Skip to content. | Skip to navigation

This Slogan Viewlet registered to qPloneThemeBrendA product

Home » COSIRES Posters
Document Actions


Anatomy and Physiology of dislocation climb – A dissection by atomistic simulations and kinetic-Monte Carlo

Alankar Alankar, Los Alamos National Laboratory;Enrique Martinez, Los Alamos National Laboratory;Ricardo Lebensohn, Los Alamos National Laboratory;Alfredo Caro, Los Alamos National Laboratory

This work presents an analysis of dislocation-vacancy interaction in a pure Fe single crystal. By means of molecular dynamics simulations, we identify the unit processes that cause dislocation climb. The methodology suggests an improved determination of energy barriers that are used in kinetic-Monte Carlo simulations. The kinetic-Monte Carlo simulations show the effect of applied external stress on the vacancy concentration profiles around pure edge dislocation core. The flux of vacancies to dislocation core as a function of applied stress enables determination of stress exponent in the creep power-law that is used in a continuum crystal plasticity model. This work was supported by the nuclear energy advanced modeling and simulation (NEAMS) program, under the FMM project.

Bismuth segregation after impact of bismuth clusters into germanium

Christian Anders, Technical University of Kaiserslautern and Optimas;Herbert M Urbassek, Technical University of Kaiserslautern and Optimas

Nano-patterning of semiconductor surfaces can be induced by ion or cluster impact. Using molecular dynamics simulations, we study the impact of Bi_n clusters into a germanium target. In addition to pure Ge, we also study a Bi/Ge mixture in which 10% bismuth atoms are randomly distributed. The Bi_n projectile clusters (with n <= 5) impact with energies in the range of 3-20 keV/atom. The impacting clusters create a melt pool in the Ge surface which shows a characteristic meniscus since liquid Ge has a smaller volume than the solid phase. The heated Bi in the melt zone starts to segregate from the Ge and moves to the surface because it has a lower surface energy and a lower melting point. On the surface a Bi layer forms on the freezing Ge melt which is efficiently cleaned (so called snow plough effect). The surface modification, the amount of Bi collected on the surface and the remaining concentration of Bi in the former melt zone are analyzed in their dependence on projectile size, impact energy and direction.

Concurrent job control of simulations for iterative radiation and post processing

Takaaki Aoki, Kyoto Univ.;Toshio Seki, Kyoto Univ.;Jiro Matsuo, Kyoto Univ.;Takeshi Iwashita, Kyoto Univ.;Tasuku Hiraishi, Kyoto Univ.;Shin-ichi Satake, Tokyo Univ. of Science;Takahiro Kenmotsu, Doshisha Univ.

The development of high performance computing resources has expanded the variations of radiation simulations and also the number of atomic coordinate data given by simulations. Now, concerning atomic data analysis, concurrent and bulk task operation framework has become important from the view point of processing time and automation. Furthermore, this framework demands that, job description should be kept simple and easy-to-learn because of variations and life-cycles of analyzing programs, and be run on both local and remote batch system. On the other hand, concurrent job control can be also applied to evaluate the accumulation of radiation damage in a target. In this simulation, one would perform a large number of radiation simulations on a same target material iteratively. When the target is enough large and the range where single radiation effects for each trial is enough narrow and limited, it is allowed to run several irradiation tasks simultaneously, if they do not overlap with each other. Additionally, job controller should be designed to block a job from running if its range of effect will overlap ones of under execution jobs, and should abort descending jobs safely when a running job is abnormally terminated. This presentation will demonstrate applications of Xcrypt (, a domain specific language framework for concurrent job operation, for above simulation and analysis of radiation effects.

Evaluation of sputtering and damage with huge cluster impact using molecular dynamics simulations

Takaaki Aoki, Kyoto Univ.;Toshio Seki, Kyoto Univ.;Jiro Matsuo, Kyoto Univ.

Gas cluster beam technique, which utilizes large clusters of atoms or molecules to be irradiated on solid targets, is expected for high performance material processing and analysis because of its enhanced sputtering process. Large-scale molecular dynamics (MD) simulations of cluster impact, in which cluster consists of 1 million fluorine atoms and target with more than 130 million silicon atoms, were performed. The MD simulation results showed that there is certain threshold energy around 0.3eV/atom, where crater is formed on the target surface and cluster-surface reaction and vaporization of silicon fluorides are enhanced. Especially, when the incident energy of cluster is 1eV/atom (totally 1MeV), it was observed that about 10000 Si are sputtered in various forms of silicon-fluoride, not only mono-silicon fluorides such as SiF3 and SiF4, but also of silicon-fluoride clusters like Si2Fx, Si3Fx, etc.. The size distribution of sputtered molecules suggests that sputtering process obeys thermal excitation model due to high-density energy irradiation at the impact point. In the collisinal process of (F2)500000 1MeV impact, the cluster atoms penetrate the target surface and reach only several nm below from the target surface. However, displacements remain which reaches 10nm below from the surface remains even 80ps after the impact due to shock propagation. In this presentation, cluster size and energy dependence on sputtering and damage formation with huge supersonic cluster impact will be discussed.

Formation and evolution of MnNi clusters in neutron irradiated dilute Fe alloys modelled by a first principle-based AKMC method

Raoul Ngayam Happy, UMET & EM2VM & EDF;Lorenzo Malerba, SCK-CEN;Christophe Domain, EDF & EM2VM;Charlotte S. Becquart, UMET & EM2VM

An atomistic Monte Carlo model parameterized on electronic structure calculations data has been used to study the formation and evolution under irradiation of solute clusters in Fe-MnNi ternary and Fe-CuMnNi quaternary alloys. Two populations of solute rich clusters have been observed which can be discriminated by whether or not the solute atoms are associated with self-interstitial clusters. Mn-Ni-rich clusters are observed at a very early stage of the irradiation in both modelled alloys, whereas the quaternary alloys contain also Cu-containing clusters. Mn-Ni-rich clusters nucleate very early via a self-interstitial-driven mechanism, earlier than Cu-rich clusters; the latter, however, which are likely to form via a vacancy-driven mechanism, grow in number much faster than the former, helped by the thermodynamic driving force to Cu precipitation in Fe, thereby becoming dominant in the low dose regime. The kinetics of the number density increase of the two populations is thus significantly different. Finally the main conclusion suggested by this work is that the so-called late blooming phases might as well be neither late, nor phases.

Coupled motion of interstitial-loaded grain boundaries in bcc tungsten as a possible radiation damage healing mechanism under fusion reactor conditions.

Valery Borovikov, Los Alamos National Laboratory;Xian-Zhu Tang, Los Alamos National Laboratory;Danny Perez, Los Alamos National Laboratory;Xian-Ming Bai, Idaho National Laboratory;Blas P. Uberuaga, Los Alamos National Laboratory;Arthur F. Voter, Los Alamos National Laboratory

As a potential first-wall fusion reactor material, tungsten will be subjected to high radiation flux and extreme mechanical stress. We propose that under these conditions, coupled grain boundary motion, and enhanced mobility of interstitial-loaded grain boundaries, can lead to a radiation-damage self-healing mechanism, in which a grain boundary absorbs interstitials produced in collision cascades, activating motion of the grain boundary, which sweeps up the less-mobile vacancies. We examine the stress-induced mobility characteristics of a number of grain boundaries in W to investigate the likelihood of this scenario.

Why nano-projectiles work differently than macro-impactors- role of plastic flow

Eduardo M. Bringa, CONICET and ICB, Universidad Nacional de Cuyo, Mendoza, Argentina;Christian Anders, Fachbereich Physik und Forschungszentrum OPTIMAS, Universitat Kaiserslautern, Kaiserslautern, Germany;Gerolf Ziegenhain, Fachbereich Physik und Forschungszentrum OPTIMAS, Universitat Kaiserslautern, Kaiserslautern, Germany;Giles A. Graham, Mineralogy Department, The Natural History Museum, London, United Kingdom;J. Freddy Hansen, Lawrence Livermore National Laboratory, Livermore, CA, USA;Nigel Park, AWE, Plc Aldermaston, Reading, United Kingdom;Nick Teslich, Lawrence Livermore National Livermore, CA, USA;Herbert M. Urbassek, Fachbereich Physik und Forschungszentrum OPTIMAS, Universitat Kaiserslautern, Kaiserslautern, Germany

Hypervelocity impacts provide a way to take localized regions of a target to extreme pressure and temperature conditions. Resulting crater features can be challenging for hydrocode simulations and test the validity of constitutive models. We will present atomistic simulation data on crater formation due to hypervelocity impact of nanoprojectiles of up to 55 nm diameter and with targets containing up to ten billion atoms, and compare them to available experimental data on micron-, mm-, and cm-sized projectiles. We show that previous scaling laws do not hold in the nano-regime and outline the reasons: within our simulations we observe that the cratering mechanism changes, going from the smallest to the largest simulated scales, from an evaporative regime to a regime where melt and plastic flow dominate, as it is expected in larger micro-scale experiments [Anders et al., PRL 108, 027601 (2012)]. The importance of strain-rate dependence of strength and of dislocation production and motion under these extreme conditions will be discussed.

Molecular dynamics study of the continuous Deposition/Implantation of Carbon and Titanium on Silicon: preliminary results

Ludovic Briquet, Department “Science and Analysis of Materials” (SAM), Centre de Recherche Public – Gabriel Lippmann, 41 rue du Brill, L-4422 Belvaux, Luxembourg;Arindam Jana, Department “Science and Analysis of Materials” (SAM), Centre de Recherche Public – Gabriel Lippmann, 41 rue du Brill, L-4422 Belvaux, Luxembourg;Patrick Philipp, Department “Science and Analysis of Materials” (SAM), Centre de Recherche Public – Gabriel Lippmann, 41 rue du Brill, L-4422 Belvaux, Luxembourg;Gérard Henrion, Institut Jean Lamour, UMR CNRS – Université de Lorraine, Department « Chemistry and physics of solids and surfaces », Ecole des Mines - Parc de Saurupt, 54042 Nancy, France ;Tom Wirtz, Department “Science and Analysis of Materials” (SAM), Centre de Recherche Public – Gabriel Lippmann, 41 rue du Brill, L-4422 Belvaux, Luxembourg

Although plasma deposition techniques are nowadays widely used, a lot is still to be gained from the detailed knowledge of the deposition mechanisms of matter on surface at the sub-monolayer regime. Depending on the nature of the deposit, on the structure of the collector’s surface and on the deposition conditions, the matter being deposited may behave differently and be eventually located at different places in the system. Depending on the solubility and the surface interactions, the deposit may segregate, form films at the surface or evenly spread inside the sub-surface region. Within this framework, molecular dynamics are much more suitable to deposition investigation than binary collision techniques because of the low energy of the incoming particles. Indeed, as the velocity of the particle is low, the complete collision cascade has to be modelled in order to consider the effects of all interactions between the particle and the surface. In this study, we use a third generation reactive force field that is able to take into account the breaking and formation of chemical bonds to study deposition of carbon and titanium at the sub-monolayer regime. Third generation force fields are computationally heavier then other types of force fields such as the bond order ones, hence they were so far limited to simulations of single impacts on pristine surface. It is however well known, that the surface conditions evolve during the deposition process, which may have a great influence on the deposition itself. Therefore, instead of depositing every atom on a pristine surface as previously done, we deposit up to 100 carbon or titanium atoms in a continuous process on a 4 nm x 4 nm silicon surface at various energies and incidence angles. We are then able to follow the deposition mechanisms as well as the sticking properties of the atoms on the silicon surface with respect to the number of atoms that have already been deposited. It is also shown that titanium tends to form a film on top of the surface, while carbon implants and is diluted in the sub-surface region.

