Symposium Organizers
Enrique Martinez, Los Alamos National Laboratory
Laurent Capolungo, Georgia Institute of Technology
Daniel Schwen, Idaho National Laboratory
Aurelien Vattre, Commissariat a l'energie atomique
CM2.1: Input to Mesoscale Modeling
Session Chairs
Tuesday PM, April 18, 2017
PCC North, 100 Level, Room 125 B
11:45 AM - *CM2.1.01
Reconciling Ab Initio Modeling of Self-Interstitial Migration with Internal Friction Experiments in Irradiated Zirconium
Emmanuel Clouet 1 , Celine Varvenne 2
1 , CEA Saclay, Gif Sur Yvette France, 2 CINAM, CNRS - Aix-Marseille University, Marseille France
Show AbstractIrradiation in hexagonal close-packed zirconium leads to the creation of point-defects, both vacancies and self-interstitials. These point defects are responsible of the internal friction peaks observed in irradiated zirconium [1]. In particular, the low temperature peaks are induced by the self-interstitial, which can take different configurations. We performed ab initio calculations to characterize all possible configurations of the self interstitial [2] and the different migration barriers between these configurations. The obtained ab initio data are then used to model the internal friction signal in irradiated zirconium. The predicted signal is compared to the experiments, so as to assess the validity of the self-interstitial energy landscape obtained in our ab initio calculations.
[1] R. Pichon, E. Bisogni and P. Moser, Radiation Effects 20, p. 159 (1973).
[2] C. Varvenne, F. Bruneval, M.-C. Marinica and E. Clouet, Phys. Rev. B 88, 134102 (2013).
12:15 PM - CM2.1.02
Point Defects in Materials—Measurement of Elastic Dipoles and Polarisability Effects
Celine Varvenne 1 , Fabien Bruneval 2 , Mihai-Cosmin Marinica 2 , Emmanuel Clouet 2
1 , CINaM - CNRS / Aix-Marseille University, Marseille France, 2 , CEA Saclay, Gif-sur-Yvette France
Show AbstractPoint defects (PDs) in crystalline solids, such as vacancies, self-interstitials, solute atoms or small clusters, play a crucial role in controlling materials properties and their kinetic evolution, particularly through their interactions with other defects, like other PDs, dislocations, or surfaces/interfaces. The dipole moment representation of those defects, within continuum elastic theory, has become a valuable tool to model those long-range interactions that are not reachable by typical ab initio calculations, and is also valid at rather short distances. Successful applications of this PD representation can be quoted, such as finite-size correction schemes, stress-driven diffusion, or sink strength optimization of semi-coherent interfaces [1-3]. Accurate measurements of elastic dipoles are thus needed, with a method working at the small simulation size of ab initio simulations.
Here, we perform a detailed study of different methods to extract elastic dipoles from atomistic simulations: strain derivatives of the energy, adjustment on elastic displacement fields, and computation of Kanzaki forces. Higher order moments and inhomogeneity effects, that contribute to the short-range PD/defect interaction, are also considered. We use empirical potentials – that allow very large simulation sizes – for the purpose of comparing carefully the different methods, considering the vacancy defect and various self-interstitials in hcp zirconium. We then discuss the feasibility of the different methods within ab initio simulations, particularly with respect to their inherent size limitations. To this end, we rely on ab initio calculations performed on the previous systems, plus self-interstitial clusters in bcc iron and vacancy in diamond silicon and fcc aluminium. We finally extend this systematic study to the case of PDs in non-dilute alloys (e.g. high entropy alloys, austenitic stainless steels, etc.), considering Fe-Ni-Cr solid-solution alloys as generic model systems.
[1] C. Varvenne, F. Bruneval, M.-C. Marinica and E. Clouet, Phys. Rev. B 88 (2013).
[2] T. Garnier, V.R. Manga, P. Bellon and D. Trinkle, Phys. Rev. B 90 (2014).
[3] A.Vattré, T. Jourdan, H. Ding, M.-C. Marinica and M.J. Demkowicz, NComms. (2016).
12:30 PM - CM2.1.03
Coarse Grained Molecular Dynamic Simulations of the Interaction a Carbon Nanotube with a Bilayer Membrane
Clarence Matthai 1 , Rangeen Salih 1
1 , Cardiff University, Cardiff, Wales, United Kingdom
Show AbstractIn a coarse grained (CG) MD simulations, small groups of atoms are treated as single particles (beads) and the forces between these particles are derived from the interatomic forces. The effect of this is to severely reduce the number of particles in a simulation, thereby allowing for the consideration of a larger number of atoms. It has also proven to be a valuable tool to probe time and length scales of systems beyond that used in all-atom (AA) simulations. The down side of this is that the inter-particle interactions are less accurate. However, if these coarse grained particles are chosen carefully, such simulations can provide much useful information.
