Symposium EE12 : Radiation Damage in Materials---A Grand Multiscale Challenge

Understanding the evolution of microstructures in irradiated materials is key to both the prediction of their future behavior and the development of advanced, radiation tolerant materials. Well controlled and carefully tailored ion irradiation has been successful at creating most of the features as well as their time and spatial evolution in LWR core structural materials such as stainless steels and zirconium alloys, and in fast reactor candidate materials such as ferritic-martensitic steels. Evolution of defect cluster distribution, loop size distribution, precipitate formation and growth, segregation to interfaces, and void nucleation and growth have all been captured using ion irradiation. The co-implantation of He to simulate transmutation is important in the procssess of void nucleation and growth, and co-injection of H accounts for uptake from the water in LWR components. Such techniques can also be applied to fuels in which fission gasses combine with radiation damage to play a key role in the evolution of microstructure and mechanical properties. This talk will focus on the application of carefully tailored ion irradiation as a means of capturing the multiscale nature of radiation effects in actinides.

The release of fission gases during normal and accident conditions is of key importance for the safe operation of nuclear fuel. The release of fission gas can result in over pressurization of the fuel clad and can alter heat transfer from the fuel. At lower temperatures, such as those in the periphery of the pellet, the gas diffusivity is found to be athermal and is said to be driven by damage cascades. In fact, it is shown experimentally to be proportional to the fission rate. Underpinning this behavior are atomic scale processes that can be investigated through molecular dynamics. Firstly, potential parameters are developed for Xe and Kr that are consistent with a previously developed manybody potential for UO₂ by force matching between molecular dynamics and density functional theory. Subsequently, we investigate the mobility of fission gases, Xe and Kr, during radiation damage cascades. By examining the effect of PKA energy and of multiple cascades occurring on the same region of crystal we attempt to link individual damage cascade simulations to the experimentally determined relationship between fission rate and gas diffusivity.

The evolution of inert gas bubbles in metals has important implications on the evolution of the mechanical properties of nuclear materials as well as materials in highly irradiating environments, such as those expected in next-generation nuclear reactors. The presence of gas bubbles in metallic lattices can profoundly alter the mechanical properties and strength of materials leading to embrittlement, swelling, and blistering. The behaviors of gas bubbles are thus important components of any evaluation of the effects of irradiation-induced aging in a material. The alpha decay of plutonium in PuGa alloys continually generates inert He atoms within the lattice of the PuGa matrix. In naturally aged Pu specimens, those He atoms form into bubbles, He-filled vacancy clusters, with a characteristic size from 2-10nm. Upon annealing, the He bubbles are subject to temperature induced changes which results in a coarsening of the bubble distribution yielding a lower bubble density but larger average bubble sizes up to 60nm. The formation of the He bubbles in PuGa alloys has been studied by TEM however concern has been raised that the preparation of very thin samples (> 1 micron) needed in the TEM experiment together with low number of voids in any particular TEM images may skew the measured He bubble concentration and distribution. To resolve these outstanding issues we have used a combination SAXS and USAXS to examine the formation and growth of He bubbles in aged and temperature annealed PuGa alloys. Development of non destructive volumetric probes for nuclear materials is needed to confirm TEM results and validate models of Pu aging.

The materials deployed in many nuclear applications suffer from the buildup of helium generated as a result of neutron bombardment. Typically the He is not soluble in the target material and forms nano-sized bubbles within it. Due to the fact that studying the buildup of He bubbles in actual active materials is obviously difficult, surrogate materials and He implantation studies are utilized to understand the underlying effects. In this work we utilize the new ORION Nanofab, an He/Ne and Ga ion beam microscope, to implant He into Cu to develop an understanding of the formation of the He bubble superlattices and their effect on the mechanical properties of the material. In situ TEM nanocompression tests are performed to quantify the changes in the mechanical properties and to observe the evolution of the He bubble structure under stress. In addition, we present first results from correlative microscopy of the He implanted surfaces.

Borosilicate glasses have been recognized as valuable materials for the conditioning of nuclear wastes. An important issue for their long-term behaviour is radiation effects which may impact their performance and stability. To address these concerns, a fundamental understanding of the origin at the atomic scale of the macroscopic property evolutions must be established. Over the last decade, magic-angle spinning nuclear magnetic resonance (MAS NMR) has firmly established itself as one of the most powerful tool to investigate a glass’s structure. It offers several probes of the local structure, nuclei such as ¹¹B, ²³Na, ²⁷Al, ²⁹Si and ¹⁷O, to probe changes either in the glass network or in the alkali distribution.
Recently, using external heavy ions irradiation (Xe, Au, Kr) to simulate alpha decays,[1-3] dramatic changes in the local network structure were evidenced : conversion of tetrahedral BO₄ units into planar trigonal BO₃ units (¹¹B), appearance of high-coordination aluminum units (AlO₅, AlO₆); glass depolymerisation (²⁹Si) and changes in the distribution of alkali cations (²³Na). Additionally, the spectra broaden globally which supports the hypothesis of an increased topological disorder after irradiation. All these structural changes are similar to those observed with increasing the glass temperature or quenching rate and support therefore the model of ballistic disordering fast quenching events which induce a new glassy state with higher fictive temperature. Effects of external electronic (beta) irradiations will be also discussed. If NMR spectra variations show similar trends -but much less pronounced- they are mainly engendered by alkali migration phenomena and formation of molecular oxygen.
Until recently, these studies were limited to externally irradiated samples (enabling the different components of irradiation to be dissociated for their precise investigation), but recently, the first MAS-NMR experiments could be performed on radioactive glasses (doped with 244Cm 0.1 % mol.) paving the way for future MAS NMR examinations of self-irradiation damages in glasses. Experiments were performed at the Joint Research Centre Institute for Transuranium Elements (JRC-ITU) where a commercial NMR spectrometer were integrated with a radioactive glovebox and a MAS commercial probe. First results will be presented. Competitive effects between the recoil nuclei and alpha decays were evidenced and the high resistance of the nuclear waste glasses corroborated.

