactants is reduced by half, this reduction being compensated by the addition of environment-friendly polysaccharides. Using atomic force microscopy (AFM), quartz crystal microbalance with dissipation (QCM-D) and ellipsometry, the deposition of formulations with or without polysaccharides on model cellulose substrates is studied. We found that 180 nm long cellulose nanocrystals (CNCs) deposited by spin-coating on amorphous silicon dioxide or on quartz crystal sensors provide a good model of cellulose surfaces. QCM-D results reveal strong electrostatic interactions of the surfactant vesicles and polysaccharides with the CNC layer. In the presence of polysaccharides, the adsorbed quantities are less (by 60%) than the sum of the respective amounts of each component, the structure of the interface being however more homogenous and rigid. This outcome suggests that surface techniques coupled with CNC coated substrates are promising for studying interactions of current formulations with cellulose surfaces.
△ Less
Submitted 29 December, 2020;
originally announced December 2020.
-
Solute-point defect interactions, coupled diffusion, and radiation induced segregation in fcc nickel
Authors:
E. Toijer,
L. Messina,
C. Domain,
J. Vidal,
C. S. Becquart,
P. Olsson
Abstract:
Radiation-induced segregation (RIS) of solutes in materials exposed to irradiation is a well-known problem. It affects the life-time of nuclear reactor core components by favouring radiation-induced degradation phenomena such as hardening and embrittlement. In this work, RIS tendencies in face-centered cubic (fcc) Ni-X (X = Cr, Fe, Ti, Mn, Si, P) dilute binary alloys are examined. The goal is to i…
▽ More
Radiation-induced segregation (RIS) of solutes in materials exposed to irradiation is a well-known problem. It affects the life-time of nuclear reactor core components by favouring radiation-induced degradation phenomena such as hardening and embrittlement. In this work, RIS tendencies in face-centered cubic (fcc) Ni-X (X = Cr, Fe, Ti, Mn, Si, P) dilute binary alloys are examined. The goal is to investigate the driving forces and kinetic mechanisms behind the experimentally observed segregation. By means of ab initio calculations, point-defect stabilities and interactions with solutes are determined, together with migration energies and attempt frequencies. Transport and diffusion coefficients are then calculated in a mean-field framework, to get a full picture of solute-defect kinetic coupling in the alloys. Results show that all solutes considered, with the exception of Cr, prefer vacancy-mediated over interstitial-mediated diffusion during both thermal and radiation-induced migration. Cr, on the other hand, preferentially migrates in a mixed-dumbbell configuration. P and Si are here shown to be enriched, and Fe and Mn to be depleted at sinks during irradiation of the material. Ti and Cr, on the other hand, display a crossover between enrichment at lower temperatures, and depletion in the higher temperature range. Results in this work are compared with previous studies in body-centered cubic (bcc) Fe, and discussed in the context of RIS in austenitic alloys.
△ Less
Submitted 23 September, 2020;
originally announced September 2020.
-
Multi-scale modeling of the effects of temperature, radiation flux and sink strength on point-defect and solute redistribution in dilute Fe-based alloys
Authors:
Liangzhao Huang,
Maylise Nastar,
Thomas Schuler,
Luca Messina
Abstract:
In this work, we investigate the radiation-induced segregation (RIS) resulting from the coupling between the atomic and point defect (PD) fluxes towards the structural defects of the microstructure. This flux coupling depends on the migration mechanisms of PDs and atoms, including thermal diffusion mechanisms and forced atomic relocations (FAR) occurring in displacement cascades. We derive an anal…
▽ More
In this work, we investigate the radiation-induced segregation (RIS) resulting from the coupling between the atomic and point defect (PD) fluxes towards the structural defects of the microstructure. This flux coupling depends on the migration mechanisms of PDs and atoms, including thermal diffusion mechanisms and forced atomic relocations (FAR) occurring in displacement cascades. We derive an analytic model of the PD and solute RIS profiles accounting for PD production and mutual recombination, the FAR mechanism, and the overall sink strength of the microstructure controlling the elimination of PDs at structural defects. From this model, we present a parametric investigation of diffusion and RIS properties in dilute Fe-$B$ ($B$ = P, Mn, Cr, Si, Ni, and Cu) binary alloys, in the form of quantitative temperature/radiation flux/sink strength maps. As in previous works, we distinguish three kinetic domains for the diffusion and RIS properties: the recombination domain, the sink domain, and the thermal domain. Both our analytical approach and numerical applications demonstrate that the diffusion and RIS behaviors of PDs and solute atoms largely differ from one kinetic domain to another. Moreover, at high radiation flux, low temperature, and large sink strength, FARs tend to destroy the solute RIS profiles and therefore reduce the overall amount of RIS by forcing the mixing of solute and host atoms, especially close to PD sinks. Finally, we provide quantitative criteria to emulate in-reactor RIS behaviors by ion irradiation.