Modelling irradiation damage in Fe-Cu alloys with an object kinetic Monte Carlo approach

Nicolas Castin, SCK-CEN, Belgium;Maria Ines Pascuet, Conicet, Argentina;Dmitry Terentyev, SCK-CEN;Ville Jansson, SCK-CEN;Lorenzo Malerba, SCK-CEN

Hardening and embrittlement under neutron irradiation of low alloy ferritic steels are the limiting factors to the lifetime of nuclear power plants. It is long established that radiation-enhanced copper precipitation is one of the major causes for these changes in mechanical response, because precipitates act as obstacles to dislocation motion. Nanostructural changes are enhanced by supersaturation of point-defects that are debris of atomic collision cascades triggered by impinging high energy neutrons. In particular, vacancies have been shown to interact strongly with Cu atoms and precipitates, leading to the formation of small vacancy-copper complexes from the early stages. The interaction is sufficiently strong to lead to the dragging of Cu atoms by vacancies, ending up with the formation of Cu precipitates that can efficiently trap vacancies and may contain a significant amount of them. It seems therefore established that the mechanism leading to Cu precipitation under irradiation is governed by the diffusion of vacancy-copper (VCu) clusters. We have recently [1] been able to simulate the complete coherent stage of copper precipitation in bcc iron under thermal ageing, at different temperatures and alloy compositions, with a model that explicitly included the possibility for copper precipitates to diffuse. This is an additional argument to suggest that mobility of copper clusters, up to large sizes, should play a central role in the predicted precipitation kinetics. We have recently calculated the kinetics of migration and dissociation of VCu clusters, in a large variety of size and composition [2], with an atomistic model that includes all effects of relaxation and long-range chemical interactions. Our results showed that irradiation reduces the mobility of VCu clusters, which partly explains why the rate of coarsening is reduced under neutron irradiation. In this work, we include this information in an object kinetic Monte Carlo (see V. Jansson et al., same proceedings) to verify if our model, naturally taking into account also the contribution of many other actors such as SIA clusters and C interstitials, can predict this trend. [1] N. Castin et al., J. Chem. Phys. 135 (2011) 064502. [2] N. Castin et al., submitted to J. Nucl. Mat.

Mechanisms of loop growth in irradiated alpha-Fe modeled by object kinetic Monte Carlo simulations

Ignacio Martin-Bragado, IMDEA Materials, Spain;Mercedes Hernandez-Mayoral, CIEMAT, Spain;María J. Caturla, Universidad de Alicante, Spain

One of the outstanding questions about damage accumulation in irradiated alpha-Fe is the growth of loops with Burgers vectors <111> and <100>. On one hand, the mobility of <111> loops measured experimentally [1] is much higher than the values obtained from atomistic simulations [2]. On the other hand experiments show the presence of <100> loops in irradiated a-Fe under many different conditions of irradiation (electrons or ions) as well as temperatures and doses [1, 3, 4], even though the <111> configuration is the most stable one at low temperatures. At high temperatures the transformation of <111> loops into <100> loops has been observed experimentally [1] and explained theoretically [5, 6], and could explain some of the experimental results. For the low temperature observations there are different explanations for the presence of <100> loops: these loops could be formed within the collision cascades at very small sizes and then grow by loop coalescence [4] or they could be the result of the interaction of <111> loops to result into <100> loops [6]. In this work we have implemented several mechanisms for the growth of two different populations of loops in the newly developed object kinetic Monte Carlo code MMonCA. The energetics of defect stability and defect mobility used are those from ab initio calculations and molecular dynamics simulations, as well as the initial damage distribution. The accumulation of <100> and <111> loops as a function of dose as well as the size of these loops is studied depending on the type of mechanism assumed for loop growth. Results are then compared to existing TEM measurements of loop density as a function of dose. References: [1] K. Arakawa, K. Ono, M. Isshiki, K. Mimura, M. Uchikoshi, and H. Mori, Science 318, 956 (2007). [2] N. Soneda and T. Diaz de la Rubia, Philos. Mag. A 81, 331 (2001). [3] Z. Yao, M. Hernandez-Mayoral, M. L. Jenkins, and M. A. Kirk, Philos. Mag. 88, 2851 (2008). [4] Z. Yao, M. Jenkins, and M. Hernandez-Mayoral, Philos. Mag. 90,4623 (2010). [5] S. L. Dudarev, R. Bullough, and P. M. Derlet, Phys. Rev. Lett. 100,135503 (2008). [6] J. Marian, B.D. Wirth, J.M. Perlado, Phys. Rev. Lett. 88 (2002) 255507.

Thermal conductivity degradation induced by cascade accumulation or point defects in irradiated silicon carbide

Jean-Paul Crocombette, CEA, DEN, Service de Recherches de Métallurgie Physique

The high thermal conductivity of silicon carbide is a key property in view of its possible use for future fusion or fission nuclear reactors. However concerns exist about its degradation, as compared to the perfect material, due to the conditions of operation. In this work we studied with molecular dynamics simulations, the deterioration of the thermal conductivity of cubic SiC due to irradiation damage. We first calculated the variation of the thermal conductivity produced by the accumulation of displacement cascades [1]. The conductivity is found to decrease with dose down to very low values characteristic of the amorphous material, in very good quantitative agreement with low temperature irradiation experiments. We then calculated the thermal resistivity associated with point defects [2] , thus allowing comparison with high temperature irradiation experiments where amorphization does not take place and only point defects are created. The additional thermal resistivity due to point defects proves to vary linearly with their concentration. Large variations in the proportionality coefficient with the nature of the defects are observed. From these calculations, an approximate scale for the concentration of vacancies in irradiated SiC is built. The concentration of defects varies by two orders of magnitude depending on the irradiation conditions and can be as high as about 4 at. % for low temperature irradiation just above amorphization critical temperature, in qualitative agreement with swelling data. References: [1] Jean-Paul Crocombette, et al. J. Appl. Phys. 101, 023527 (2007) [2] Jean-Paul Crocombette and Laurent Proville, Applied Physics Letters 98 (2011), 191905

Cluster-induced desorption from metal organic surfaces: Structural effects

Arnaud Delcorte, Université catholique de Louvain;Emilien van Hoecke, Université catholique de Louvain;Oscar A. Restrepo, Université catholique de Louvain

Kiloelectronvolt molecular (C60) and cluster (Au{n}, Bi{n}) beams are routinely used as projectiles for secondary ion mass spectrometry (SIMS). In comparison with atomic ions, they induce much larger sputtering and secondary ion yields in inorganic as well as organic materials. In addition, projectiles such as C60 remove a large part of the damage they create in the solid, so that their use in dynamic SIMS opened the door to the molecular depth profiling of many organic materials. If the basic mechanisms are now well understood [1], many questions related to the behavior and performance of cluster projectiles in the bombardment of hybrid and nanostructured solids made of different materials are still open. In this contribution, we use molecular dynamics simulations to model the 10 keV bombardment of Au-nanoparticle (NP)-covered polymeric samples by C60 and Au400 projectiles, at normal incidence [2]. While Au400 impacts on the Au-NPs favor the polymer emission, by increasing the energy deposition in the polymer surface, the Au-NPs reflect or deflect much of the lighter C60 projectile energy, thereby reducing the molecular emission. In contrast C60 is more successful when impinging directly on the organic material. Our results show that these trends are valid for kDa polymers (which can be emitted intact) as well as for virtually infinite length chains (which require fragmentation), but that the polymer sputtered mass is consistently >5 times larger in the case of the kDa molecules for all impact points and projectiles. However, burying the Au-NPs by ~3 nm in the organic material leads to completely different results, with, upon C60 bombardment, the largest sputtered masses observed for impacts above the NPs. For Au400 bombardment, the burial of the Au-NPs leads to comparatively lower sputtered masses. These new results demonstrate the complexity of the sputtering of nanostructured hybrid materials by cluster projectiles and explain various artifacts that should complicate the analysis and depth profiling of such materials. [1] B. J. Garrison, Z. Postawa, Mass Spectrom. Rev. 2008, 27, 289–315. [2] O. A. Restrepo, A. Delcorte, J. Phys. Chem. C 2011, 115, 12751–12759.

Water nanodroplet impacts on surfaces: Effect of the substrate nature

Arnaud Delcorte, Université catholique de Louvain;Barbara J. Garrison, Penn State University

Charged water nano- and micro-droplets, with velocities in the range 102-104 m/s, are used in a variety of analytical techniques to produce molecular ions from analyte molecules upon impact, either through the desolvation of molecules embedded in the droplets (IDEM) or via the desorption of molecules from the bombarded surfaces (EDI MS and DESI MS). Several research groups have performed atomistic MD simulations of km/s nanodroplet impacts, modeling the substrate by a repulsive wall [1,2] or by a nonpolar surface [3]. However, in the aforementioned experiments, the surface may be naturally polar or already covered by a water film. In this contribution we report on the simulation of 1-10 km/s (H2O)n impacts on planar surfaces, with normal or oblique incidence (60°). The focus is placed on the effect of the substrate, one of the targets being a rigid atomic layer with a repulsive interaction to the droplet, while the other one is a polar surface modeled by three layers of NaCl. In our simulations at 60° incidence, the velocity limit for droplet fragmentation is between 2 and 3 km/s for both substrates. However, the dynamics and the energetics of the interaction are very dependent on the substrate nature. While <2km/s droplets glide on the repulsive substrate, keeping most of their translational energy, they stick and stop on the polar substrate, transforming their energy into heat. The influence of the substrate is also pronounced for velocities above the fragmentation threshold, with much higher internal energies and more extensive fragmentation observed for the polar substrate. The results are mainly discussed on the basis of the energy partitions and particle distributions obtained upon interaction. The effects of the droplet size and initial temperature are also considered. In conclusion, our simulations demonstrate that the nature of the substrate cannot be overlooked in such impact processes, which may have important consequences for the experiments. [1] A. Tomsic, H. Schroder, K.L. Kompa, C. R. Gebhardt, J. Chem. Phys., 2003, 119, 6314. [2] S. N. Sun, H. M. Urbassek, J. Phys. Chem. B, 2011, 115, 13280. [3] A. Tomsic, P. U. Andersson, N. Markovic, W. Piskorz, M. Svanberg, J. B. C. Pettersson, J. Chem. Phys., 2001, 115, 10509.