[1] S. Peuget, C. Mendoza, E.A. Maugeri et al. Procedia Materials Science 7, (2014) 252-261
[2] C. Mendoza, S. Peuget, T. Charpentier et al., Nuclear Instruments & Methods in Physics Research Section B-Beam Interactions with Materials and Atoms 325 (2014) 5-65
[3] S. Peuget, E.A. Maugeri, T. Charpentier et al. J. Non-Cryst. Solids 378 (2013) 201-212.

Because of their internal radiation and their applications, nuclear materials are often subject to extreme radiation fields that affect their structures and properties over time. In plutonium, every atom is displaced from its lattice position once every ten years on average. Although almost all of the displaced nuclei rapidly return to the lattice quickly, some do not, becoming defects in the material. The conventional idea is that these will accumulate randomly, causing increasing disorder in the material that will eventually result in its becoming amorphous. Alternatively, these defects might interact strongly with each other and other defects or inhomogeneities in the material. If they do then the possibility exists that they will stabilize themselves and aggregate into defect-enriched domains until these move so far away from the original composition that they could potentially transofrm into domains with altered structures that are still trapped in the host lattice. This process describes the formation of helium bubbles from the emitted alpha particles, but also other such nanometer and larger scale structures. We have now observed exactly this type of phenomenon in delta plutonium-gallium alloys and also in ion irradiated uranium dioxide. Although delta plutonium eventually succumbs to accumulated radiation damage and does lose order, prior to that stage it displays a cycle based on the formation of particular locally ordered structures that deviate from the fcc one that is its long range average. These alternative structures are similar to the ones observed in new materials that are caused by strong interactions between the alloys atoms that cause it to cluster to form a quasi-intermetallic. Similarly, the propensity of adventitious O to cluster in uranium dioxide is emulated as the result of ion irradiation. Uranyl type species with higher valences are observed just as with oxidation to mixed valence compounds, although in the case of ion irradiation they must be mirrored by lower valence species as well. This complication demonstrates the substantial stability of these non-equilibrium structures. In the case of radiation damage, and also other forms of aging, this phenomenon of the formation of structures on increasingly large length scales is, in fact relatively common. The formation of these structures is coupled with the irreversibility of damage accumulation, with the formation of such structures corresponding to phase transtions. Aging effects may therefore be best described as non-equilibrium thermodynamic processes with these irreversible steps corresponding to critical points on the path.

One challenge for the development of Gen IV nuclear reactors is to improve significantly the effectiveness of the design and selection of innovative fuels. To this aim, multiscale modelling approaches are developed to build more physically based kinetic and mechanical mesoscale models to enhance the predictive capability of fuel performance codes. Atomic scale methods, in particular electronic structure calculations, form the basis of this multiscale approach. It is therefore essential to know the accuracy of the results computed at this scale if we want to feed them into higher scale models.
Electronic structure calculation methods, especially density functional theory (DFT), have been used extensively on molecular and solid systems during the last thirty years. Numerous assessments of these methods have been performed, which show that they are powerful tools yielding precise and predictive results for a large number of solid and molecular systems, therefore contributing to the understanding of numerous phenomena. The application to nuclear materials under irradiation and especially to fuels, however, is more delicate and calls for bespoke developments. A specific assessment of the atomic scale methods for the description of nuclear fuel under irradiation is therefore necessary.
We will present the result of the extensive assessment effort of the results of state-of-the-art electronic calculations on uranium dioxide performed in the framework of the Working Party on Multiscale Modelling of Fuels and Structural Materials for Nuclear Systems (WPMM) of the OECD/NEA.

Models used in fuel performance codes to predict the change in the fuel thermal conductivity are typically empirical fits to experimental data, and are independent of other models such as fission gas or grain size. As part of the Nuclear Energy Advanced Modeling and Simulation program, we are developing a system of new materials models for fuel performance that are based on microstructure rather than burn-up, for use in Idaho National Laboratory’s (INL’s) BISON code. In order to obtain a mechanistic model of thermal conductivity, we have developed a preliminary model that couples the fission gas release model to the thermal conductivity. Atomistic and mesoscale simulations were used to quantify the impact of three distributions of fission gas on the thermal conductivity: dispersed gas atoms, small intragranular gas bubbles, and grain boundary bubbles. The model was implemented in BISON and the results were compared to reactor test data.

In UO₂ nuclear fuel, the retention and release of fission gas atoms such as xenon (Xe) are important for nuclear fuel performance. We use multi-scale simulations to determine fission gas diffusion mechanisms as well as the corresponding rates in UO₂ under both intrinsic and irradiation conditions. Density functional theory (DFT) calculations are used to study formation, binding and migration energies of small clusters of Xe and vacancies. Empirical potential calculations enable us to determine the corresponding entropies and attempt frequencies for migration as well as investigate the properties of large clusters or small fission gas bubbles. A continuum reaction-diffusion model is developed for Xe and point defects based on the mechanisms and rates obtained from atomistic simulations. Effective fission gas diffusivities are then obtained by solving this set of equations for different chemical, irradiation and microstructure conditions using the MARMOT phase field code. Emphasis is put on understanding how the diffusion rates evolve as function of the irradiation dose and its coupling to defect concentrations and microstructure. The predictions are compared to available experimental data. The importance of the large Xe_U3O cluster (a Xe atom in a uranium + oxygen vacancy trap site with two bound uranium vacancies) is emphasized, which is a consequence of its high mobility and high binding energy. However, all simple vacancy-mediated diffusion mechanisms underestimate the Xe diffusivity compared to the empirical radiation-enhanced model used in most fission gas release models. We investigate the possibility that diffusion of small fission gas bubbles or extended Xe-vacancy clusters may be responsible for the radiation-enhanced diffusion coefficient. These studies highlight the importance of U divacancies and a cluster composed of an octahedron coordination of uranium vacancies encompassing a Xe fission gas atom. The latter cluster can migrate via a multistep mechanism with a low effective barrier.