It may also be noted that some applications do not need detailed atomistic information making them very suitable for CGMD simulations. However, using CG methods can also result in an inaccurate description of systems, including membrane systems, so CG simulations are not always recommended. There are different levels of how the coarse grains are constructed. For example, CG systems have been developed using tens or hundreds of atoms per CG bead in some studies of amino acids in biological science. By contrast, for other systems, a single CG bead is used to replace just two or three atoms.
The interaction potential between CG beads are simplified and do not include bending or torsional interactions. Electrostatic interactions are only taken into account if absolutely necessary. The important point is that CG simulations are used to provide a collective description of observed phenomena using only significant interactions. The main feature of the CG approach is that it is faster than all atom simulations. For example, in a simulation of water, 4 water molecules each comprising 3 atoms can be replaced by a single CG water bead. Additionally, the time step in the simulation can be increased by a factor of 10 or 15. So, in such a case, using a coarse grained model would result in a computational speed-up of about 100.
In this paper, the interaction of a carbon nanotube with a lipid bilayer membrane is studied using both coarse grained and atomistic MD in an effort to understand the usefulness of the CGMD method for such simulations. Our preliminary studies of the interaction of a CNT with a lipid bilayer points indicates that such nano-tubes inserted into a membrane could be stable. This means that it could be used as an agent in the delivery of drugs. It would be good if these simulations could be repeated using AA simulations to confirm the validity of these results.
CM2.2: Beyond Atomistic Models I
Session Chairs
Tuesday PM, April 18, 2017
PCC North, 100 Level, Room 125 B
2:30 PM - *CM2.2.01
A Coupled Continuum-Atomistic Framework for 2D Layered Heterostructures
Ellad Tadmor 1
1 , University of Minnesota, Minneapolis, Minnesota, United States
Show AbstractThe synthesis of graphene, a one-atom thick 2D graphitic sheet, was a revolution in materials physics. Since then a host of other 2D materials have been discovered that can be stacked to create layered heterostructures with remarkable properties. Due to the weak van der Waals interaction between layers, the resulting structures can be incommensurate and therefore challenging to model. We describe recent work on developing a hybrid continuum-atomistic computational framework for simulating the mechanical response of 2D heterostructures.
3:00 PM - CM2.2.02
Quantifying the Complex Dynamics of Moving Dislocations from the Atomic to the Microscale
Rigelesaiyin Ji 1 , Shuozhi Xu 2 , David McDowell 2 3 , Youping Chen 4 , Liming Xiong 1
1 Department of Aerospace Engineering, Iowa State University, Ames, Iowa, United States, 2 Woodruff School of Mechanical Engineering, Georgia Institute of Technology, Atlanta, Georgia, United States, 3 School of Materials Science and Engineering, Georgia Institute of Technology, Atlanta, Georgia, United States, 4 Department of Mechanical and Aerospace Engineering, University of Florida, Gainesville, Florida, United States
Show AbstractIn this work, the dynamics of moving dislocations in metallic materials are quantified using a concurrent atomistic-continuum (CAC) methodology. The velocity-dependent dislocation line energy, core stress field, and wavelength of the phonon waves emitted from moving dislocations in metallic materials are measured from the atomic to the microscale. Results show that the higher the dislocation velocity the shorter the wavelength of the phonon waves emitting from moving dislocations. It is also observed that the short-wavelength phonons cannot smoothly pass from the atomic to the coarse-grained (CG) atomistic domain. They are seen to be reflected back to the fully atomistic domain. When the dislocation velocity is high, in CAC without any special treatments, such reflections lead to an inaccurate prediction of dislocation dynamics, such as mobility. As such, a digital filter which can absorb or selectively pass the short-wavelength phonons from the atomic to the CG domain becomes necessary and is implemented into CAC in this work. Our preliminary results demonstrate that such a CAC furnished with the digital filter does accurately predict the mobility of high-speed dislocations, which mainly emit short-wavelength phonons. The CAC equipped with the digital filter is then applied to exam the mechanisms of the interactions between obstacles and high-speed dislocations in heterogeneous metallic materials under intermediated and high strain-rate loadings. The barrier strength of obstacles to dislocations is found to be dislocation-velocity dependent. The limitation, the future development as well as the potential applications of the extended CAC in predicting the dynamic properties of heterogeneous materials will be also discussed.