Molecular dynamics simulation of the irradiation of carbon nanotubes and graphene layers

Cristian D. Denton, Universidad de Alicante;Santiago Heredia-Avalos, Universidad de Alicante;María J. Caturla, Universidad de Alicante

The irradiation of carbon based nanostructures with ions and electrons has been shown to be an appropriate tool to tailor their properties [1]. The defects induced in the nanostructures during irradiation are able to modify their mechanical and electronic properties. In particular, in graphite, experiments show magnetism induced by irradiation, that is attributed to either the formation of vacancies or the presence of hydrogen [2]. Here we simulate the irradiation of carbon nanotubes and graphite monolayers with ions using a molecular dynamics code. We use empirical interaction potentials, such as the Brenner or the Tersoff potentials [3] joined smoothly to the Universal ZBL potential [4] at short distances. We evaluate the damage produced during the irradiation for different ions and projectile energies. In graphene we observe the formation of some very stable structures of low coordination atoms such as monatomic chains, which are also observed experimentally [5]. [1] A. V. Krasheninnikov and K. Nordlund, J. Appl. Phys. 107 (2010) 071301. [2] P. Esquinazi, D. Spemann, R. Höhne, A. Setzer, K.-H. Han, and T. Butz, Phys. Rev. Lett. 91 (2003) 227201. [3] L. Lindsay and D. A. Broido, Phys. Rev. B 81 (2010) 205441. [4] J. F. Ziegler, J. P. Biersack and M. D. Ziegler, SRIM: The stopping and range of ions in matter, Lulu Press, Napa (2008). [5] C. Jin, H. Lan, L. Peng, K. Suenaga and S. Iijima, Phys. Rev. Lett. 102 (2009) 205501.

Temperature dependent irradiation response of nanoporous Au foams

E. G. Fu, Los Alamos National Laboratory;L. Zepeda-Ruiz, Lawrence Livermore National Laboratory;M. Caro, Los Alamos National Laboratory;Y. Q. Wang, Los Alamos National Laboratory;E. Bringa, Universidad de Cuyo, Argentina;J. Baldwin, Los Alamos National Laboratory;M. Nastasi, University of Nebraska, Lincoln;A. Caro, Los Alamos National Laboratory

In a recent computational study [1] we proposed the existence of a window of radiation tolerance for nanoscale foams, defined by two fundamental scales, a length scale relating filament size to cascade diameter, and a time scale relating diffusion of point defects across the filaments to dose rate. In this presentation, we report on experiments and computer simulations of the temperature dependent irradiation response of nanoporous Au foams under Ne ion irradiations, and analyze our findings in the framework of this proposed window for radiation endurance. The co-deposited Au-Ag thin films on single crystal NaCl substrate by electron beam evaporation were first dealloyed to form Au foams by immersing them in dilute nitric acid. Then, the Au foams were irradiated by 400 keV Ne ions at room (RT) and liquid nitrogen temperatures (LNT). Rutherford backscattering spectrometry was used to examine the composition and thickness of the Au-Ag thin films and transmission electron microscopy was used to characterize the microstructure of the Au foams before and after ion irradiation. The results show significant changes of the microstructure of the nanoporous Au foams after ion irradiation. Stacking fault tetrahedra (SFTs) were observed at RT irradiation of Au foams whereas no SFTs were found at LNT temperature irradiation. The possible mechanisms to explain the observations are explored with computer simulations of irradiation of filaments with different diameters, various PKA’s energies, impact locations and orientations. We find that while interstitials disappear after rapid migration to the surface, vacancies remain and accumulate, leading to the spontaneous formation of SFTs. These results represent a first step towards a systematic exploration of the response of nanofoams to radiation. Work supported by the Laboratory Directed Research and Development Program at Los Alamos National Laboratory. [1] E. M. Bringa, J. D. Monk, A. Caro, A. Misra, L. Zepeda-Ruiz, M. Duchaineau, F. Abraham, M. Nastasi, S. T. Picraux, Y. Q. Wang, and D. Farkas., “Are Nanoporous Materials Radiation Resistant?”, to appear in Nano Letters, July 2012.

Effect of Grain Boundary on Lattice Thermal Conduction of Tungsten revealed by Molecular Dynamics Simulations

Baoqin Fu, Tsinghua University;Wensheng Lai, Tsinghua University;Wei Liu, Tsinghua University;Yue Yuan, Tsinghua University;Haiyan Xu, Tsinghua University

Tungsten (W) and tungsten-based materials are being considered to use in the divertor and first wall as plasma facing materials (PFMs) in the fusion reactor owing to their low sputtering yield, low tritium retention, no chemical erosion, good thermo-mechanical properties and high thermal conductivity. Melting, cracking and blistering might occur for W under such work condition. These radiation damage phenomena are closely related to lattice thermal conduction of W. Moreover, Bai et al [1] recently described that grain boundaries might have exceptional ability for materials to resist radiation damage, suggesting that polycrystalline W may be a better choice for PFMs. However, PFMs must reliably withstand thermal cycles with heat fluxes of the order of 20 MW/m2 up to several thousand times as required in ITER [2]. Therefore, it is important to study the effect of grain boundary on lattice thermal conduction of W. Various grain boundaries have been constructed and the lattice thermal conductivities (TCs) have been calculated by molecular dynamics (MD) simulations with Finnis and Sinclair’s empirical N-body potential [3]. It turns out that there exist the sharp temperature drops across these grain boundaries, indicating that TCs near grain boundary are much smaller than those in the bulk. The grain boundary effect on TCs of polycrystalline W samples has been analyzed by combining MD and finite element results. These research results are potentially helpful for the design of ITER and the choice of PFM. [1] X.M. Bai, A.F. Voter, R.G. Hoagland, et al, Science, 327 (2010) 1631. [2] G. Federici, Phys. Scr., T124 (2006) 1. [3]M.W. Finnis, J.E. Sinclair, Philos. Mag. A, 50 (1984) 45. * Work supported by National Magnetic Confinement Fusion Science Program (Grant 2009GB106003), and the National Nature Science Foundation of China (Contract No.51071095).

Computer simulation of kinetic excitation in atomic collision cascades in metals: Progress towards a ballistic treatment of electronic excitation transport

Stefanie Hanke, University of Duisburg-Essen;Andreas Duvenbeck, University of Duisburg-Essen;Boris Weidtmann, University of Duisburg-Essen;Andreas Wucher, University of Duisburg-Essen

The impact of a keV ion onto a metal surface leads to a sequence of collisions among the target atoms, which may lead to sputtering. The space- and time evolution of this atomic collision cascade is accompanied by a kinetic excitation of the electronic system. These electronic excitations lead to phenomena like electron emission or secondary ion formation at ion-bombarded solids. In a series of previous studies [2] we have developed a computer simulation model to describe ion impact induced atomic collision cascades. The model consists of (i) a Molecular Dynamics (MD) -treatment of the cascade dynamics, (ii) an implementation describing the transfer of kinetic into electronic excitation energy and (iii) a diffusive approach to describe the transport of excitation energy. The model calculations yield a time- and space-dependent excitation energy density profile that has been successfully employed to calculate ionization probabilities of sputtered particles or, respectively, kinetic external electron emission yields for various bombarding parameters. The physical picture extracted from our previous studies is that electron emission processes mainly originate from electronic excitations generated during the very initial stage of the cascade (~10 fs) when the projectile penetrates the uppermost layer. On this time scale, however, the diffusive treatment of excitation energy transport is afflicted with a high degree of uncertainty. In order to get insight into the accuracy of a diffusive approach on that time scale, we apply our model to the description of internal electron emission across a buried metal-insulator-metal tunnel junction. The propagation of the initial electronic excitation peak generated at the very surface towards deeper layers in the bulk crucially determines the internal electron emission yield across such a junction. Thus, the comparison of calculated internal electron emission yields with experiments [1] provides a unique opportunity to study the underlying transport physics on a fs-time scale. The results will be discussed in terms of a more general, ballistic description of electronic transport process via the Boltzmann transport equation. [1] Meyer et al, PRL, 2004, 93, 137601 [2] Wucher et al, NIMB, 2011, 269, 1655

The effects of comprehensive factors on displacement cascade in iron

Nengwen Hu, Department of Applied Physics, Hunan University;Huiqiu Deng, Department of Applied Physics, Hunan University;Shifang Xiao, Department of Applied Physics, Hunan University;Wangyu Hu, Department of Applied Physics, Hunan University

Large mounts of Helium atoms can be generated by (n, a) transmutation reactions in fusion environment. The accumulation of Helium in materials can cause the formation of helium bubble, void swelling, blistering and so on, which degrade the mechanical properties of materials. It is, therefore, important to understand the interaction of helium atoms with metal atoms. First Principles approach, Molecular Dynamics and Kinetic Monte Carlo simulations have been widely used to investigate the behaviors of helium in iron. Results from ab initio calculations show that Helium prefer to stay at tetrahedral interstitial sites rather than octahedral interstitial sites [Chu-Chun Fu et al., Phys. Rev. B 72, 064117 (2005)]. Recently it was found that the initial locations of Helium greatly influence the number of Frenkel pairs after cascade in α-Fe [G Lucas et al., J. Phys.: Condens. Matter 20 (2008) 415206 (12pp)]. Though lots of results concerning Helium and cascade in α-Fe have been published, the behavior of Helium atoms in iron is still unclear. In present work, comprehensive influencing factors have been taken into consideration. Large mounts of displacement cascades with different Helium concentrations have been performed under different EPKA at different temperatures to make out their effects on displacement cascades. It is indicated that defect production efficiency increases with the contents of Helium atoms, EPKA and the radiation temperature increasing. The size distribution of defects clusters are greatly affected by Helium atoms and the size of the largest clusters decrease with the concentration of Helium increasing. Most of the clusters contain only three point defects and most of these clusters present in the manager of dumbbells in the direction of <110>. Moreover, subcascades have been observed only under high energy radiation and vacancy clusters stay along the motion path of the PKA in iron with 1.0% Helium impurities but only in the core of the primary cascade and subcascade in pure Fe.

Computer Simulation of Few Layer Graphene Nanostructures Modified by Bridge-Like Radiation Defects

Arkady M Ilyin, National NanoLab Kazakh National University

The paper presents results of computer simulation and study of energetic and structural characteristics of a special type of bridge-like radiation defects in few- layer graphene structures. The main feature of bridge-like defect is the interstitial atom arranged between two graphene sheets with two vacancies, faced each other, bonding two graphene sheets fast together [1,2]. Such few layer graphene elements can be used as reinforcement particles in composites as well as safe and capacious cells for hydrogen storage. Recently nanomaterials based on ultra-thin graphite have found a use in the technological field relating to electrical sources, in particular, for production of lithium-ion rechargeable batteries. But there are some obstacles on this way of application, linking for example, with deformation of cells which results in limitation of a charge capacity and dimension instability of devices. The paper presents preliminary results of computer simulation and density functional theory calculations of possible configurations of nanoscale few layer graphene cells which serve as storing elements for lithium atoms. Our calculations show that radiation modifying of few layer graphene nanomaterials by bridge-like radiation defects provides stiffness and dimension stability of graphene cells that can provide essential increase of charge capacity and dimension stability of electrical sources based on Li. Moreover, the existence of bridge-like defects essentially improves one more important feature of lithium ion sources: it enlarges the permeation and mobility of Li ions through graphene cells. Therefore, radiation modification of graphene cells with bridge-like defects might become a key technology to improve mechanical properties of ultra-thin graphite particles and few layer graphene structures, that results, in particular, in the essential enlargement of charge capacity and dimension stability of Li –ion-based batteries. References [1] A.M.Ilyin. Computer Simulation of Radiation Defects in Graphene and Relative Structures. In “Graphene Simulation”, Ed. by Jian Ru Gong, “InTech”, 2011, 39-52. [2] A.M.Ilyin. Simulation of End-Bridge-Like Radiation Defects in Carbon Nanotubes.COSIRES -2010, Poland, Krakov, July 19-23,2010,P.123

Modelling the Growth of Thin Films Oxides

Sabrina Blackwell, Loughborough University;Steven D Kenny, Loughborough University;Roger Smith, Loughborough University;Mike Walss, Loughborough University

We present results for modelling the growth of thin film oxides, our emphasis on rutile TiO2; recent results on ZnO will also be reported. We illustrate how long time scale dynamics techniques can be used to model growth at realistic temperatures, over experimental timescales. Modelling is carried out using a multi-timescale technique involving molecular dynamics and an on-the-fly Kinetic Monte Carlo (otf-KMC) method, which finds diffusion pathways and barriers, in parallel, without prior knowledge of transitions. We examine effects of varying parameters, such as substrate bias, plasma density and stoichiometry of the deposited species. Growth of rutile TiO2 via three deposition methods has been investigated; evaporation, ion-beam assist and reactive magnetron sputtering. It was concluded that evaporation produced a void filled, incomplete structure even with the low energy ion-beam assist. The sputtering process, however, produced crystalline TiO2. Results show that the energy of the deposition method plays a vital role in the resulting oxide thin film quality. ZnO is a very interesting yet challenging material to model. We will present the results of ZnO growth via the same deposition methods used for TiO2. It is already clear from early results that ZnO growth incorporates some highly interesting and complicated growth mechanisms, involving concerted motions, which would not have been discovered without the use of otf-KMC, thus demonstrating the power of the technique.