Radiation damage effects in materials and their ability to recover and resist damage is a crucial consideration when designing a nuclear waste host or for any feature exposed to a radiation environment. Structure defects such as interstitials, vacancies or dislocations are all consequences of radiation damage that affect the macroscopic and mechanical properties of the material. Another consequence of radiation, especially in nuclear fuel, is the presence fission products (among them rare gases such as Xe, Kr or He). Microstructural change arises from the diffusion of fission gasses through a lattice causing more defects or the formation of bubbles of trapped fission gasses in voids created by radiation damage. Since practically all crystals contain dislocations any diffusion may contain a dislocation-mediated contribution. In this work, atomistic simulations are used to investigate the influences of a number of dislocations and grain boundaries in UO₂ on He diffusion from 2300 – 3000 K. Their effect on He diffusion properties such as diffusivity, activation energy and anisotropy of diffusion are examined depending on temperature and distance to dislocation cores.

As the burn-up of a nuclear fuel is increased there is a marked increase in both the dislocation density and the concentration of fission products in the matrix. In particular, dislocations with a Burgers vector along <110> axes are formed during high energy events, and have important consequences on fuel properties at a macroscopic scale. Whilst previous works have shown that such dislocations can play a role in xenon bubble nucleation, the aim of the present study is to investigate on their effect on pre-existing Xe bubbles. To this intent, we used empirical potential Molecular Dynamics to simulate the dislocation-bubble interaction depending on bubble size and density. These simulations showed how dislocations can help stabilising bubbles by decreasing their internal pressure, as well as the pinning effect of such bubbles on mobile dislocations.

Technology Computer Aided Design (TCAD), developed for the accelerated modeling-assisted development of semiconductor devices and maturing in the 80’s and 90’s of the previous century, is the mother of what is currently known as “Integrated Computational Materials Engineering (ICME)”. In process TCAD, the different fabrication steps of semiconductor devices such as ion implantation, annealing, or oxidation are modeled with well calibrated models in order to develop the best fabrication recipes while minimizing costly and slow experimentation. Since the modeling methods and tools developed include radiation damage creation (in the form of ion implantation) and damage annihilation, they can be transferred with small modifications to radiation problems in nuclear environments. Since the abundance of experimental data, which exists for silicon-based semiconductor systems, is not available for most nuclear materials, density-functional theory based methods can be used to gain insight into the modeling parameters. In this talk, we will outline basic concepts of process TCAD developed for ion implantation and annealing, and then apply them to the case of light attenuation in sapphire fibers in high-temperature irradiating environments. Such fibers have been proposed as fiber optic conductors and sensors which can still function in conditions well above 1000 ^oC where the traditional silica fibers cease functioning and thus can be used for sensing applications in future-generation high-temperature reactors as well as in accident scenarios.

Bottom-up growth of microscopic pillars is observed at room temperature on GaN irradiated with a Ga⁺ beam in a gaseous XeF₂ environment. Ion bombardment produces Ga droplets which evolve into pillars, each comprised of a spherical Ga cap atop a Ga-filled, gallium fluoride tapered tube (sheath). The structures form through an interdependent, self-ordering cycle of liquid cap growth and solid sheath formation. The sheath and core growth mechanisms are not catalytic, but instead consistent with a model of ion-induced Ga and F generation, Ga transport through surface diffusion, and heterogeneous sputtering caused by self-masking of the tapered pillars. This is the first example of self-assembly driven by FIB chemistry.

Emergent phenomena such as spontaneous pattern formation, self-assembly, and self-organization [1–3] have stimulated much research into the underlying mechanisms, and applications in bottom-up growth [3] at length scales ranging from the atomic to macroscopic. Here, we report a spontaneous, room temperature growth mechanism that yields microscopic pillars each comprised of a solid, tapered, gallium fluoride sheath and a Ga core that protrudes from the sheath and forms a liquid spherical cap at the pillar tip. The growth process was observed on GaN irradiated by a Ga⁺ beam in a gaseous XeF₂ environment. Pillar growth is initiated by the formation of a spherical liquid Ga droplet and concurrent growth of a solid sheath, caused by chemical conversion of liquid Ga to gallium fluoride. Tapered pillars emerge from an interdependent, self-ordering cycle of Ga droplet (i.e., pillar cap) growth and sheath formation. The underlying mechanisms are noncatalytic and physically distinct from others reported in the literature, such as vapor- liquid-solid, solid-liquid-solid, and solution-liquid-solid growth [4].



[1] G. M. Whitesides and B. Grzybowski, Science 295, 2418
(2002); R. M. Bradley and P. D. Shipman, Phys. Rev. Lett. 105, 145501 (2010); S. Facsko, T. Dekorsy, C. Koerdt, C. Trappe, H. Kurz, A. Vogt, and H. L. Hartnagel, Science 285, 1551 (1999).
[2] Q. Wei, J. Lian, W. Lu, and L. Wang, Phys. Rev. Lett. 100, 076103 (2008).
[3] J. V. Barth, G. Costantini, and K. Kern, Nature (London) 437, 671 (2005); W. Lu and C. M. Lieber, Nat. Mater. 6, 841 (2007).
[4] K. W. Kolasinski, Curr. Opin. Solid State Mater. Sci. 10, 182 (2006).

GaN nanowires are comprehensively studied based on the crystal structure, orientation and surface states using transmission electron microscope techniques. Individual nanowires and also nanowires in nanoFET devices are then presented for heavy ion beam radiation in the National Superconducting Cyclotron Laboratory. Radiation damage and effects are studied with and without current transport being present; electron transport is modeled using our unique multi-parameter fitting method. This study presents the fundamental insight of radiation damage on atomic level within nanowires and its effect on the material’s electronic performances.

The formation of stable radiation damage in crystalline solids often proceeds via complex dynamic annealing (DA) processes, involving migration and interaction of ballistically-generated point defects. The dominant DA processes remain unknown even for crystalline Si, which is the best studied and arguably the simplest material. Here, we use a pulsed ion beam method to study defect dynamics in Si as a function of temperature. We measure the effective time constant of defect interaction by studying the dependence of damage on the passive portion of the beam duty cycle. Results reveal two well-defined Arrhenius regimes with unique activation energies. These activation energies suggest that, for the irradiation conditions studied, inter-cascade defect interaction is dominated by migrating self-interstitials and vacancies. This work was performed under the auspices of the U.S. DOE by LLNL under Contract DE-AC52-07NA27344.