3:15 PM - *CM2.2.03
Static Strain Aging—A Diffusive Molecular Dynamics Study
Chad Sinclair 1
1 , University of British Columbia, Vancouver, British Columbia, Canada
Show AbstractSegregation Engineering as a way of controlling the structure and properties of engineering alloys has been growing in popularity. Solute segregation to interfaces and defects is driven by atomistic effects but prediction of the time evolution of the composition is difficult as segregation occurs on timescales that far exceed those accessible with standard molecular dynamics simulations. Here, a dynamical version of the variational Gaussian method for binary alloys based on the recently proposed Diffusive Molecular Dynamics approach developed to simulate kinetics of vacancy diffusion in crystalline structures at the atomic level will be presented. In this approach, a phonon-free description of solids is coupled with statistical averaging over various configurations, from which free energies can be efficiently computed. Atomic positions are represented by Gaussian density fields, whose amplitude is interpreted as an occupation probability that indicates the relative local concentration of each atomic species. Interactions arise from a “thermalized” embedded atom (EAM) potential, and the free energy of the alloy is minimized by optimizing atomic positions and vibrational amplitudes coupled with relaxational dynamics of the concentration fields.
This model will then be used to predict the segregation of Mg solute to dislocations in binary Al-Mg alloys. The stress required to pull these dislocations away from the solute atmosphere will then be calculated and discussed in relation to the phenomenon of static strain aging.
4:15 PM - CM2.2.04
First-Principles and Grand Canonical Monte Carlo Simulations of Pseudocapacitive Response of MnO2 Electrodes
Yasuaki Okada 1 2 , Nathan Keilbart 2 , Ismaila Dabo 2 , Shin'ichi Higai 1 , Kosuke Shiratsuyu 1
1 , Murata Manufacturing Co., Ltd., Nagaokakyo-shi, Kyoto, Japan, 2 Materials Research Institute, Pennsylvania State University, University Park, Pennsylvania, United States
Show AbstractElectric double layer capacitors exhibit higher power density and longer cycle life than batteries; however, there is a need to increase their energy density to broaden their technological use. Pseudocapacitors are energy storage devices that have a much larger energy density than electric double layer capacitors due to their ability to store charges at the electrode–electrolyte interface through both charge accumulation within the electric double layer and electron transfer to adsorbed ionic species. Among existing electrode materials for use in pseudocapacitors, ruthenia (RuO2) is known to exhibit an exceptionally large pseudocapacitance. But ruthenium is a precious metal, which limits the prospects of mass production. Therefore, considerable efforts have been devoted to developing cheaper pseudocapacitive electrode materials whose charge storage ability is comparable to that of RuO2. At the molecular level, charge storage in RuO2 pseudocapacitors is thought to involve the double insertion of electrons and protons into the electrode. However, the detailed mechanisms underlying pseudocapacitive charge storage under applied voltage are still not understood. In particular, the inclusion of configurational entropy contributions poses a major challenge to first-principles simulations due to their high computational cost. In this research, we have carried out grand canonical simulations using an accurate extrapolation of first-principles free energies taking into account solvent effects via efficient quantum-continuum models and configurational entropy contributions via finite-temperature Monte Carlo sampling in an effort to clarify the charge storage mechanisms underlying the operation of pseudocapacitive materials. Importantly, we have elucidated the sequence of adsorption reactions that take place on beta-MnO2 stoichiometric (110) surfaces in water, as well as the sequence of adsorption/intercalation reactions for delta-MnO2 layers and lithium ions upon raising the applied voltage. We have furthermore calculated the voltammetric peaks corresponding to transitions in the adsorption/intercalation configurations, and we have predicted realistic charge-voltage curves taking into account configurational entropy contributions in close agreement with experimental data. This comprehensive method provides a powerful approach to predict the intrinsic voltage-dependent response of pseudocapacitive materials and ultimately guide their chemical and structural optimization.