Reactive MD study of small Si nanowire oxidation: Control of Si-core radius

Umedjon Khalilov, University of Antwerp;Geoffrey Pourtois, IMEC, University of Antwerp;Adri C. T. van Duin, Penn State University;Erik C. Neyts, University of Antwerp

Core shell silicon nanowires (Si-NWs) have been used for application as field-effect transistors. For this purpose, size control of Si-NWs is particularly important for precise control of the band gap. Usually, small Si-NWs are obtained by oxidizing large diameter Si-NWs and removing the oxide shell. Therefore, we here focus on the Si-NW oxidation in order to unravel the oxidation mechanisms of Si nanowires with a small diameter (< 3 nm) on the atomic scale. In this work, we report on the simulated formation of such core shell Si-NWs by dry thermal oxidation of (100) Si nanowires with initial diameters from 1.0 nm to 3.0 nm in the temperature range 300 K - 1273 K, by reactive molecular dynamics simulations using the ReaxFF potential. Temperature dependent oxidation mechanisms of small diameter Si-NWs are discussed. Self-limiting small Si-NW oxidation is studied, i.e., control over the Si-core radius and the SiOx (x ≤ 2.0) oxide shell is possible by controlling the growth temperature and by choosing the initial Si-NW diameter. Two different structures were obtained, i.e., ultrathin SiO2 silica (i.e., dielectric) nanowire, and a Si core | ultrathin SiO2 silica shell (i.e., semiconductor + dielectric) nanowire. Both structures are analyzed in detail. Furthermore, we also investigated the stress build-up in the nanowires as a function of the oxidation progress. This study is important for the fabrication of nanowire based nanoscale electronic devices.

Self-limiting hyperthermal Si oxidation: a reactive MD study

Umedjon Khalilov, University of Antwerp;Geoffrey Pourtois, IMEC, University of Antwerp;Adri C. T. van Duin, Penn State University;Erik C. Neyts, University of Antwerp

We have been investigating hyperthermal Si oxidation at the atomic-scale, using reactive molecular dynamics simulations by means of the ReaxFF potential. Oxidation of Si(100){2x1} surfaces by both atomic and molecular oxygen was investigated in the energy range 1-5 eV. As part of our investigations, we report the oxidation and growth mechanism of ultrathin silica (SiO2) layers during hyperthermal oxidation at room temperature. In the case of oxidation by molecular O2, silica is quickly formed and the thickness of the formed layers remains limited compared to oxidation by atomic oxygen. Oxygenated Si structures are found to be divided into three regions, i.e., silica bulk, a transition layer, and the Si surface. We analyzed the structures in terms of partial charges, mass, radial and angle distributions. The obtained structures of the ultrathin SiO2 films are amorphous, including some intrinsic defects. Furthermore, we also investigated various intrinsic defects in the Si|SiO2 interface, which arise due to interfacial stresses. During oxidation, the compressive stress increases in the oxygenated Si, i.e., the Si|SiO2 interface and ultrathin amorphous silica (a-SiO2). This compressive stress significantly slows down the inward silica growth. In the interface, the compressive stress is found to be about 2 GPa, which is close to the experimental value in the Si|SiO2 interface obtained by traditional thermal oxidation. Finally, we demonstrate how the silica thickness can be controlled by controlling the initial kinetic energy and type of incident oxygen species. We compared our results with experimental and ab-initio data, and good agreement was found. This study is important for the fabrication of silica-based devices in the micro and nanoelectronics industry, and more specifically for the fabrication of metal-oxide-semiconductor devices.

Reactive MD Study of Hyperthermal Si Oxidation: Effect of Growth Temperature

Umedjon Khalilov, University of Antwerp;Geoffrey Pourtois, IMEC, University of Antwerp;Adri C. T. van Duin, Penn State University;Erik C. Neyts, University of Antwerp

In this work, we studied the growth mechanism of ultrathin silica (SiO2) layers during hyperthermal oxidation as a function of temperature in the range 100 K – 1300 K by means of reactive molecular dynamics simulations based on the ReaxFF potential. The oxidation of Si(100){2x1} surfaces by both atomic and molecular oxygen was investigated for hyperthermal impact energies in the range of 1 to 5 eV. Two different growth mechanisms are found, corresponding to low temperature oxidation and high temperature oxidation. The transition temperature between these mechanisms is estimated to be about 700 K. Below this temperature, the oxide thickness only depends on the impact energy of the impinging species. Above this temperature, the oxide thickness depends on both the impact energy and the surface temperature. Also, the initial step (i.e., direct oxidation stage) of the Si oxidation process is analyzed in detail. Where possible, we validated our results with experimental and ab-initio data, and good agreement was obtained. These results are of importance for the fabrication of silica-based devices in the micro and nanoelectronics industry.


Ane Lasa, University of Helsinki;Krister O. E. Henriksson, University of Helsinki;Kai H. Nordlund, University of Helsinki

ITER aims to be the first Tokamak fusion reactor producing more energy than it consumes, based on a deuterium (D)-tritium (T) plasma. Due to the ITER TOKAMAK design, the conditions (temperature, particle flux and thermal loads), and thus the material requirements, vary with location in the reactor. The divertor area is where the plasma is designed to be touching the wall, meaning that the materials there will suffer the most extreme conditions and the most intense plasma-wall interactions. The divertor of the new ITER-like wall JET, as well as the one planned for ITER itself, is made of Tungsten (W), due to its high sputtering threshold and low chemical interaction. A W-divertor would minimize harmful effects, such as T retention and wall erosion. On the other hand, Helium (He) will also be present in the plasma, as it is one of the products of the D-T fusion. When the He leaves the reactor, the W-divertor is exposed to a He-containing plasma. Experiments by Baldwin and Doerner [1] showed the formation of a fuzz-like nano-morphology in W when it was exposed to the He-plasma. These results pose a new undesired phenomenon for the divertor, that needs to be well understood in order to minimize its possible effects. The atomic-scale dynamics in materials can be analyzed using Molecular Dynamics simulations (MD), allowing the study of the onset of the fuzz formation mechanisms, such as structure growth and material mixing. We simulated He impacts on W, analyzing the effect of the cell size, irradiation energy (30-60 eV) and impurities -W irradiated by He and C-seeded He- on the structure formation and its mechanism. Our results show that deeper, larger and more He clusters form when increasing the cell size and irradiation energy. Moreover, atoms are pushed to the surface by interstitial and interstitial loop punching due to the high pressure in the bubbles, providing a likely explanation to the onset of growth. [1] Baldwin and Doerner, Nucl. Fusion 48 (2008) 035001

Nucleation and Growth of He Bubbles at Grain Boundaries in α-Fe

Li Yang, Pacific Northwest National Laboratory;Fei Gao, Pacific Northwest National Laboratory;Richard J Kurtz, Pacific Northwest National Laboratory;Howard L Heinisch, Pacific Northwest National Laboratory

The formation of He bubbles at grain boundaries (GBs) can have major consequences for the mechanical integrity of first-wall materials in future fusion power systems. To gain insight into He effects, the accumulation of He atoms and nucleation of He bubbles in the Σ3 <110> {112} and Σ73b <110> {661} GBs in α-Fe are studied by molecular dynamics methods that use a newly developed Fe-He potential. For this study He atoms are inserted randomly within a region consisting of the 8 atom layers straddling the GB plane at local concentrations of 1%, 5% and 10%. The local He concentration is calculated as the ratio of He-to-Fe atoms within the 8-layer region. We recognize the anticipated end-of-life bulk He concentration is about 0.15%, but it is well known that He segregates to GBs, and other excess volume defects, so it is reasonable to postulate higher local He concentrations than in the bulk. Our results show that He accumulation, He bubble formation, and GB structural evolution all depend on the He concentration, temperature and initial GB structure. For the Σ3 GB at a He concentration of 1%, small He clusters are formed, and a few Fe self-interstitial atoms (SIAs) are ejected from those clusters at temperatures greater than 600 K. The SIAs easily migrate to nearby atomic planes and form <111> crowdions. At a He concentration of 5%, a few large He clusters are found, in platelet form, along with dislocations at the periphery of these large clusters. At a He concentration of 10%, the GB structure changes significantly due to formation of large He clusters and emission of a large number of SIAs. The SIAs rearrange to form an extra atomic plane within the GB. This results in GB dislocation climb along the direction, and healing of the deformation caused by the He clusters. In contrast, He clustering in the Σ73b GB is more difficult compared to the Σ3 GB at the same concentration and temperature. With increasing He concentration GB dislocation climb along the <112> direction is observed, which is driven by He cluster nucleation at the original position of the GB dislocation core.

Comprehensive modelling of Solid Phase Epitaxial Growth using Lattice Kinetic Monte Carlo

Ignacio Martin-Bragado, Institute IMDEA Materials

Damage evolution of irradiated silicon is and has been a topic of interest for the last decades for its applications to the semiconductor industry. In particular sometimes the damage is heavy enough to collapse the lattice and locally amorphize the silicon, while in other cases amorphization is introduced explicitly to improve other implanted profiles. Subsequent annealing of the implanted samples heals the amorphized regions through Solid Phase Epitaxial Regrowth (SPER). SPER is a complicated process. It is anisotropic, it generates defects in the recrystallized silicon, has a different amorphous/crystalline (A/C) roughness for each orientation, leaving pits in Si(110), and in Si(111) produces two “modes” of recrystallization with different rates. We have used the recently developed code MMonCa (Modular Monte Carlo simulator), presented in this conference for the first time, to introduce a physically-based comprehensive model using Lattice Kinetic Monte Carlo that explains all the above singularities of silicon SPER. The model operates by having as building blocks the silicon lattice microconfigurations and its four twins. It detects the local configurations, assigns microscopical growth rates, and reconstructs the positions of the lattice locally with one of those building blocks. The overall results reproduce the a) anisotropy as a result of the different growth rates, b) localization of SPER induced defects, c) roughness trends of the A/C interface, d) pits on Si(110) regrown surfaces, and e) bimodal Si(111) growth, while providing physical insights of the nature and shape of deposited defects and how they assist in the occurance of all the above effects.