Thermoelectric generators (TEGs) have potential applications in extreme environments of high radiation flux, such as nuclear reactors and space. TEGs could be used in nuclear power plants to power remote sensor networks, offering the potential to expand remote monitoring of facilities for increased safety and cost savings. In space, TEGs can be powered by the thermal energy of radioactive materials, providing electricity for decades. In recent years, thermoelectric materials have experienced a significant improvement in performance mainly due to nanostructuring to reduce thermal conductivity while maintaining high electrical conductivity. Despite these advances in material performance, the effect of radiation on nanostructured thermoelectric materials remains largely unexplored. Under the influence of radiation, it is possible to have exchanges in atomic positions as well as structural changes in bond configuration in thermoelectric materials, thereby altering their properties and performance.

The effect of proton and ion-beam irradiation on nanostructured half-Heusler and bismuth telluride materials is presented. Seebeck coefficient, electrical conductivity, and thermal conductivity are compared before and after irradiation. To understand the fundamental mechanisms of potential radiation damage, high-resolution concurrent in-situ TEM is used to examine the crystal structure and characterize polyhedral configurations within the lattice during the impact of ion radiation. Since the goal is to design materials that can withstand radiation environments, it is desirable to quantify local thermoelectric properties and correlate these with the above-mentioned microstructure characterizations. A novel scanning thermal microprobe technique was developed to simultaneously characterize Seebeck coefficient and thermal conductivity with sub-micron spatial resolution. In this way, essential understanding is gained of the nanoscale original of thermal and thermoelectric property changes due to irradiation, which will in turn allow for informed material and device design for radiation environments.

Particle irradiation is an effective method for manipulating properties of individual carbon nanostructures such as nanoparticles, nanotubes, and graphene sheets. This potential, however, remains essentially unexplored for macroscopic 3D assemblies of cross-linked nanostructures. Here, we study a family of open-cell monolithic nanoporous carbons (aerogels) exposed to bombardment with 1-2 MeV noble gas ions at room temperature. We focus on the highly unusual behavior of dopant retention. Contrary to expectations based on diffusion-limited processes, results show that the noble gas retention is independent of ion energy and the average diameter of nanoligaments but scales nonlinearly with the pore size and, hence, monolith density. These findings have important implications for understanding radiation processes in unconstrained nanostructures and for designing the next generation of swelling-resistant nuclear materials.

This work was performed under the auspices of the U.S. DOE by LLNL under Contract DE-AC52-07NA27344.

Irradiation enhanced precipitation hardening is the primary cause of in-service embrittlement of reactor pressure vessel (RPV) steels. Odette and Chao long ago predicted that at very high fluence Mn-Ni-Si Precipitates (MNSPs), so called late-blooming phases, would form large mole fractions of nano-scale precipitates, even in low Cu RPV steels. Since these phases could result in severe and unanticipated embrittlement, and since they are not treated in current regulatory models, MNSPs may limit currently planned light water reactor plant life extension. Unfortunately, due to their slow formation, MNSPs can only be experimentally investigated by analyzing very limited high fluence-low flux surveillance data, combined with accelerated irradiation experiments that require interpolation and extrapolation to service conditions. Thus accurate models are an essential tool to allow prediction for the multitude of RPV conditions relevant to light water reactor life extension.

In this talk we describe a cluster dynamics model we have developed to predict the formation and evolution of MNSPs under a range of composition, temperature, flux, and fluence. The model integrates CALPHAD thermodynamics, literature thermal diffusion coefficients, and fitted parameterizations for radiation-enhanced diffusion, interfacial energies, and homogeneous and heterogeneous nucleation mechanisms. Insight and quantitative model parameterization are provided by comparisons of experimental observations and model predictions for a range of both irradiation and post irradiation annealing conditions. We demonstrate known equilibrium intermetallic Mn-Ni-Si phases are thermodynamically stable in dilute Fe-Mn-Ni-Si alloys at RPV relevant temperatures and that our thermodynamic models can be used to predict precipitate volume fraction and composition for highly irradiated materials.[1-3] The model demonstrates a critical role for heterogeneous nucleation for systems with weaker precipitation driving forces, and suggests that some RPVs may experience significant embrittlement due to MNSPs formed during life-extension.


[1] W. Xiong, H. Ke, R. Krishnamurthy, P. Wells, L. Barnard, G. R. Odette, and D. Morgan, Thermodynamic models of low-temperature Mn–Ni–Si precipitation in reactor pressure vessel steels, MRS Communications, p. 1-5 (2014).
[2] P. B. Wells, T. Yamamoto, B. Miller, T. Milot, J. Cole, Y. Wu, and G. R. Odette, Evolution of manganese–nickel–silicon-dominated phases in highly irradiated reactor pressure vessel steels, Acta Materialia 80, p. 205-219 (2014).
[3] D. J. Sprouster, J. Sinsheimer, E. Dooryhee, S. K. Ghose, P. Wells, T. Stan, N. Almirall, G. R. Odette, and L. E. Ecker, Structural Characterization of Nanoscale Intermetallic Precipitates in Highly Neutron Irradiated Reactor Pressure Vessel Steels, To be publised in Scripta Materialia (2015).