△ Less
Submitted 7 September, 2020;
originally announced September 2020.
-
Elastic dipole tensors and relaxation volumes of point defects in concentrated random magnetic Fe-Cr alloys
Authors:
Jan S. Wróbel,
Marcin R. Zemła,
Duc Nguyen-Manh,
Pär Olsson,
Luca Messina,
Christophe Domain,
Tomasz Wejrzanowski,
Sergei L. Dudarev
Abstract:
Point defects in body-centred cubic Fe, Cr and concentrated random magnetic Fe-Cr are investigated using density functional theory and theory of elasticity. The volume of a substitutional Cr atom in ferromagnetic bcc Fe is approximately 18\% larger than the volume of a host Fe atom, whereas the volume of a substitutional Fe atom in antiferromagnetic bcc Cr is 5\% smaller than the volume of a host…
▽ More
Point defects in body-centred cubic Fe, Cr and concentrated random magnetic Fe-Cr are investigated using density functional theory and theory of elasticity. The volume of a substitutional Cr atom in ferromagnetic bcc Fe is approximately 18\% larger than the volume of a host Fe atom, whereas the volume of a substitutional Fe atom in antiferromagnetic bcc Cr is 5\% smaller than the volume of a host Cr atom. Elastic dipole $\boldsymbol{P}$ and relaxation volume $\boldsymbolΩ$ tensors of vacancies and self-interstitial atom (SIA) defects exhibit large fluctuations, with vacancies having negative and SIA large positive relaxation volumes. Dipole tensors of vacancies are nearly isotropic across the entire alloy composition range, with diagonal elements $P_{ii}$ decreasing as a function of Cr content. Fe-Fe and Fe-Cr SIA dumbbells are more anisotropic than Cr-Cr dumbbells. Fluctuations of elastic dipole tensors of SIA defects are primarily associated with the variable crystallographic orientations of the dumbbells. Statistical properties of tensors $\boldsymbol{P}$ and $\boldsymbolΩ$ are analysed using their principal invariants, suggesting that point defects differ significantly in alloys containing below and above 10\% at. Cr. The relaxation volume of a vacancy depends sensitively on whether it occupies a Fe or a Cr lattice site. A correlation between elastic relaxation volumes and magnetic moments of defects found in this study suggests that magnetism is a significant factor influencing elastic fields of defects in Fe-Cr alloys.
△ Less
Submitted 30 July, 2020;
originally announced July 2020.
-
The dominant mechanisms for the formation of solute-rich clusters in low-Cu steels under irradiation
Authors:
N. Castin,
G. Bonny,
A. Bakaev,
F. Bergner,
C. Domain,
J. M. Hyde,
L. Messina,
B. Radiguet,
L. Malerba
Abstract:
The formation of nano-sized, coherent, solute-rich clusters (NSRC) is known to be an important factor causing the degradation of the macroscopic properties of steels under irradiation. The mechanisms driving their formation are still debated. This work focuses on low-Cu reactor pressure vessel (RPV) steels, where solute species are generally not expected to precipitate. We rationalize the processe…
▽ More
The formation of nano-sized, coherent, solute-rich clusters (NSRC) is known to be an important factor causing the degradation of the macroscopic properties of steels under irradiation. The mechanisms driving their formation are still debated. This work focuses on low-Cu reactor pressure vessel (RPV) steels, where solute species are generally not expected to precipitate. We rationalize the processes that take place at the nanometre scale under irradiation, relying on the latest theoretical and experimental evidence on atomic-level diffusion and transport processes. These are compiled in a new model, based on the object kinetic Monte Carlo (OKMC) technique. We evaluate the relevance of the underlying physical assumptions by applying the model to a large variety of irradiation experiments. Our model predictions are compared with new experimental data obtained with atom probe tomography and small angle neutron scattering, complemented with information from the literature. The results of this study reveal that the role of immobilized self-interstitial atoms (SIA) loops dominates the nucleation process of NSRC.
△ Less
Submitted 17 January, 2025; v1 submitted 14 December, 2019;
originally announced December 2019.