The Effect of Hydrocarbon Chemistry on Sputtering in Mixed Be-C-H Materials

Andrea Meinander, EURATOM-Tekes, Department of Physics, P.O. Box 43, 00014 University of Helsinki, Finland;Carolina Björkas, EURATOM-Tekes, Department of Physics, P.O. Box 43, 00014 University of Helsinki, Finland, and Institute of Energy and Climate Research – Plasma Physics, Forschungszentrum Jülich GmbH, Association EURATOM-FZJ, Partner in the Trilateral Euregio Cluster, Jülich, Germany;Kai Nordlund, EURATOM-Tekes, Department of Physics, P.O. Box 43, 00014 University of Helsinki, Finland

Significant sputtering yields are observed for certain materials at energies well below the threshold for physical sputtering. At these levels, where the energy of the incident ions is comparable to the strength of chemical bonds, chemical effects can be expected to contribute to material erosion. This is a critical issue already during the developmental phases of the ITER fusion reactor, where ions with kinetic energies of the order of 1 - 100 eV will interact with the complex Be/W/C-based plasma materials surface. Erosion of carbon based materials at these energies has been observed in numerous experiments, and has been shown by molecular dynamics (MD) simulations to occur through the process of swift chemical sputtering (SCS). Hydrocarbon chemistry plays a crucial role in SCS of carbon, which can be seen by comparing the predictions of different interaction models, i.e. inter-atomic potentials. The Brenner hydrocarbon potential takes into account bond conjugation in carbon chains, providing a chemically motivated description of the C-H bonding behavior. Using this potential, it has been shown that terminating methyl-like groups on the surface are the precursors to sputtered CH_x molecules. Tersoff type potentials can successfully describe both covalent and metallic bonds within the same formalism, facilitating the construction of a potential for the Be-C-H system. This potential predicts the formation of areas of Be2C interspersed with amorphous C in compounds with BeC and BeC2 stoichiometry. Our simulations indicate a higher rate of Be sputtering from these grain boundaries. However, we observe no C sputtering, and inspection shows a total lack of the critical methyl groups. Thus, for a complete picture of SCS of Be-C compounds, it is necessary to take into account the special bonding behavior of carbon. To enable this , we have combined the Brenner hydrocarbon potential with the potential for Be-C-H, allowing investigation of the effect of carbon chemistry on the complete ternary system. This also opens the way for study of other mixed systems, such as W-C-H, which can be described within the same framework. We present the results from MD simulations of consecutive bombardment by incident ions at energies from 10 to 100 eV on Be-C compounds.

Density functional theory and molecular dynamics simulations of hydrocarbon deposition on silicon

Lotta Mether, Department of Physics, University of Helsinki;Krister Henriksson, Department of Physics, University of Helsinki;Kai Nordlund, Department of Physics, University of Helsinki;Canan Turgut, Department “Science and Analysis of Materials, Centre de Recherche Public – Gabriel Lippmann;Patrick Philipp, Department “Science and Analysis of Materials, Centre de Recherche Public – Gabriel Lippmann

Plasma surface treatments have come to provide both an efficient and ecological tool for surface functionalization. In this context, a proper understanding of the plasma-surface interaction during the deposition process is of great importance. The structure of the first monolayer of deposited matter largely determines the general properties of the deposited surface, as well as its adhesion properties in the case of reversible adhesion. However, the adhesion of molecules in the sub-monolayer range is difficult to study by plasma deposition techniques. In this work, we use molecular dynamics (MD) simulations along with density functional theory (DFT) calculations in order to gain insight into the molecule-surface interactions during the initial stages of the plasma deposition process. To this end, we study the deposition of hexane molecules and fragments onto a pure silicon surface, using both classical and ab-initio MD. The sticking probability of the various species is determined, while systematically varying the incidence angle and the deposition energy in the range of 1 to 10 eV. In addition, we perform DFT calculations to study the chemical bonds between the molecules and the substrate. We discuss the details of our numerical studies and present the obtained results. In general, the simulations show no adhesion of intact hexane molecules, whereas sticking is frequent for molecular fragments, with the probability roughly decreasing for increasing fragment size and angle of incidence.

Molecular dynamic simulations of thermal spikes applied to a thermochronological model

Pedro A. F. P. Moreira, Instituto de Física Glab Wataghin;Sandro Guedes, Instituto de Física Glab Wataghin;Julio C. Hadler, Instituto de Física Glab Wataghin

Zircon (ZrSiO4) is used in thermochronology to unravel thermal histories from mineral samples of geological interest. Uranium fission fragments interact with the zircon lattice resulting in a damage region, called latent track. The lattice can be recovered by natural heating over geological periods (annealing). Many annealing models use this fact to infer thermal histories of minerals such as zircon. In one of these models*, parameters can be determined by computational simulation. Simulations of thermal spikes in Zircon were run out. A threshold energy for fission-track formation could be established with simulation and be used to define one of the model parameters. ____________ * Guedes, S, Hadler, J.C., Oliveira, K.M.G., Moreira, P.A.F.P., Iunes, P.J., Tello, C.A., 2006. Kinetic model for the annealing of fission tracks in minerals and its application to apatite. Radiat. Meas. 41, 392-398.

MD study on low-energy sputtering of carbon material by Xe ion bombardment

Tetsuya Muramoto, Okayama University of Science;Toru Hyakutake, Yokohama National University

The low-energy sputtering of carbon material under Xe ion bombardment is studied through the molecular dynamics (MD) simulation. For the normal incidence of Xe, the MD result of sputtering yield almost agrees with the experimental result by Williams et al. (AIAA-2004-3788). However, the experimental result shows less incident angle dependence than the MD result because it performed on a rough surface. It is found that the sputtered particles has memory of the projectile because the sputtered particles by the low-energy projectile undergo only a few collisions to the ejection.

A cellular Monte Carlo code for the prediction of flux couplings in alloys under irradiation

Maylise Nastar, CEA; Thomas Garnier, CEA

Starting from an atomic interaction model, a new mesoscopic kinetic simulation method is developed. The alloy is represented by a mesoscopic rigid lattice of cells characterized by the number of solute atoms they contain. A Cellular Monte Carlo (CMC) algorithm is designed to perform simulations at this scale. The thermodynamic description relies on a concentration, temperature and cell-size dependent mesoscopic Hamiltonian. Atomic Monte Carlo (AMC) simulations and analytical calculations of free energies are performed to parameterize this Hamiltonian, with an emphasis on finite size effects and boundary conditions. The CMC code simulates equilibrium alloys with thermodynamic properties very similar to the ones predicted by the reference AMC method. The kinetic version of the cellular Monte Carlo code introduces effective exchange probabilities of atoms and vacancies between cells. The latter are defined in terms of the mesoscopic Hamiltonian and the phenomenological transport coefficients. Spectacular flux couplings such as the solute drag by vacancies are simulated. The example of Fe-Cu alloys is presented.

Surface enhancement of defect production in semiconductor nanowires

Kai Nordlund, Department of Physics, University of Helsinki;Wei Ren, Department of Physics, University of Helsinki;Antti Kuronen, Department of Physics, University of Helsinki;Sanni Hoilijoki, Department of Physics, University of Helsinki;Eero Holmström, Department of Physics, University of Helsinki

Semiconductor nanowires (NW) are considered to be of great importance in future nanotechnology applications. Due to the roles of dimensionality and small system size, NWs have great potential for testing and understanding fundamental concepts, like optical, electrical, and mechanical properties. Potential practical applications of nanowires range from field-effect transistors to biological applications. Recent experiments show that irradiation of nanomaterials with energetic ions or electrons induces atomic defect production in the target. This effect can be used to modify the mechanical and electronic properties of the nanomaterials. It is not obvious whether irradiation of a nanowire enhances or reduces the damage compared to bulk. Thus, a fundamental understanding of defect formation under irradiation and how defects affect the material properties, is of great interest. We have used classical molecular dynamics (MD) simulations to examine the irradiation induced defect formation process in small-diameter hexagonal Si and GaN NWs [1,2]. The density distribution of the defects was analyzed to examine the importance of a possible surface enhancement on defect production. We also compared the defect production in NWs with that in bulk which was irradiated by self-recoiling atoms, and compared defect production in Si and GaN NWs under the same Ar irradiation conditions. We found that most of defect productions in the nanowires was concentrated on the surface. The strong surface enhancement caused the defect production of NWs to become greater than bulk in the low energy range. The defect production in the smaller cross-section GaN nanowire showed a greater enhancement than the larger one, because of the larger surface-to-volume ratio of the smaller nanowires. Additionally, defects in the GaN NW caused a small decrease in the elastic Youngs modulus. The defect production of both GaN and Si materials were quite similar and in both of them a strong surface enhancement was observed. Thus, our simulations showed that defect production in semiconductor NWs seem to be in general enhanced at the surface.

Molecular dynamics simulation on threshold displacement energies in lithium metatitanate

Takuji Oda, University of Tennessee, Knoxville;William J. Weber, University of Tennessee, Knoxville;Satoru Tanaka, University of Tokyo

In nuclear fusion reactors, ternary lithium oxides such as Li2TiO3 and Li4SiO4 are planned to be used as tritium breeding materials. During reactor operation, a large number of defects are formed by high-energy particle irradiation. Consequently, the thermal and mechanical properties are degraded, and the tritium release rate, which should be high enough to keep balance between tritium burning amount/rate in the plasma and tritium recovering amount/rate from the breeding materials, declines as irradiation defects trap tritium. Hence, understanding of irradiation damage processes is important to predict long-term material performance realistically. However, detailed damage processes in those materials are not sufficiently understood. Even threshold displacement energies (Ed), one of the key parameters for evaluation of irradiation damage, have not been reported except for LiAlO2 [1]. Therefore, the present study aimed to evaluate Ed in Li2TiO3 by using molecular dynamics (MD) simulation. A potential model developed by Kerisit et al. [2] was utilized in the MD simulations. Although this model is of core-shell form originally, we modified it to a rigid-ion form by neglecting the spring forces and fit it to the ZBL potential model at short interatomic distances (< ~1.2 Å) in order to improve description of collision events. It was confirmed that these modifications hardly affected the thermal and mechanical properties of Li2TiO3 crystals. All simulations were conducted at 0 K under NVE ensemble using DL_POLY code. Since strong dependence of Ed on PKA displacement direction have been found in LiAlO2 [1], Ed was determined over several hundred directions. Lithium basically had lower displacement energies than titanium and oxygen. Replacement-collision sequences were sometimes observed when a lithium atom was the PKA, especially along directions parallel with a and b lattice vectors. The obtained results will be discussed in comparison with LiAlO2 in order to extract general trends in ternary lithium oxides. References [1] H. Tsuchihira, T. Oda, S. Tanaka, Nucl. Instrum. Methods B 269 (2011) 1707. [2] S. Kerisit, K.M. Rosso, Z. Yang, J. Liu, J. Phys. Chem. C 114 (2010) 19096.