Successful demonstration of energy producing burning plasmas in next-step magnetic fusion experiments requires a solution to the challenges of plasma-material interactions in a radiation damaged environment. During their operational lifetime, the plasma facing wall and high heat flux components in a magnetic fusion reactor must function in an energetic neutron environment that induces several 10s of displacements per atom (dpa) on components that are exposed to severe hydrogenic and helium particle fluxes and very high steady-state and transient thermal loads. These components must avoid excessive erosion and trapping of tritium, and avoid swelling and embrittlement due to H/D/T and He absorption in the material while allowing the removal of the heat load. The near surface (first few nm’s to 10’s of microns) of plasma facing surfaces present a particular challenge due to the combination of displacement damage, D, T and He ion implantation and subsequent diffusion, trapping, and void and bubble formation in this region that can lead to profound changes in the continuum thermo-mechanical material properties in this zone.
This talk presents results on D retention, He nano-bubble formation, thermal conductivity, hardness and elasticity in tungsten materials that have undergone a combination of divertor-like plasma and heavy ion-beam displacement damage exposures. Nuclear reaction analysis (NRA) and thermal desorption spectroscopy (TDS) are used to characterize the effect of displacement damage at elevated substrate temperatures on D implantation, diffusion, trapping and retention. Nano-indentation and nano-scale thermal diffusivity studies provide thermo-mechanical data in similar samples.
Plasma ions implant in the first few nm of the surface, and then diffuse into the bulk. By controlling the dpa from Cu ion beams in the first micron of the surface, a controlled number of vacancy trap sites are created, and D diffusion and trapping is then studied. At lower exposure temperatures (300-600 K)he trapped D inventory increases as expected with dpa, Experiments at higher temperatures show evidence for vacancy annealing and reduction in D retention, with nearly no retention observed for samples damaged at 1000K. A 1D diffusion model with distributed trap sites can reproduce the observed D spatial profiles at low temperatures. Radiation damaged surfaces exhibit a significant (factor of >3x) reduction in thermal conductivity and an increase in hardening. We discuss these initial experimental results, plans for future experimental and damage modeling work, and implications for the performance of plasma-facing components tritium inventory management, and fuel self-sufficiency in next-generation fusion energy systems.

*Submitted for the UCSD-LANL Collaboration on PMI in Radiation Damaged Materials: R. Doerner, R. Chen, S. Cui, J. Barton and M. Simmonds, G.R. Tynan (UCSD) and Y. Wang, N. Mara and S. Pathak (LANL)

Beryllium is a strong candidate first wall material for the ITER nuclear fusion reactor, and has been proposed as a neutron multiplier for tritium breeding in the future DEMO reactor. As such, the radiation response of this material must be understood. Atomic scale molecular dynamics simulations of radiation damage have been performed on Be to calculate the threshold displacement energy. Two approaches were considered: direct threshold displacement simulations along a geodesic projection of directions and damage cascade simulations from which peak and residual defect concentrations can be calculated. This allows for a comprehensive investigation of the directional dependence of the threshold displacement energy. Based on these results, we offer a new energy dependent damage efficiency factor for the Norgett, Robinson and Torrens model of radiation damage.

Nuclear fusion is a promising path towards future clean energy production. However, finding materials that withstand extreme conditions due to plasma-surface interactions (PSI) is an outstanding challenge, even more pronounced by the lack of available facilities to experimentally determine material response to such extreme environments. This emphasizes the need for physically-based, predictive multiscale materials modeling. In this work, we use Molecular Dynamics (MD) and Kinetic Monte Carlo (KMC) to tackle two key questions in fusion research: tungsten (W) fuzz formation and beryllium (Be) erosion.

In our first study, a yet not fully understood and possibly harmful phenomenon is addressed in a bottom-up approach: fuzz-like nano-morphology formation in W when it is exposed to helium (He) plasma [1]. Our MD simulations reproduce the experimentally found square root of time dependence of the surface growth, hinting that the driving mechanisms for fuzz formation onset (surface growth by bubble formation and coalescence, consequent loop-punching and surface lowering by bubble rupture) were successfully identified. To follow the long-term evolution of the system, these mechanisms were implemented in an in-house developed KMC code. The outcome is in quantitative agreement with experiments, in the surface growth time dependence, its rate and surface morphology. Thus, we explain W fuzz growth be an stochastic growth process caused by the balance of the above mentioned surface growth and lowering mechanisms.

As a second exercise, we follow up on recent experiments [2,3] and MD simulations [4], showing that Be erosion by deuterium (D) plasma critically depends on surface temperature and the related D concentration. Despite its relevance for reliably predicting material lifetime and fuel retention, little is understood how these three parameters relate under reactor relevant conditions. A top-to-bottom multi-scale scheme is presented here to address these uncertainties. The KMC code MMonCa is used to estimate equilibrium D concentrations in Be at different surface temperatures and defect concentrations. Then, mixed Be-D surfaces – according to KMC profiles – are generated and irradiated by D in MD, to calculate the resulting Be-D molecular erosion yields. With this new database, impurity transport studies aiming to estimate wall lifetime and fuel retention can be revisited with higher confidence.

[1] M.J. Baldwin et al, NF 48 3 (2008) 035001
[2] D. Nishijima et al., Plasma Phys Contr F 50 (2008) 125007
[3] S. Brezinsek et al., NF 55 (2015) 063021
[4] E. Safi et al., JNM 463 (2015) 805-809

Metallic alloys used as structural materials in the nuclear core of pressurized water reactors suffer from irradiation creep deformation. A proper understanding of the mechanisms which control the deformation is essential in order to predict the dimensional changes under irradiation. At the macroscopic scale, many experimental data are available. However, the microscopic mechanisms are still not yet fully understood.
In this study, a novel approach based on the testing of on-chip thin freestanding structures is evaluated. This on-chip test method, developed at Université catholique de Louvain, is for the first time used in the context of irradiation studies. An elementary test structure is composed of three main elements: (i) a thin specimen of the material of interest, (ii) an actuator layer of silicon nitride with strong internal tensile stresses to deform the attached specimen layer and (iii) a sacrificial layer of silicon dioxide to release the test structure from the underlying substrate. The small thickness of the material, below a few hundreds nanometers, allow full and homogenous irradiation by heavy ions with a kinetic energy of a few hundreds keV.
After adapting the method to in situ irradiation conditions, sets of test structures were successfully used to assess the room temperature creep behaviour of PVD copper films with a thickness of 200 nm and 500 nm. The contribution of the irradiation creep to the deformation has been quantified by performing several irradiation steps up to a dose of 3 dpa and measuring after each step the deformation of the specimens.
Preliminary results show that while the creep behaviours before and after irradiation are very similar, the strain rate under irradiation is several times higher than out of flux. The activation volumes and the strain rate sensitivity were extracted from these experiments. Theses quantities were found to be nearly identical for the unirradiated samples and the irradiated ones. Hence the same mechanisms are thought to take place in both cases. The study of the mechanisms responsible for irradiation creep is currently in progress.
In parallel to these nanomechanical tests, the microstructure of the copper film before and after irradiation is characterized by Transmission Electron Microscopy. In particular, the influence of the grain size on the irradiation defects is investigated. No significant effect on the irradiation defects for grain size between 100 nm and 10 µm is found in this study.