4:30 PM - *CM2.2.05
Predicting Materials Strength in BCC Alloys Using Parameter-Less Mesoscale Approaches
Jaime Marian 1 , David Cereceda 2 , Dierk Raabe 3 , Franz Roters 3 , Yue Zhao 1 , Martin Diehl 3
1 , University of California-Los Angeles, Los Angeles, California, United States, 2 , Johns-Hopkins University, Baltimore, Maryland, United States, 3 , Max Planck Institute for Iron Research, Düsseldorf Germany
Show AbstractThe yield and flow stresses of body-centered cubic (BCC) metals and alloys display a strong dependence with temperature and loading orientation. We will present a computational approach based on crystal plasticty calculations and kinetic Monte Carlo simulations capable of predicting the dependence of the yield strength with temperature and orientation entirely from first-principles, using atomistic calculations to calculate all model parameters and without fitting to any experimental data. Key physics input in the form of thermally activated dislocation mobilities, non-Schmid effects on the resolved shear stress, and solute effects are all included in the model by construction, yielding excellent agreement with experiments, and predicting soft and hard regions of the stereographic triangle. Although the methodology is general to all BCC systems, we apply the model to W-Re alloys due to their high technologiucal importance.
5:00 PM - CM2.2.07
Order Parameter Aided Efficient Phase Space Exploration
Amit Samanta 1 , Eric Schwegler 1 , Jonathan Belof 1 , Alexander Chernov 1
1 , Lawrence Livermore National Laboratory, Livermore, California, United States
Show AbstractPhysical processes in nature exhibit disparate time-scales, for example time scales associated with processes like phase transitions, various manifestations of creep, sintering of particles etc. are often much higher than time the system spends in the metastable states. The transition times associated with such events are also orders of magnitude higher than time-scales associated with vibration of atoms. Thus, an atomistic simulation of such transition events is a challenging task. Consequently, efficient exploration of configuration space and identification of metastable structures in condensed phase systems is challenging. In this talk I will illustrate how we can define a set of coarse-grained variables or order parameters and use these to systematically and efficiently steer a system containing thousands or millions of atoms over different parts of the configuration. This order parameter aided sampling can be used to identify metastable states, transition pathways and understand the mechanistic details of complex transition processes. I will illustrate how this sampling scheme can be used to study phase transition pathways and phase boundaries in prototypical materials, like SiO2, Ge.
Symposium Organizers
Enrique Martinez, Los Alamos National Laboratory
Laurent Capolungo, Georgia Institute of Technology
Daniel Schwen, Idaho National Laboratory
Aurelien Vattre, Commissariat a l'energie atomique
CM2.3: Beyond Atomistic Models II
Session Chairs
Wednesday AM, April 19, 2017
PCC North, 100 Level, Room 125 B
10:00 AM - CM2.3.02
A New Implementation of the Concurrent Atomistic-Continuum Method with Quasistatic Algorithms and Mesh Refinement Schemes
Shuozhi Xu 1 , Yongchao Liu 1 , Youping Chen 2 , David McDowell 1
1 , Georgia Institute of Technology, Atlanta, Georgia, United States, 2 , University of Florida, Gainesville, Florida, United States
Show AbstractIn recent years, the concurrent atomistic-continuum (CAC) [1] approach has been widely employed as an effective tool for coarse-grained modeling of dislocation-mediated metal plasticity, including screw dislocation cross-slip, dislocation/void interactions, dislocations bowing out from obstacles, dislocation multiplication from Frank-Read sources, sequential slip transfer of dislocations across grain boundaries, and transonic dislocation dynamics. Here, the CAC method is augmented by implementing quasistatic algorithms and mesh refinement schemes [2,3]. Other new features of the code include: (i) common atomistic techniques (Newton’s third law, Verlet list and link cell methods for short range neighbor search) have been employed, (ii) periodic boundary conditions become available by filling in extra atoms at the jagged interstices, (iii) integration points in individual elements are shared among multiple processors to minimize the amount of data communication, (iv) simulation results are visualized using ParaView, a free software, such that CAC results are accessible to a larger community, and (v) besides the force and stress, energy on each node is calculated, by virtue of a novel approach to implement the state-of-the-art embedded-atom method potential in the coarse-grained domain. The new CAC code enables energy minimizing the dislocated ensemble that evolves away from equilibrium during a sequence of nonequilibrium configurations, as well as modelling brittle-to-ductile transition in dynamic fracture.