Configurational Contribution to the free energy of embedded nanoclusters

M. Posselt, Helmholtz-Zentrum Dresden-Rossendorf, Institute of Ion Beam Physics and Materials Research;A. T. Al-Motasem, Helmholtz-Zentrum Dresden-Rossendorf, Institute of Ion Beam Physics and Materials Research

Nanostructure evolution is a common phenomenon occurring during ion and neutron irradiation as well as during thermal treatment. It is characterized by diffusion and reaction processes that can cause the formation of embedded nanoclusters which often leads to a modification of the materials properties. Multiscale modeling can substantially contribute to a better understanding of nanostructure evolution. Atomic-scale molecular dynamics and kinetic Monte Carlo simulations are applicable on relatively small length and time scales whereas coarse-grained methods such as object kinetic Monte Carlo simulations and rate theory can be used on scales more easily accessible by experiments. The latter methods need a number of input parameters. One of the most important is the free binding energy of a monomer to a cluster which can be hardly obtained by experimental investigations but can be provided by atomistic simulations. The fundamental quantity that must be determined is the free formation energy of the clusters which consists not only of the formation energy but also of vibrational and configurational contributions. The focus of the present work is on the evaluation of the configurational part of the free formation energy. The simple example of coherent Cu nanoclusters in bcc-Fe is considered. First, at T=0 the most stable cluster configurations are determined by Metropolis Monte Carlo simulations and their formation and binding energies are calculated. Second, a modified Wang-Landau Monte Carlo method is employed in order to determine the contribution of all possible geometrical configurations of nanoclusters to the free formation energy. Finally, the total and monomer free binding energies are calculated. It is shown that even at moderate temperatures such as 600 K the configurational contributions to the free formation energy cannot be neglected. The calculation scheme applied in this work can be extended to other types of embedded nanoclusters in solids. The presented method should be especially important for nanoclusters with relatively low formation energies. Further investigations are required in order to estimate the vibrational contribution to the free formation energy and to perform a comparison with the configurational part.

Ion-beam mixing in crystalline and amorphous Si und Ge

Manuel Radek, WWU Muenster, Germany;Hartmut Bracht, WWU Muenster, Germay;Matthias Posselt, Helmholtz-Zentrum-Dresden-Rossendorf, Germany ;Bernd Schmidt, Helmholtz-Zentrum-Dresden-Rossendorf, Germany ;Dominique Bougeard, Universitaet Regensburg, Germany;Eugene E. Haller, Lawrence Berkeley National Laboratory, USA;Kohei Itoh, Keio University, Japan

Molecular dynamics simulations using Stillinger-Weber-type interatomic potentials were performed in order to investigate ion-beam mixing by 400 eV self-ion implantation at different fluences and temperatures. In general the magnitude of mixing in an amorphous structure was found to be higher than in its crystalline counterpart. This supports the results of earlier calculations (K. Nordlund et al., J. Appl. Phys. 83 (1998) 1238). The trends observed in our simulations are compared to our experimental results on ion beam mixing in crystalline and amorphous isotopically modulated Si and Ge multilayer structures.

Modeling of electron and lattice thermodynamics in swift heavy ion-irradiated semiconductors

Tzveta Apostolova, Institute for Nuclear Research and Nuclear Energy, Sofia, Bulgaria;M. V. Ivanov, Facultad de Ciencias Fisicas, Universidad Complutense de Madrid, CEI Moncloa, Spain;Jose Manuel Perlado, Instituto de Fusion Nuclear, Universidad Politecnica de Madrid, Spain;Antonio Rivera, Instituto de Fusion Nuclear, Universidad Politecnica de Madrid, Spain

In order to study the evolution of the strong electronic excitation induced by swift heavy ion irradiation in semiconductors, we have developed a formalism based on Boltzmann’s scattering equation that describes the relative scattering motion of electrons in the conduction band and we have applied it to bulk n-doped GaAs semiconductor. The formalism includes impurity- and phonon-assisted photon absorption processes as well as Coulomb scattering between two electrons. Ion projectiles with intermediate non-relativitic velocities are considered. Charge transfer and ionization across the band-gap are not taken into account at this stage. In addition, we have developed an alternative method to describe electron excitation in bulk GaAs during ion irradiation. It is based on a semi- equivalent process of interaction of the semiconductor material with two orthogonal electromagnetic radiation pulses. The envelopes of the radiation pulses are chosen to simulate the electric field produced by the incoming ion during the ion interaction process. Our approaches provide quantitative information on the electron distribution. Upon irradiation we observe a high-energy tail in the Fermi-Dirac distribution. The thermodynamics of hot electrons is studied by calculating the average kinetic energy (effective electron temperature) as a function of the impact parameter of the ion projectile, its velocity and the charge number. The time evolution of the lattice-temperature is determined by a thermal diffusion equation for phonons containing the rate of the energy-exchange between the electrons and the lattice per unit volume. This equation can be directly derived from the Boltzmann transport theory. The power exchange density between electrons and phonons is obtained by keeping only the leading order interaction between electrons and phonons in the Heisenberg equation for electrons. In this work we will discuss the applicability of our formalism to study the evolution of irradiation-induced electronic excitation and how to link our results to experimental data through phenomenological models.

Effect of ion flux on helium retention in helium-irradiated tungsten

Antonio Rivera, Instituto de Fusion Nuclear, Universidad Politecnica de Madrid, Spain;Gonzalo Valles, Instituto de Fusion Nuclear, Universidad Politecnica de Madrid, Spain;Maria J. Caturla, Universidad de Alicante, Spain;Ignacio Martin-Bragado, Institute IMDEA Materials, Spain

Helium retention in irradiated tungsten leads to swelling, pore formation, sample exfoliation and embrittlement with deleterious consequences in many applications. In particular, the use of tungsten in future nuclear fusion plants is proposed due to its good refractory properties. However, serious concerns about tungsten survivability stems from the fact that it must withstand severe irradiation conditions. In magnetic fusion as well as in inertial fusion (particularly with direct drive targets), tungsten components will be exposed to low and high energy ion (helium) irradiation, respectively. A common feature is that the most detrimental situations will take place in pulsed mode, i.e., high flux irradiation. There is increasing evidence on a correlation between a high helium flux and an enhancement of detrimental effects on tungsten. Nevertheless, the nature of these effects is not well understood due to the subtleties imposed by the exact temperature profile evolution, ion energy, pulse duration, existence of impurities and simultaneous irradiation with other species. Physically based Kinetic Monte Carlo is the technique of choice to simulate the evolution of radiation-induced damage inside solids in large temporal and space scales. We have used the recently developed code MMonCa (Modular Monte Carlo simulator), presented in this conference for the first time, to study He retention (and in general defect evolution) in tungsten samples irradiated with high intensity helium pulses. The code simulates the interactions among a large variety of defects and impurities (He and C) during the irradiation stage and the subsequent annealing steps. In addition, it allows us to vary the sample temperature to follow the severe thermo-mechanical effects of the pulses. In this work we will describe the helium kinetics for different irradiation conditions. A competition is established between fast helium cluster migration and trapping at large defects, being the temperature a determinant factor. In fact, high temperatures (induced by the pulses) are responsible for large vacancy cluster formation and subsequent additional trapping with respect to low flux irradiation.

Single-ion bombardment of the Si(001) surface by keV ions: Sub-surface channeling and the influence of monolayer steps

Yudi Rosandi, University of Kaiserslautern;Herbert M. Urbassek, University of Kaiserslautern

Using molecular-dynamics simulation, the impact of keV ions at glancing incidence onto Si(001) is studied. The role of surface reconstruction for the damage formation is highlighted. The surface usually exhibits a (2x1) reconstruction during which dimer rows, aligned in [110] direction, are created. These dimer rows form sub-surface channels, which are able to trap the incoming ions. We choose two azimuth angles of the impinging ions, parallel and perpendicular to the dimer rows on the target surface. Two different ion species are used as projectiles in order to observe the effect of the ion mass. We find that already on a clean terrace the impact azimuth leads to different impact features, which can be traced back to the geometry of the target surface. At oblique incidence (at angles of ~45°-70° to the surface normal) ions are generally implanted, while for glancing incidence(~ 80°) the probability of the ion being channeled (surface-channeling) comes into play. Our simulation results provide information on the dependence of surface-defect production and sputtering on the incidence which allows one to identify surface-channeling events by their characteristic damage features. At glancing incidence, sub-surface channeling of the ion becomes possible if defects exist on the surface which steer the ion into the channel. We simulate ion impact onto surface monolayer steps to investigate the channeling mechanism. Two types of steps are constructed - perpendicular and parallel to the dimer row - in such a way that the ion impacts perpendicular to the step. Our data demonstrate significant differences with respect to the ion incidence azimuth and mass. Our results provide detailed insight into the mechanism of surface and sub-surface channeling on Si(001), which will be useful for understanding ion-beam-assisted surface modification. In addition, the characteristic surface damage patterns presented here will be helpful for identifying subsurface channeling in experiment.

Effect of temperature on the surface composition of the i-Al-Pd-Mn quasicrystal by LEIS technique

Feridoun Samavat, Bu Ali Sina University;Reza kooroshi mahmood, Bu Ali Sina University

In this paper effect of temperature on the surface composition of the i-Al-Pd-Mn quasicrystal by LEIS technique has been studied. The sample was initially cleaned in vacuum by cycles of helium ion sputtering followed by annealing. A 2keV He ion beam was used. It was found that the Al surface concentration decreases smoothly to the steady state under bombardment as a result of preferential sputtering. The Al/Pd surface concentration ratio remained constant with annealing temperatures between 295K and 575K. At 575K this ratio begins to increase with temperature until around 875K. Lastly, the Al/Pd ratio decreases slightly at temperatures more than 875K.

Nano-hillock formation in diamond-like carbon induced by swift heavy projectiles in the electronic stopping regime: experiments and atomistic simulations

Daniel Schwen, Los Alamos National Lab;Eduardo M. Bringa, Universidad Nacional de Cuyo;Johann Krauser, Hochschule Harz;Alois Weidinger, Helmholtz-Zentrum Berlin für Materialien und Energie;Christina Trautmann, GSI Helmholtzzentrum;Hans Hofsaess, Universität Göttingen

The formation of surface hillocks in diamond-like carbon is studied experimentally and by means of large-scale molecular dynamics simulations with 5×10^6 atoms combined with a thermal spike model. The irradiation experiments with swift heavy ions cover a large electronic stopping range between ~12 and 72 keV/nm. Both experiments and simulations show that beyond a stopping power threshold the hillock height increases linearly with the electronic stopping, and agree extremely well assuming an efficiency of approximately 20% in the transfer of electronic energy to the lattice. The simulations also show a transition of sp^3 to sp^2 bonding along the tracks with the hillocks containing almost no sp3 contribution.