Silicon carbide (SiC) has been attracting an increasing interest for many applications in extreme environments such as structural components in fission and fusion reactors or for microelectronics devices. For these applications, a comprehensive understanding of its behaviour under ion irradiation appears as a fundamental issue. In this work, we present a combined experimental and computational study of both the amorphization and recrystallization processes that can take place in SiC under ion irradiation.
First, 3C-SiC single crystals were irradiated with 100 keV Fe ions at different fluences up to full amorphization, and characterized using ion channelling (RBS/C) and X-ray diffraction. Strain and damage levels have been monitored and compared to values obtained from molecular dynamics simulations of tailored, defective SiC cells. A good agreement was obtained between experimental and computed results. For instance, disorder accumulation kinetics were found to be very similar, with a two-step process for both approaches, and a same maximal lattice volume swelling of ~5% was observed. Furthermore, a stimulated amorphization process has been confirmed and was attributed to the elastic energy stored in the defective layer containing irradiation defects.
Second, damaged 3C-SiC single crystals have been submitted to swift heavy ion irradiation (870 MeV Pb ions) and an annealing effect has been evidenced. This effect has been observed experimentally using RBS/C and transmission electron microscopy, and reproduced by MD calculations combined with a thermal spike modelling. It is found that annealing takes place only at the amorphous/crystalline interface in fully amorphized layers, while recovery occurs over the entire damaged layer when this latter is initially, only partially amorphized. Cross-sections for recrystallization, as well as maximum recrystallization levels have been estimated in both cases. These parameters clearly depend on the defect spectrum.

Since high purity SiC exhibits inherent irradiation tolerance, its potential utilization across a number of fission and fusion platforms is the subject of much research today. Whether utilized as a structural material or a component of the integral fuel system, presence of gradients in temperature and flux may result in transient or permanent differential swelling in this material. This is the case since the magnitude of swelling strain is sensitive to temperature. Coupled with the high value of elastic modulus in SiC, this differential swelling strain in turn results in large stresses in the material. If a robust creep mechanism is present, it can significantly alleviate these stresses. However, one can prudently ignore thermal creep at temperatures below 1400°C and is only left with irradiation creep at temperatures typical to fission reactor systems. This study reports preliminary results from instrumented in-pile tests on high purity chemical vapor deposited SiC specimens at 300°C. The specimens are maintained in stress-free or stressed (100 MPa) conditions to determine the swelling and creep strains separately as a function of neutron dose. Although a strong dose dependence of creep compliance is observed, consistent with prior observations, the estimated magnitude of creep compliance is much larger than what is reported previously in literature. Early results from the ongoing tests are complemented by a series of discussions that intend to interpret the source of this discrepancy.

Silicon carbide is known to be a highly irradiation-stable materials. However, the relative stability of the two primary SiC crystal structures has been called into question. As part of this work the fundamental swelling behavior leading to amorphization of various crystals of SiC is presented. Single crystal a-SiC (6H), b-SiC (3C), highly faulted polycrystalline CVD b-SiC, and single crystal Si have been irradiated in a fast neutron spectrum near 60°C from 5 x 10²³ to 2 x 10²⁶ n/m² (E>0.1 MeV), or about 0.05 to 20 dpa, in order to study the effect of irradiation on swelling. Single crystal and powder diffractometry, TEM, as well as precise bulk density measurements have been carried out to observe and compare irradiation-induced microstructural evolution. For all neutron doses where the samples remained crystalline all SiC materials demonstrated equivalent swelling behavior. Moreover the 6H-SiC expanded isotropically. The magnitude of the swelling followed a ~0.8 power law against dose. Silicon demonstrated nil swelling over all doses. An extraordinarily large ~7.8% volume expansion in SiC was observed prior to amorphization. Above ~ 0.9 x10²⁵ n/m² (E>0.1 MeV) all SiC materials became amorphous with an identical swelling: a 11.7% volume expansion, lowering the density to 2.84g/cm³. The amorphous density was the same at the 2 x 10²⁵ and 2 x 10²⁶ n/m² (E>0.1 MeV) levels. The swelling rates and critical threshold for swelling will be discussed in terms of historic data and amorphization models.

Silicon carbide (SiC) is a good candidate for nuclear applications, fusion reactors as well as Gen IV reactors, because of its interesting properties, such as low activation, high temperature resistance and chemical inertness.
3C-SiC samples were prepared by Spark Plasma Sintering (SPS) from a 15-nm-nanopowder. A microstructure with average grain size of about 70 nm was obtained with a densification ratio of about 95%. SiC pellets were then implanted with 800 keV ¹²⁹Xe⁺⁺ ions at two fluencies: 5.10¹⁵ at.cm^-2 and 1.10¹⁷ at.cm^-2 to simulate gaseous fission product presence near the surface. The projected range (Rp) was 207 nm and the maximal concentration was about 0.4% and 8% respectively. Implantations were carried out at room temperature. Implanted samples were characterized by ion beam analysis (Rutherford Backscattering Spectrometry (RBS)) and electronic microscopy. Noticeable changes in surface morphology are observed after implantation at high fluence with a global increase of surface roughness due to swelling. The composition of the material is almost not modified after implantation at low fluence but a strong oxidation occurred at high fluence. In each case, xenon retention measured by RBS is almost total. Transmission Electron Microscopy analyses show a complete amorphization of the material even at low fluence. At high xenon fluence, amorphous SiO₂ is formed in the implanted region even if isolated SiC phase remains in some places. Large gas bubbles are also observed in the implanted region.

