Correlating TEM images of damage in irradiated materials to molecular dynamics simulations (original) (raw)
Related papers
Journal of Nuclear Materials, 2006
Kinetic Monte Carlo is used extensively in the field of radiation effects to understand damage accumulation and growth under irradiation. These calculations require previous knowledge on the formation of these defects, the relative stabilities of the different types of defects, their interactions and their mobilities. Many of these parameters can be extracted from molecular dynamics calculations using empirical potentials or from ab initio calculations. However, the number of parameters necessary for a complete picture is rather large. Kinetic Monte Carlo can be used as a tool to isolate those parameters that most influence the outcome of the calculations. In this paper, we focus on one aspect: the form of the damage after the collision cascade. We describe the effect of the form of the cascade as obtained from molecular dynamics simulations on damage accumulation. In particular, we demonstrate that the form of the cascade drastically changes the nucleation and growth of helium-vacancy clusters, possible precursors of voids and bubbles. Finally, we point to those open questions that need to be resolved to develop a truly predictive kinetic Monte Carlo model.
First DAE Computational Chemistry Symposium, 2019
Special computational tools are necessary in simulating radiation induced changes in the micro-structure of materials. At the atomistic scales, where molecular dynamics simulations (MD) are used, the following tasks have to be carried out: (i) interatomic potentials have to be stiffened to account for interaction of high energy atoms which can get close to each other, (ii) electronic stopping has to be introduced since MD simulates only nuclear-nuclear interactions amongst atoms, (iii) defects and defect clusters have to be indentified; it is also desirable to classify the defect clusters so that the subsequent transport behavior of the clusters can be studied, and (iv) transport parameters of the defects, like, jump-correlations, pre-factor for diffusion and migration energies of self interstitial diffusion in presence of dumbells and crowdions have to be obtained from MD simulations. Once the details of the defects and their transport are obtained using atomic simulations, kinetic Monte Carlo (KMC) simulations can be used to study their diffusion and agglomeration at the micron levels for time scales of several hours. In this document, state of the art computational methods to carry out the above tasks, which are all developed in-house, are described. New methods like use of unsupervised machine learning to classify clusters occuring in collision cascades in W and Fe have been developed.
Journal of Nuclear Materials, 2006
The existence of an interplay between the structure of displacement cascades and point defect mobility that influences the long-term evolution of primary damage in a-Fe is revealed by applying an object kinetic Monte Carlo (OKMC) method. The investigation was carried out using different parameter sets, which primarily differ in the description of self-interstitial atom (SIA) cluster mobility. Two sets of molecular dynamics cascades (produced with the DYMOKA and the MOLDY codes, using different interatomic potentials) and one set of cascades produced in the binary collision approximation with the MARLOWE code, using a Ziegler-Biersack-Littmark (ZBL) potential were separately used as input for radiation damage simulation. The point defect cluster populations obtained after reaching 0.1 dpa were analyzed in each case and compared. It turns out that the relative influence of using different input cascade datasets on the damage features that evolve depends on which OKMC parameter set is employed. Published by Elsevier B.V.
2006
The multiscale modeling scheme encompasses models from the atomistic to the continuum scale. Phenomena at the mesoscale are typically simulated using reaction rate theory (RT), Monte Carlo (MC), or phase field models. These mesoscale models are appropriate for application to problems that involve intermediate length scales (µm to >mm), and timescales from diffusion (~µs) to long-term microstructural evolution (~years). Phenomena at this scale have the most direct impact on mechanical properties in structural materials of interest to nuclear energy systems, and are also the most accessible to direct comparison between the results of simulations and experiments. Recent advances in computational power have substantially expanded the range of application for MC models. Although the RT and MC models can be used simulate the same phenomena, many of the details are handled quite differently in the two approaches. A direct comparison of the RT and MC descriptions has been made in the domain of point defect cluster dynamics modeling, which is relevant to both the nucleation and evolution of radiation-induced defect structures. The relative merits and limitations of the two approaches are discussed, and the predictions of the two approaches are compared for specific irradiation conditions.