Helium diffusion in tungsten: a molecular dynamics study

Xiaolin Shu, Department of Physics, Beihang University, China;Xiaochun Li, Department of Physics, Beihang University, China;Yi Yu, Department of Physics, Beihang University, China;Guanghong Lu, Department of Physics, Beihang University, China

Because tungsten (W) has high melting point, low-sputter yield, and high sputtering threshold energy for helium (He) and hydrogen (H) isotopes, it becomes an important candidate material as plasma facing materials. However, recent experiments demonstrated that He bubble can be formed on W surface even when W is exposed at low energy He ion irradiation. The W dust due to the break of He bubble would significantly influence on the plasma reaction operation. The He diffusivity in W has a large effect on the formation of He bubbles. Amino studied the diffusivity of 3He in perfect crystal tungsten is D =5.4×10-7exp(-0.28 eV/kT) m2/s measured by the atom probe field-ion microscope. Wagner measured the enthalpy change of migration of 4He in W is 0.24-0.32 eV. Becquart and Zhou calculated the diffusion barrier of He from one stable tetrahedral site (TIS) to the nearest TIS by the first-principles method, which is only 0.06 eV. The experimentally observed much higher diffusion barrier is attributed to a very large binding energy to form a He-He cluster. Here we simulate the diffusion process of He in W by a molecular dynamics (MD) method with self-developed W-H-He analytic bond-order potentials [JNM]. We employ the mean squared displacement (MDS) method to determine diffusivity at different temperatures, and then fit the D-T relationship by the Arrhenius equation to obtain the prefactor and the diffusion barrier. Based on the determined MDS of a He atom, the diffusivity D of He in W is fitted as 3.59×10-8 exp(-0.039 eV/kT) m2/s with the diffusion barrier of 0.039 eV. However, we note that the diffusivity of He is shown to be different at different temperature range. The diffusivity is 9.58×10-9 exp(-0.021 eV/kT) m2/s at the temperature range of 100-300 K, 3.70×10-8exp(-0.062 eV/kT) m2/s at 400-1200 K, and 9.56×10-8exp(-0.168 eV/kT) m2/s at 1300-2800 K, with the diffusion barrier of 0.021 eV, 0.062 eV, and 0.168 eV, respectively.

Ion beam induced surface pattern formation : A travelling wave solution

Roger Smith, Loughborough University, Loughborough, LE11 3TU, UK;Satoshi Numazawa, HZD Rossendorf, 01314 Dresden, Germany

The formation of ripple structures on ion bombarded semiconductor surfaces is examined theoretically. Previous models are discussed and a new non-linear model is formulated, based on the local atomic flow and associated density change in the near surface region. Within this framework ripple structures are shown to form without the necessity to invoke surface diffusion or large sputtering as important mechanisms. The model can also be extended to the case where sputtering is important and it is shown that in this case, certain 'magic' angles can occur at which the ripple patterns are most clearly defined. The results are in very good agreement with experimental observation.

Modelling the diffusion of H and He in delta-Pu

Christopher Scott, Loughborough University;Steven D Kenny, Loughborough University;Marc Robinson, Loughborough University;Roger Smith, Loughborough University;Mark T Storr, AWE, Aldermaston

A computational investigation into the effects of the presence of H and He in Ga stabilised delta-Pu has been undertaken. Understanding radiation effects in Pu is important due to its uses as an energy source and in nuclear weapons. Of particular interest are the diffusion of H and He and the formation of He bubbles. Classical Molecular Dynamics and long time scale on-the-fly kinetic Monte Carlo techniques were applied during the investigation. Interatomic interactions were modelled using the Modified Embedded Atom Method.

Defect formation energies and diffusion barriers have been calculated for H and He in Pu-Ga. Subsequently the diffusivity of H was investigated in a damage free Pu-Ga system over a range of temperatures. Both Molecular Dynamics and on-the-fly kinetic Monte Carlo were used, depending on the temperature of interest.

The diffusion of H and He in the presence of radiation damage, over realistic time scales, was then studied using a hybrid technique. Molecular Dynamics was used to model the cascade events, while on-the-fly kinetic Monte Carlo was used to model diffusion between events. This has enabled us to build a realistic picture of the behaviour of H and He in Ga stabilised delta-Pu subjected to radiation events.

Atomistic modeling of decomposition and segregation kinetics in Fe-Cr alloys

Frederic Soisson, CEA Saclay;Enrique Martinez, Lawrence Livermore National Laboratory;Oriane Senniger, CEA Saclay

High chromium ferritic and martensitic steels are considered as essential materials for several concepts of the next generation of fission reactors as well as for future fusion reactors. The addition of Cr has a positive effect on mechanical, corrosion and radiation resistance properties, so that the stability of Fe-Cr solid solutions versus the decomposition is of great technological importance. In this study, the kinetics decomposition during thermal ageing is modeled by Atomistic Kinetic Monte Carlo (AKMC) simulations, using a rigid lattice approximation with composition dependent pair interactions that can reproduce the change of sign of the mixing energy with the alloy composition – a key feature of Fe-Cr alloy thermodynamic properties. The interactions are fitted on ab initio mixing energies and on the experimental phase diagram, as well as on the migration barriers in iron and chromium rich phases. Simulated kinetics is compared with 3D atom probe and neutron scattering experiments. The modeling of radiation induced segregation phenomena is also discussed.

Defect-interface interactions and radiation tolerance in oxides

Blas P Uberuaga, Los Alamos National Laboratory

It is well established that interfaces and grain boundaries can act as efficient sinks for radiation-induced defects. Exactly how interfaces interact with defects and how this interaction depends on the structure of the interface, however, are still uncertain. Here, we examine coherent hetero-interfaces in oxide thin film bilayers to determine how radiation-induced defects interact with those interfaces and modify the radiation tolerance of the material. In particular, we focus on the interface between thin film oxides and the SrTiO3 substrate on which they are grown. Even though these interfaces are often nearly fully coherent, with no special atomic structure that leads to trap states at the interface, the interface nevertheless greatly influences how the materials on both side respond to the produced defects. Both enhancement and degradation of radiation tolerance is observed, depending on which side of the interface is examined. We complement irradiation experiments with atomistic calculations to gain insight into the origins of the behavior, identifying differences in the bulk behavior of defects on each side of the interface as the important determiners. In particular, differences in chemical potential and bulk migration of defects in each phase are hypothesized to be the controlling factors for this behavior.

Sputtering of a silicon surface: Preferential sputtering of impurities and dimer emission

Maureen L. Nietadi, Physics Dept., University Kaiserslautern, Germany;Michael Kopnarski, IFOS, Kaiserslautern, Germany;Jan Lorincik, Faculty of Science, J. E. Purkinje University, Ústí nad Labem, Czech Republic;Herbert M. Urbassek, Physics Dept., University Kaiserslautern, Germany

Impurities may be contained in Si in small amounts. When performing sputtering experiments, one can measure the fraction of impurities contained in the sputtered flux. If the preferential sputter coefficients of these impurities were known, one could conclude on the stoichiometry of the sample. We employ molecular dynamics simulation to learn about the preferential sputter yields. Linear sputtering theory predicts the sputter yields to depend on the mass and surface binding energy of the impurities only. We use this knowledge to perform simulations on a pure Si target in which only one impurity –with well-defined mass and surface binding energy – is present. 2 keV Ar impact at perpendicular incidence is employed, as it is typical for SIMS or SNMS experiments. To enhance the effectivity of our simulation we use the following strategy: If an atom (let us call it ‘atom i’) has been sputtered, we rerun this simulation, but with exchanging before the start of the simulation atom i by the impurity atom X. We can thus determine the probability that the impurity X is sputtered relative to the probability that the Si atom was sputtered. Additionally we discuss molecular sputter yields.

Defect mobility and recovery within Cu grain boundaries

Louis J Vernon, LANL

The mobility of defects in the neighbourhood of various Cu grain boundaries has been investigated using the adaptive kinetic Monte Carlo method. Large numbers of transitions searches have allowed us to catalog long range Frenkel defect recombination processes through which a defect loaded grain boundary may anneal. The results reveal that the relative recovery ranges for different grain boundary orientations are strongly dependent on the local atomic structure and temperature - it is not necessarily enough to only consider the spontaneous recombination at 0K. Combined with the evolution of defect clusters within the grain boundary we begin to form a more complete picture of nano-crystalline Cu radiation tolerance.

Statistics of primary radiation defects formed in collision cascades in the vicinity of edge, screw and extended dislocations

Roman Voskoboinikov, Australian Nuclear Science and Technology Organisation

Existing programmes of primary damage modelling usually consider collision cascade phenomena in idealized defect-free materials. Their real polycrystalline multiphase microstructure with developed dislocation networks is neglected at the displacement cascade stage of the interaction with fast particles. In relevant studies grain boundaries, precipitates/voids, dislocations etc. are regarded as sinks for point defects and at longer time scales they are typically taken into account in the context of diffusion redistribution of residual radiation defects and alloying elements. In this report we present the results of the computer modelling of the direct interaction of displacement cascades with edge, screw and extended dislocations. The velocity-Verlet molecular dynamics method has been applied for modelling of collision cascades in Al and Ni crystals containing an isolated screw or edge dislocation with 1/2[110] Burgers vector. The simulation programme over a wide range of temperature, 100K<T<600K, and primary knock-on atom (PKA) energy, 5keV<Epka<20keV has been performed. For each (Epka, T) at least 64 different cascades were simulated in the vicinity of the dislocations. In order to quantify the interaction of collision cascades with screw (source of anti-plane strain) and edge (source of plane strain) dislocations in Al we calculated the number of residual vacancies and self-interstitials in the two cases as a function of (Epka, T). The obtained data has been compared with the number of Frenkel pairs created in collision cascades in the pristine material. The same approach has been implemented in order to analyse the interaction of displacement cascades with extended dislocations in pure Ni.

MD Study of Primary Damage in Ti-Al Based Intermetallics

Roman Voskoboinikov, Australian Nuclear Science and Technology Organisation

Along with the reliability, nuclear waste management, longer residence time of nuclear fuel and its extended burn-up, higher operating temperatures determine the sustainable economic and environmental perspectives of nuclear power. The increase of the working temperature requires new refractory materials capable of withstanding both high thermal flux and fast particle irradiation without exhibiting significant degradation of service properties. Ti-Al based intermetallic refractory alloys have already been used in a number of engineering applications in automotive and aviation industries. They can be regarded as candidate reactor structural materials as well if their radiation tolerance is good enough. In order to qualify Ti-Al based intermetallics for use in radiation environment at high temperatures, it is important to quantify any adverse external impact. The experimental work aimed to reach this objective is often long and costly. At the same time, not all phenomena are amenable to experimental investigation. It is therefore essential to complement experiments by using mathematical modelling and scientific computing. Computer modelling by molecular dynamics has been applied to study the radiation damage created by collision cascades in L10 TiAl and D019 Ti3Al intermetallic compounds. Either Al or Ti primary knock-on atoms (PKA) with energy 5keV<Epka<20keV were introduced in the intermetallic crystals at temperatures ranging from 100K to 900K. At least 24 different cascade for each (Epka, T, PKA type) set were modelled in order to simulate a random spatial and temporal distribution of PKAs and provide statistical reliability of the results. The total yield of more than 1500 simulated cascades is the largest yet reported for these intermetallic materials. A comprehensive treatment of the modelling results has been carried out. The number of Frenkel pairs, fraction of Al and Ti vacancies, self-interstitials and anti-sites as a function of (Epka, T, PKA type) is established. Preferred formation of Al self-interstitial atoms has been detected in both TiAl and Ti3Al intermetallics. Typical point defect clusters formed in displacement cascades have been identified and their internal structure and mobility have been examined thoroughly.