[1] L. Xiong, G.J. Tucker, D.L. McDowell, Y. Chen, Coarse-grained atomistic simulation of dislocations, J. Mech. Phys. Solids, 59, 160 (2011)
[2] S. Xu, R. Che, L. Xiong, Y. Chen, D.L. McDowell, A quasistatic implementation of the concurrent atomistic-continuum method for FCC crystals, Int. J. Plast., 72, 91 (2015)
[3] S. Xu, L. Xiong, Q. Deng, D.L. McDowell. Mesh refinement schemes for the concurrent atomistic-continuum method, Int. J. Solids Struct., 90, 144 (2016)
10:00 AM - CM2.3
CM2.3.01 Youpin Chen moved to CM2.9.04 Thursday 1130am
Show Abstract10:15 AM - CM2.3.03
Finite-Difference Time-Domain and Monte-Carlo Ray Tracing Hybrid Modeling of Optical Devices and Structures
Mark Portnoi 1 , Christian Sol 1 , Clemens Tummeltshammer 1 , Ioannis Papakonstantinou 1
1 Department of Electrical and Electronic Engineering, University College London (UCL), London United Kingdom
Show AbstractComputerised modelling of structures and materials has become a vital tool in the development and optimisation of materials and devices, saving time and money incurred from prototype manufacture.
Finite-difference time-domain modelling (FDTD) method is used extensively for solving sub-wavelength scale electromagnetic scattering problems, however as modelling dimensions are extended beyond a few μm, demand on computational resources become vast, and the technique is no longer viable. On a macroscopic scale, Monte-Carlo ray tracing, based on classical ray optics, is a robust technique for device simulation. While Monte-Carlo methods are appropriate for modeling macroscale optical systems, the technique can not, on its own, deal with structures whose dimensions approach the wavelength of light.
Combined, however, these techniques can be a powerful instrument for the simulation of complex structures of a large scale and reasonable time. In this work, we present such a FDTD / Monte-Carlo hybrid model. We explore two different scenarios often met in electromagnetic problems; i) nanoparticles embedded in a bulk structure, and ii) nanostructures at the interface between two media.
Differential scattering, scattering, and absorption cross sections are calculated using FDTD techniques and are converted to probability distributions for individual events in the ray tracing model. Polarisation and wavelength are considered, resulting in a multitude of possible yields; transmission, reflection, absorption, spatial light distributions, photoluminescent conversions (both up and down conversions), and energy transfer (such as FRET).
Within our group the platform is used to simulate; nanocomposite films, luminescent solar concentrators, structured surfaces, plasmon and phase change materials. The flexibility of the program allows for easy extension to many diverse applications. At the conference we will present simulation examples of both surface structures and nanocomposite films.
10:30 AM - CM2.3.04
Phase-Field Crystal Model for Ordered Crystals
Eli Alster 1 , Ken Elder 3 , Jeffrey Hoyt 2 , Peter Voorhees 1
1 , Northwestern University, Evanston, Illinois, United States, 3 , Oakland University, Rochester, Michigan, United States, 2 , McMaster University, Hamilton, Ontario, Canada
Show AbstractThe phase-field crystal (PFC) model is a phase-field variant that promises a method for simulating crystalline materials on diffusional time scales but with atomic resolution. However, PFC models are currently limited to crystals with simple crystallographic symmetries. We describe a general PFC model suitable for materials with complex sublattice ordering, such as monolayer BN and L10 CuAu. As a test case, a generic B2 compound is investigated. This model of B2 compound can be manipulated to produce either first-order or second-order order-disorder phase transitions, and it also reproduces classical results for antiphase boundaries. Dynamical simulations of ordering across small-angle grain boundaries illustrate the utility of this model by predicting that dislocation cores pin the evolution of antiphase boundaries. Extensions of this model to ordered 2D materials such as BN and MoS2 will be discussed.
10:45 AM - CM2.3.05
Coupling Deterministic and Stochastic Simulations—An Application to Cluster Dynamics in Materials
Pierre Terrier 1 , Manuel Athenes 2 , Thomas Jourdan 2 , Gilles Adjanor 3 , Gabriel Stoltz 1
1 , Cermics, Ecole des Ponts, Champs-sur-Marne France, 2 SRMP, CEA Saclay, Gif-sur-Yvette France, 3 MMC, EDF R&D, Moret-sur-Loing France
Show AbstractModeling the aging of irradiated materials requires predicting the microstructural evolution from the nucleation of unstable defect clusters of small sizes to the growth and coarsening of larger defect clusters. Cluster dynamics method has been successfully used to study such a phenomenon. It consists in solving a set of rate equations (i.e. a set of ODEs) describing the evolution of the concentration of clusters of various sizes. However, it becomes computationally prohibitive when large clusters appear. In order to reduce the numerical complexity of the model, we develop a versatile coupling approach between the ODEs and different stochastic approaches. In particular, we highlight two stochastic approaches. The first one is a jump process that exactly describes the dynamics. The second one is based on a limiting model, in the form of a Fokker-Planck equation, obtained in the asymptotic regime of large size clusters. We propose a stochastic approach to solve this equation. The coupling method then allows to simultaneously evolve the set of rate equations (for small size clusters) and the PDE part. The accuracy of this hybrid deterministic/stochastic coupling algorithm is studied on a simple case [1]. We also report results for complex materials with different types of defects.