-
Solute diffusion by self-interstitial defects and radiation-induced segregation in ferritic Fe-X (X=Cr, Cu, Mn, Ni, P, Si) dilute alloys
Authors:
Luca Messina,
Thomas Schuler,
Maylise Nastar,
Mihai-Cosmin Marinica,
Pär Olsson
Abstract:
This work investigates solute transport due to self-interstitial defects and radiation induced segregation tendencies in dilute ferritic alloys, by computing the transport coefficients of each system based on ab initio calculations of binding energies and migration rates. The implementation of the self-consistent mean field method in the KineCluE code allows to extend the calculation of transport…
▽ More
This work investigates solute transport due to self-interstitial defects and radiation induced segregation tendencies in dilute ferritic alloys, by computing the transport coefficients of each system based on ab initio calculations of binding energies and migration rates. The implementation of the self-consistent mean field method in the KineCluE code allows to extend the calculation of transport coefficients to arbitrary interaction ranges, crystal structures, and diffusion mechanisms. The results show that the diffusivity of P, Mn, and Cr solute atoms is dominated by the dumbbell mechanism, that of Cu by vacancies, while the two mechanisms might be in competition for Ni and Si, despite the fact that the corresponding mixed dumbbells are not stable. Systematic enrichment at defect sinks is expected for P and Mn solutes due to dumbbell diffusion, and for Si due mainly to vacancy drag. Vacancy drag is also responsible for Cu and Ni enrichment below 1085 K. The RIS behavior of Cr is the outcome of a fine balance between dumbbell enrichment and vacancy depletion. Therefore, for dilute Cr concentrations global enrichment occurs below 540 K, and depletion above. This threshold temperature grows with solute concentration. The findings are in agreement with experimental observations of RIS and clustering phenomena, and confirm that solute-defect kinetic coupling plays an important role in the formation of solute clusters in reactor pressure vessel steels and other alloys.
△ Less
Submitted 15 April, 2020; v1 submitted 24 October, 2019;
originally announced October 2019.
-
KineCluE: a Kinetic Cluster Expansion code to compute transport coefficients beyond the dilute limit
Authors:
T. Schuler,
L. Messina,
M Nastar
Abstract:
This paper introduces the KineCluE code that implements the self-consistent mean-field theory for clusters of finite size. Transport coefficients are obtained as a sum over cluster contributions (in a cluster expansion formalism), each being individually computed with KineCluE. This method allows for the calculation of these coefficients beyond the infinitely dilute limit, and is an important step…
▽ More
This paper introduces the KineCluE code that implements the self-consistent mean-field theory for clusters of finite size. Transport coefficients are obtained as a sum over cluster contributions (in a cluster expansion formalism), each being individually computed with KineCluE. This method allows for the calculation of these coefficients beyond the infinitely dilute limit, and is an important step in bridging the gap between dilute and concentrated approaches. Inside a finite volume of space containing the components of a given cluster, all kinetic trajectories are accounted for in an exact manner. The code, written in Python, adapts to a wide variety of systems, with various crystallographic structures (possibly under strain), defects and solute amount and types, and various jump mechanisms, including collective ones. The code also features a set of useful tools, such as the sensitivity study routine that allows for the identification of the most important jump frequencies to get accurate transport coefficients with minimum computational cost.
△ Less
Submitted 14 September, 2018;
originally announced September 2018.
-
Smart energy models for atomistic simulations using a DFT-driven multifidelity approach
Authors:
Luca Messina,
Alessio Quaglino,
Alexandra Goryaeva,
Mihai-Cosmin Marinica,
Christophe Domain,
Nicolas Castin,
Giovanni Bonny,
Rolf Krause
Abstract:
The reliability of atomistic simulations depends on the quality of the underlying energy models providing the source of physical information, for instance for the calculation of migration barriers in atomistic Kinetic Monte Carlo simulations. Accurate (high-fidelity) methods are often available, but since they are usually computationally expensive, they must be replaced by less accurate (low-fidel…
▽ More
The reliability of atomistic simulations depends on the quality of the underlying energy models providing the source of physical information, for instance for the calculation of migration barriers in atomistic Kinetic Monte Carlo simulations. Accurate (high-fidelity) methods are often available, but since they are usually computationally expensive, they must be replaced by less accurate (low-fidelity) models that introduce some degrees of approximation. Machine-learning techniques such as artificial neural networks are usually employed to work around this limitation and extract the needed parameters from large databases of high-fidelity data, but the latter are often computationally expensive to produce. This work introduces an alternative method based on the multifidelity approach, where correlations between high-fidelity and low-fidelity outputs are exploited to make an educated guess of the high-fidelity outcome based only on quick low-fidelity estimations, hence without the need of running full expensive high-fidelity calculations. With respect to neural networks, this approach is expected to require less training data because of the lower amount of fitting parameters involved. The method is tested on the prediction of ab initio formation and migration energies of vacancy diffusion in iron-copper alloys, and compared with the neural networks trained on the same database.
△ Less
Submitted 28 August, 2018; v1 submitted 21 August, 2018;
originally announced August 2018.