An Energy Deposition Profile Model for Low Energy Cluster Impacts with Molecular Solids

Jaydeep Mody, University of Surrey;Roger Webb, University of Surrey

A simple model to describe the desorption of material from keV cluster bombardment of molecular solids can be formulated based upon the deposition and spreading of impacting cluster’s initial kinetic energy is formulated. This simple model assumes that the deposited energy spreads via an isotropic diffusive process. The initial cluster impact is clearly forward directed and hence not isotropic. We use Molecular Dynamics simulation of a benzene molecular solid to compare the behaviour of an energetic fullerene cluster with that of a spherically symmetric distribution of kinetic energy placed in the initially perfect material. It is found that after a short time period the two systems compare favourably suggesting that it is not essential to calculate the initial cluster impact. This potentially allows much simpler energy spreading models to be used to calculate the effects of a cluster impact on a molecular solid.

The effect of flux super focusing in the origin of high-yield shoulders in ion channeling angular scans

Dharshana N. Wijesundera, University of Houston;Ki Ma, University of Houston;Xuemei Wang, University of Houston;Buddhi P. Tilakaratne, University of Houston;Lin Shao, Texas A&M University;Wei-Kan Chu, University of Houston

We have investigated the effect of ion channeling flux super-focusing on the origin of high near-surface shoulders in channeling angular scans of single crystals. By simulating 2 MeV He ion planar channeling in Si {1 0 0}, we observe that at the angle of incidence corresponding to the channeling shoulder, the primary channeling focus overlaps with lattice atoms and dramatically enhances the ion flux density at atomic sites, increasing the ion-atom close encounter probability. We show that the so increased close encounter probability and the resulting enhancement of scattering yield originate high near-surface shoulders in channeling.

Modeling of cluster formation at low energy ion sputtering of Cu (100) and Ag (100)

Ishmumin D Yadgarov, Arifov Institute of Electronics;Vasiliy G Stelmakh, Arifov Institute of Electronics;Sirojidin E Raxmatov, Arifov Institute of Electronics;Akbarali M Rasulov, Tashkent University of Information Technologies;Abdirauf A Dzhurakhalov, University of Antwerp

Cluster emission processes at ion bombardment of solids attract many scientists due to their unique formation mechanisms. Recently it was shown that small clusters are predominantly emitted in the collisional phase of the sputter process (for times up to 1 ps) or due to fragmentation of large cluster while large clusters (with 10 or more atoms) are emitted well after the collision cascade is thermalized, and can be understood in terms of thermal and hydrodynamic processes [1]. In the present work the formation mechanisms of small clusters at grazing ion bombardment are studied. As numerous experimental studies show that cluster formation depends on some conditions: the energy and mass of the bombarding particles, surface structure and binding energy of substrate atoms. The idea of using grazing angle ion sputtering is to transfer the same energy to several neighbor atoms in order to eject them together. We have carried out theoretical studies, which are based on some possible models and mechanisms of cluster formation during ion sputtering of a solid surface. Due to the problems of analytical description of the cluster formation process a considerable attention has been devoted to computer simulation by classical molecular dynamics method. An interaction between atoms is described by well-defined embedded atom model. Calculations were performed for the cases of self sputtering as well as for the mixed combinations of Cu and Ag projectile-solid system. Internal and kinetic energy distributions as well as yields of particles emitted with different sizes are calculated. An influence of the initial energy varying from 0.5 to 5 keV and grazing angle (0-30) of incident particles to the process of cluster formation are analyzed. On the basis of obtained results and performed analysis the nature of cluster formation during grazing ion sputtering of a solid surface is discussed. [1] G. Betz, W. Husinsky, Phil. Trans. R. Soc. Lond. A 362 (2004) 177.

Stress Dependence of Oxigen Diffusion in ZrO2 Film

Yasunori Yamamoto, Graduate School of Energy Science, Kyoto University, Uji-shi, Kyoto 611-0011, Japan;Hirotomo Iwakiri, Faculty of Education, University of the Ryukyu, Nakagami-gun, Okinawa 903-0213, Japan;Kazunori Morishita, Institute of Advanced Energy, Kyoto University, Uji-shi, Kyoto 611-0011, Japan;Yoshiyuki Watanabe, Japan Atomic Energy Agency, Kamikita-gun, Aomori 039-3212, Japan;Yasunori Kaneta, Faculty of Engineering, The University of Tokyo, Bunkyo-ku, Tokyo 113-8656, Japan;Daiji Kato, National Institute for Fusion Science, Toki-shi, Gifu 509-5292, Japan

For the purpose of more efficient use of nuclear resource and effective reduction in spent fuel, the maximum fuel burnup allowed in commercial light water reactors has been gradually increased. One of the main issues for high burnup fuels is the embrittlement of fuel cladding tubes which is caused by absorbing hydrogen. The fuel cladding is composed of zirconium alloys and easily absorbs hydrogen, where hydrogen is produced by the Zr-water oxidation reaction. Therefore, oxide film of zirconium alloys may be useful as an indicator for monitoring the amount of absorbed hydrogen. In order to prevent fuel cladding from hydrogen embrittlement, the oxidation should be controlled and suppressed as low as possible. However, the precise mechanism of oxidation process on the surface of zirconium alloys is not yet understood well. The diffusion of an oxygen atom in an oxide film is considered to be a key to the oxidation process, where the film may act as a barrier against the diffusion. ZrO2 is one of the compositions of Zr oxide film. Here, we considered two typical crystal structures: one is monoclinic and the other is tetragonal. Because oxygen atom is supposed to diffuse in these ZrO2 structures by the vacancy mechanism, evaluation of the formation and migration energies of an oxygen vacancy is important to understand the oxygen diffusion process. In the present study, first principle calculations were performed with a ZrO2 supercell which contains an oxygen vacancy by the SIESTA code [1]. Usually in the fuel cladding, the compressive stress is applied to Zr oxide on Zr metal, because of a difference in equilibrium volume between the oxide and the metal. It is reported in the experiments [2] that an approximate maximum value of stress applied at the oxide/metal interface was 1 GPa. Since an increment of the total energy due to a stress of 1 GPa is only 0.012 eV and 0.013 eV for the monoclinic and tetragonal ZrO2, respectively, the stress seems to have little impacts on the vacancy formation energy. The stress dependence of the vacancy migration energy will be discussed. [1] J. M. Soler, E. Artacho, J. D. Gale, et al., J. Phys.: Condens. Matter, 14 (2002) 2745. [2] J. Godlewski, Tenth International Symposium ASTM STP, 1245 (1994) 663.


Alexander V. Bakaev, Center for Molecular Modeling, Ghent University, Technologiepark 903, B-9052, Zwijnaarde, Belgium;Dmitry Terentyev, Structural Material Group, Institute of Nuclear Materials Science, SCK•CEN, Boeretang 200, Mol, B2400, Belgium;Evgeny E. Zhurkin, Department of Experimental Nuclear Physics K-89, Faculty of Physics and Mechanics, St.Petersburg State Polytechnic University, 29 Polytekhnicheskaya str., 195251, St.Petersburg, Russian Federation

Iron-based materials such as austenitic steels are important for many structural applications requiring high strength and good ductility, such as components in nuclear and fusion setups. With this respect, a good resistance against irradiation is another important requirement for such kind of materials. Neutron irradiation causes degradation of mechanical properties due to the appearance of lattice defects obstructing dislocation movement. In the case of austenitic steels, neutron irradiation mainly causes the production of Frank loops, primary, and later perfect loops, precipitates and formation of voids. The formation and growth of different kinds of lattice defects is determined by its free energy, thus its knowledge allows one to predict the morphology and structure of the defects expected to appear under neutron irradiation. Here, we calculate the formation energy of different radiation defects (such as Frank loops, perfect dislocation loops, stacking fault tetrahedral (SFT) and voids) at zero temperature (T=0К) in Fe-10%Ni-20%Cr model alloy using classical Molecular Dynamics method. In the considered range of the defect sizes (1-10 nm), hexagonal Frank loop with sides orientated along <110> directions were found to have the lowest formation energy among the interstitial types of defects, and SFT among the vacancy type of defects. The obtained formation energy was compared to the predictions derived from the elasticity theory. Overall, we found a good agreement, except for the case of hexagonal Frank loops with sides oriented along <110> directions, for which the elasticity theory overestimates the formation energy. The discrepancy can be explained by the splitting of the Frank dislocation segments into Shockley and stair-rod dislocations. Thermal stability of the radiation defects was also studied by annealing at different temperatures. All kinds of considered defects are found to be stable in the temperature range 300-1200K except for Frank loops with sides orientated along <112> direction. In the latter case, the transformation of the Frank loop to a perfect dislocation loop was observed via nucleation of two partial dislocations and their propagation along the faulted area, above 300K.

Sputtering of Al nanoclusters by 1 – 13 keV monatomic or polyatomic ions studied by Molecular Dynamics simulations

Evgeny E. Zhurkin, Department of Experimental Nuclear Physics K-89, Faculty of Physics and Mechanics, St.Petersburg State Polytechnic University, 29 Polytekhnicheskaya str., 195251, St.Petersburg, Russian Federation;Petr Yu. Grigorev, Department of Experimental Nuclear Physics K-89, Faculty of Physics and Mechanics, St.Petersburg State Polytechnic University, 29 Polytekhnicheskaya str., 195251, St.Petersburg, Russian Federation

The sputtering mechanisms of freestanding and supported on a substrate nanoclusters are of interest from both fundamental and applied points of view. Atomic scale simulations in frame of classical Molecular Dynamics (MD) method were performed to study the sputtering of spherical Al nanoclusters with diameters of 2-10 nm under bombardment by Al1 and Al13 projectiles with energies of 1-13 keV at normal and oblique impact direction. MD code utilizes a many-body interatomic potential which is based on the Second Moment Approximation of the Tight Binding model combined with ZBL potential at short distances. Nanoparticles were randomly rotated prior to each projectile impact. The impact parameter was chosen randomly over the whole cross-sectional area of the projectile-target interaction. The simulation was performed as a sequence of independent impacts (100 events), and each event was traced up to 100 ps. Both monatomic and clusterized yields of the secondary emission are found to be much larger than those for (111) flat surface of the bulk Al (at the equal irradiation conditions). In case of Al1 projectile the sputtering yield was shown to depend weakly on the impact parameter whereas in case of polyatomic projectile the large cluster emission dominates at peripheral impacts. In some events, the target nanocluster receives a backward momentum and therefore its major part (more than 2/3 of the mass) is ejected. These cases are realized either due to transmitting of the incident projectile through the target or under peripheral oblique impact into the “bottom” part of the target nanoparticle, which so produces the secondary emission mainly towards the substrate direction. As a result, the remaining part of the target nanocluster can get some recoil momentum in the opposite direction. The probability of this “recoil effect” is measured up to 45% and 53% for Al1 and Al13 projectiles, correspondingly, and its value decreases with increase of the target nanocluster size. For a given size of the target, the “recoil” effect gets more pronounced in case of cluster bombardment. A restricted number of MD simulations were performed to verify whether the “recoil” effect described above is strong enough to desorb an Al nanocluster off a Al(111) substrate.