Most of the studies investigating the effects of radiation on the structure and properties of various materials have been performed on single crystals or sintered pellets having large grain sizes. However, there are some specific applications that require the use of thin films to protect some areas or encapsulate nuclear fuel. Such thin films are polycrystalline or even nanocrystalline. Recent investigations of the effects of radiation on such films found differences with respect to the results obtained for single crystal samples. The small grain sizes in such films allow for short diffusion distances of the created defects before encountering grain boundaries that act as sinks. The role of strain present in thin films should also be taken into account when investigating the lattice parameter changes induced by various defects. The Pulsed Laser Deposition (PLD) technique is very suitable to grow nanocrystalline thin films of almost any materials starting from inexpensive targets. By changing the deposition parameters, films possessing different chemical compositions and/or structures could be easily obtained. The surface morphology of the deposited films is very smooth, allowing for the use of characterization techniques such as X-ray reflectivity, grazing incidence X-ray diffraction, X-ray photoelectron spectroscopy, or nanoindentation that all possess depth resolutions of the order of few nm. The effect of ion irradiation on the structure, chemical composition, mechanical, optical and electrical properties of ZrC and ZrN thin films grown by PLD on Si substrates were investigated and compared with those obtained on single crystalline materials.

The first stages of radiation damage are characterized by fast moving particles, primary knocked atoms, and a cascade of electron and lattice excitations. In this atomistic picture, molecular dynamics (MD) would be, and often is, the fundamental tool to simulate these events. Electronic structure-based, ab initio molecular dynamics (AIMD) has seen an immense progress in the last 30 years. However, the investigation of radiation damage has been historically outside the realm of application of AIMD for diverse reasons; in part due to the limit simulated system sizes (up to few hundreds of atoms), and more fundamentally because of the lack of electronic excitations. (AIMD is usually formulated for the ground state or some well defined thermal state, i.e. adiabatic electron evolution.) For these reasons, empirical methods dominated the field, but not without assumptions that wouldn't be necessary with a full AIMD (for example, charge state, dissipation thresholds, extrapolation of force fields under extreme conditions and transferability).

Finally, in the later years a combination of algorithmic development and computational power has made possible for AIMD to include non-adiabatic explicit dynamics on the quantum electrons in the simulations, via time-dependent density functional (TDDFT) and Ehrenfest dynamics (classical ions). In this presentation, I will describe our contributions to the field, and how these new simulations capabilities can start answering, with a minimal set of hypothesis, questions regarding electronic stopping power of swift ions, the role of semicore electrons, the quantification of dynamical charge states, Coulomb explosion, and the rising of non-adiabatic (e.g. velocity dependent) forces between ions. We show predictions of AIMD in the field of radiation damage, in diverse systems, protons, alpha particles, and self irradiated metals and insulators. Also, we show that AIMD, although limited to fraction of picosecond time scales and tens of angstroms, can still inform empirical models (for large scale simulations) about the effects of electron-ion interaction from the stopping power energetic regime (initial energy deposition) down to the electron-phonon regime (temperature equilibration).


Work supported by the Energy Dissipation to Defect Evolution Center (EDDE), an Energy Frontier Research Center funded by the U.S. Department of Energy, Office of Science.

The aggregation of radiation-induced point defects in materials is a critical step in microstructural evolution, and consequently can have significant effects on the thermo-mechanical properties of materials. In iron-based reactor pressure vessel steels, radiation can creates both interstitial loops and voids. In addition, the radiation also accelerates the precipitation of low-solubility impurities or alloy elements such as copper. These radiation-induced defect clusters and precipitates can cause radiation hardening and embrittlement of the steels, which are the primary concerns for the consideration of the life extension of current reactors. Here we use mean-field cluster dynamics simulations to model the evolution of interstitial and vacancy clusters in iron. The defect and cluster energetics and kinetics are obtained from atomistic simulations or experiments. In addition, the radiation-enhanced precipitation of copper clusters will also be modeled with cluster dynamics simulations. The obtained size distribution of defect clusters and precipitates will be used as input for Orowan’s model to predict the radiation hardening by individual type of clusters. Finally, the superstition laws will be applied to predict the irradiation-hardening by these microstructural features under different irradiation conditions.

This work investigates the mechanisms controlling irradiation induced microstructural evolution in iron chrome alloys over a wide range of temperatures. Cluster dynamics is an expansion of simpler rate theory models which tracks the populations of defect clusters from production in cascades to annihilation at sinks or mutual interactions to form observable radiation damage features such as dislocation loops and voids. Primary damage is introduced with a multiscale approach by using a database of molecular dynamics simulations of displacement cascades. Insights from ab initio calculations, molecular dynamics studies, and observation in a variety of experimental techniques are combined to inform the fundamental properties of the clusters, such as mobility and configuration. The model can accommodate various assumptions about the kinetics of defect clusters. Particular attention is given to interactions with trapping sites and one dimensional diffusion of crowdion bundles. The performance of the model is evaluated by comparison with experimental data, particularly in situ heavy ion irradiation studies.

To understand radiation effects in single-phase concentrated solid-solution alloys (also referred as high entropy alloys), irradiation-induced point defect evolution in fcc pure Ni, Ni0.5Fe0.5, and Ni0.8Cr0.2 was studied using molecular dynamics simulations. Interstitials form dislocation loops of 1/3 <111>{111}-type, consistent with our irradiation experimental. The kinetics of formation is considerably slower in NiFe and NiCr than in pure Ni, indicating that migration barriers and extended defect formation energies could be higher in the alloys. Consequently, while larger size clusters are formed in pure Ni, smaller and more clusters are observed in the alloys. Surviving Frenkel pairs are composition dependent and are largely Ni dominated.
The vacancy diffusion leads to formation of stacking fault tetrahedra (SFT), also consistent with our experiments. The formation and growth of SFT are captured by vacancy cluster diffusion and aggregation mechanisms in Ni. The vacancy-tetrahedron acts as a nucleation point for SFT formation. Simulations also show that perfect SFT can grow to the next size perfect SFT via a vacancy aggregation mechanism. The stopping and range of ions in matter (SRIM) calculations and transmission electron microscopy (TEM) observations reveal that SFT can form farther away from the initial cascade-event locations, indicating the operation of diffusion-based vacancy-aggregation mechanism.
This work was supported by Energy Dissipation to Defect Evolution (EDDE), an Energy Research Frontier Center supported by the U.S. Department of Energy, Office of Science, Basic Energy Sciences.