Nuclear Instruments and Methods in Physics Research Section B: Beam Interactions with Materials and Atoms, 2005
Displacement cascades obtained by full molecular dynamics and in its binary collision approximation, as well as random point defect distributions, all having similar overall morphologies, are used as input for long-term radiation damage simulation by an object kinetic Monte Carlo method in a-iron. This model treats naturally point defect fluxes on cascades regions, resulting from cascades generated in other regions, in a realistic way. This study shows the significant effect of the internal structure of displacement cascades and, in particular, SIA agglomeration on the long-term evolution of defect cluster growth under irradiation.
A detailed approach for the classification and statistical analysis of irradiation induced defects
Nuclear Instruments and Methods in Physics Research Section B: Beam Interactions with Materials and Atoms, 2015
New criteria are presented for the classification and statistical analysis of defects from irradiation cascades that allow a more detailed description of the diversity of damage, especially amorphous regions. Classical molecular dynamics simulations are used to analyze the damage produced by 2 keV Si recoils annealed at 1000 K for 1 ns. Based on a density grouping criterion of elementary defects (displaced atoms and empty lattice sites) the nonuniformity of local defect density within damage regions is revealed. The density criterion is able to distinguish dense damage regions which evolve independently upon annealing (although they are connected by some defects), while keeping small and compact regions unaltered. Damage regions are classified according to the size, net number of defects and compactness, calculated by averaging the distance among all defects, parameters that have a direct impact on their stability.
Atomistically-informed modeling of point defect clustering and evolution in irradiated ThO2
Chemical Physics
A cluster dynamics (CD) model has been developed to investigate the nucleation and growth of point defect clusters, i.e., interstitial prismatic loops and nanoscale and sub-nanoscale voids, in ThO2 during irradiation by energetic particles. The model considers cluster off-stoichiometry due to the asymmetry of point defect generation on the O and Th sublattices under irradiation, as well as the point defect diffusivities and the defect binding energies to clusters. The energies were established using detailed molecular dynamics simulations considering the statistical variability of cluster configuration. A high-order adaptive time-integration has been used to solve the model. The predicted loop density and their average size is in good agreement with reported experimental observations for proton irradiated ThO2 at 600 o C. The model did not predict void evolution due to the sluggish kinetics of cation vacancies, explaining the absence of voids in proton irradiated ThO2 (and other oxides) at relatively low temperatures.
Modeling Defect Evolution in Irradiated 800H using Cluster Dynamics
2017
A reaction-diffusion reaction rate theory based cluster dynamics was used to model the microstructure evolution of Alloy 800H under conditions similar to that of current and proposed nuclear reactors. The predicted interstitial and vacancy faulted loop densities grew orders of magnitude larger than experimentally measured in similar environments. The large calculated densities were determined to result from the over-nucleation of faulted loops directly generated by irradiation. In order to reduce the number densities, an additional reaction term is proposed that would approximate the enhanced recombination and reduced damage production caused by the damage cascade volume overlapping with the physical volume of defect clusters. To correctly parameterize the modified recombination and production terms, molecular dynamics simulations need to be performed to provide a computational database on the effects of a cascade overlapping with pre-existing defects.
Journal of Nuclear Materials, 1996
The evolution of high-energy displacement cascades in iron has been investigated for times up to 200 ps using the method of molecular dynamics simulation. The simulations were carried out using the MOLDY code and a modified version of the many-body interatomic potential developed by Finnis and Sinclair. Previously reported results have been supplemented by a series of 10 keV simulations at 900 K and 20 keV simulations at 100 K. The results indicate that the fraction of the Frenkel pairs escaping in-cascade recombination is somewhat higher and the fraction of the surviving point defects that cluster is lower in iron than in materials such as copper. In particular, vacancy clustering appears to be inhibited in iron. Many of the larger interstitial clusters were observed to exhibit a complex, three-dimensional morphology. The apparent mobility of the <111> crowdion and clusters of 411> crowdions was very high.