The in-service behaviour of structural materials is a key issue for the successful development of advanced nuclear systems, such as fusion machines and generation IV reactors. ODS steels and Ferritic Martensitic (F/M) steels are potential candidates. These alloys, derived from FeCr alloys, will be submitted to irradiation by intense fast neutron flux, which will lead to the formation of point defect clusters (dislocation loops, voids, He bubbles) and induce intergranular segregation and/or precipitation of Cr. These microstructural modifications are among the main causes of hardening and embrittlement under irradiation

Since neutron irradiations are long and access to facilities is restricted, ions are often used to simulate neutron damage. However, irradiation conditions depend on particle type and flux.
Differences in irradiation conditions (dose rate, recoil spectrum, damage morphology etc.) may have a pronounced effect on microstructural evolution.We report here on Atom Probe Tomography (APT) and Transmission Electron Microscopy (TEM) investigations carried out in a Fe-15at.%Cr high purity alloy irradiated with 1 MeV electrons at 300°C to different doses up to 0.7 dpa using a High Voltage Electron Microscope (HVEM). While to our knowledge Cr–rich α’ precipitation has so far never been detected in FeCr alloys irradiated with heavy ions, α – α’ decomposition was observed under neutron irradiation and in all our electron irradiated samples. Based on the analysis of APT data, the unmixing kinetics was determined as a function of irradiation dose. In addition, the dislocation loop microstructure was characterized by TEM in order to detect a possible correlation between point-defect clusters and the chemical evolutions. These observations are compared to APT results obtained on the same material irradiated with 2 MeV Fe²⁺ ions to identical doses and at the same temperature and dose rate. To better understand the effects of dose rate, ion irradiations up to a high dose using a high dose rate have also been carried out. The results will be compared to published neutron irradiation data.

Stacking fault tetrahedra (SFTs) are ubiquitous defects in face-centered cubic metals. They are produced during cold work plastic deformation, quenching experiments or under irradiation. From a dislocation point of view, the SFTs are comprised of a set of stair-rod dislocations at the (110) edges of a tetrahedron bounding triangular stacking faults. These defects are extremely stable, increasing their energetic stability as they grow in size. At the sizes visible within transmission electron microscope they appear nearly immobile. Contrary to common belief, we show in this report, using a combination of molecular dynamics and temperature accelerated dynamics, how SFTs can diffuse by temporarily disrupting their structure through activated thermal events. Moreover, we demonstrate that the diffusivity of defective SFTs is several orders of magnitude higher than perfect SFTs, and can be even higher than isolated vacancies. We show how SFTs can coalesce, forming a larger defect in what is a new mechanism for the growth of these omnipresent defects. Finally, we analyze how the mobile SFTs interact with various types of free surfaces, dislocations and interfaces in Cu and Cu-Nb systems and how that interaction modifies the final microstructure. We observe a direct relation between the energetics of a single vacancy interaction with the studied extended defects and the propensity for the SFT to be absorbed. Using mesoscale modeling, we show how the fact that SFTs can migrate might influence the system microstructure and potentially other important observables of interest such as the void denuded zones around defect sinks.

The rate theory method in its mean field approximation has been the workhorse of the irradiation damage community for over five decades to perform damage accumulation calculations. Although computationally efficient, mean field rate thepry (MFRT) suffers from the problem of combinatorial explosion, i.e. an exponential growth in the number of ODEs with the number of species dealt with in the models. New paradigms, mostly motivated by fusion energy requirements, have revealed strong synergies among multiple species (e.g. He and H, in addition to lattice damage), which require an explicit consideration of all the different species involved and their interactions. In addition, doses of up to tens and even hundreds of dpa must be accessible to the calculations in order to provide long-term microstructural evolution in these scenarios. Here we describe "stochastic cluster dynamics" (SCD), a methodology that solves the combinatorial explosion bottleneck by populating the coefficient matrix one event at the time and bypassing the need to anticipate the number and size of damage species a priori, as in MFRT. SCD uses a kinetic Monte Carlo algorithm to integrate the system of ODEs in the mean field approximation, thus allowing for natural stochastic and volume fluctuations and enabling a more meaningful comparison with experimental data. Calculations of damage/He/H implantation in Fe as well as fusion neutron irradiation simulations in W will be presented.

Simulation of primary radiation damage formation in metals typically involves collisions between atoms with kinetic energies up to a few hundred keV. During these collisions, the distance between two colliding atoms can approach 0.05 nm. For such small atomic separations, interatomic potentials fitted to equilibrium properties tend to significantly underpredict the potential energy of the dimer. The common practice to enable molecular dynamics simulations of high-energy collisions involves using a screened Coulomb pair potential to describe the high-energy interactions and to smoothly join this to the equilibrium potential. However, there is no accepted standard method for choosing the joining parameters and defect production has been shown to be sensitive to how the joining is done. A new procedure has been developed which involves the use of ab initio calculations to determine the magnitude and spatial dependence of the close-pair interactions, and systematic criteria for choosing the joining parameters. Results are presented for the case of nickel which demonstrate the use and validity of the procedure

Development of damage-tolerant structural materials against radiation environments plays an essential role in improving the capability and reliability of nuclear power generation. While the majority of current researches focus on strategies to increase sinks for defect annihilation, single-phase concentrated solid solution alloys, a novel class of materials containing multiple principal elements, may provide a fundamentally new avenue to enhance radiation damage tolerance via increasing compositional complexity. In this presentation, defect evolution in several Ni-based concentrated solid solution alloys (NiCoFeCr, NiCo, and NiFe) is characterized by using in situ ion (1 MeV Kr⁺, 500 °C) and electron (400~1250 kV, 400 °C) irradiation inside a transmission electron microscope. Defects produced by ion irradiation include full dislocation loops, interstitial-type Frank loops and stacking fault tetrahedra. By contrast, defects produced by electron irradiation are interstitial-type Frank loops and polygonal full loops. During ion irradiation, the defect density steadily increases with damage level (up to 2 dpa) in all tested materials, while the production rate of defects decreases in the following order: Ni > NiCo > NiFe > NiCoFeCr. The mobility and recombination rate of loops, as directly observed in situ, decrease in the same order. Moreover, electron irradiation shows the growth rate of loop sizes in NiCoFeCr is more than 15 times lower than that in Ni. In summary, our results indicate concentrated solid solution alloys as promising candidate for nuclear materials with enhanced radiation damage tolerance.