[1] Terrier et al., arxiv:1610.02949 (2016)
CM2.4: Advance Phase Field Models
Session Chairs
Wednesday PM, April 19, 2017
PCC North, 100 Level, Room 125 B
11:30 AM - *CM2.4.01
On the Incorporation of Environmental Effects in Multi-Scale Modelling Approaches
Esteban Busso 1
1 , ONERA, Palaiseau France
Show AbstractThis work deals with the study of environmental effects on the behaviour of high temperature metallic alloys. Attention is particularly focused on the interaction between oxidation, time-dependent behaviour and microstructure and the local conditions responsible for the formation of surface and grain boundary micro-cracks. Methodologies based upon fundamental physical processes are presented for understanding and predicting oxide growth and the nucleation of associated damage. First, the surface oxidation of Ni-base superalloys is described using a coupled diffusion-single crystal plasticity approach. When superalloys operate at high temperatures, the microstructure is not stable: the γ' precipitates coarsen rapidly and, in the vicinity of free surfaces, there is a depletion of the g’ precipitates due to surface oxidation. Both mechanisms lead to a local softening of the material which in turns reduces the local stresses. Here, the local material behaviour is made to depend explicitly on the characteristics of the γ’-reinforcing phase which are, in turn, affected by the local concentrations of Ni, Al, and O2. The degradation of the local notch root region fields caused by the surface oxidation of a Ni-base superalloy notched compact tension (CT) specimen is discussed. Finite element analysis of the CT specimen revealed how environmental effects can severely reduce the time to crack initiation due to the oxidation-induced material softening.
The second part of this work deals with the growth of oxides along easy diffusion paths such as certain grain boundaries and localised deformation/slip bands intersecting free surfaces. Here, simultaneous surface and grain boundary oxidation phenomena in polycrystalline Fe-Cr-Ni austenitic steels are described through a coupled diffusion-crystal plasticity-phase field formulation. Oxidation is considered at the grain level where the effects of plasticity and grain boundaries on oxide growth are accounted for. Full field finite element simulations of a plastically deformed and oxidising bi-crystal are performed to describe the preferential oxide growth. The resulting preferential oxidation of the grain boundary is shown to depend on the wetting properties of the oxide interface with the grain boundary, the grain boundary energy and the relative grain boundary and bulk diffusivities.
12:00 PM - CM2.4.02
Phase Field Modelling of θ' Precipitation during Aging of Al-4wt.%Cu Alloys—A Multiscale Approach
Hong Liu 2 , Gustavo Esteban-Manzanares 2 , Barbara Bellon 2 , Ilchat Sabirov 2 , Javier Llorca 1
2 , IMDEA Materials Institute, Getafe, Madrid, Spain, 1 , IMDEA Materials Institute & Technical University of Madrid, Getafe, Madrid Spain
Show AbstractThe high strength of Al-Cu alloys is due to the presence of θ' precipitates that hinder dislocation slip. It is well established from the experimental viewpoint that the strength of these alloys is controlled by the size, shape, orientation and spatial distribution of these precipitates but there are not multiscale simulations tools that can predict the size and shape of these precipitates during high temperature aging. In this investigation, the evolution and equilibrium morphology of the θ' precipitates in 3 dimensions during high temperature aging is studied by means of a multiscale approach. The lattice parameters and elastic constants of θ' precipitates and of the α-Al matrix were calculated using first principles density functional theory, whereas the interfacial energy between θ' phase and α-Al matrix was determined by means of molecular dynamics. The elastic strain energy is calculated based on Khachaturyan’s microelasticity theory, and the nucleation of θ' precipitates is simulated by random noise. The effect of elastic strain energy and interfacial energy anisotropy on the equilibrium shape of θ' precipitates are investigated quantitatively. The equilibrium shape and the evolution of θ' precipitates with and without including the presence of dislocations was studied using the phase field method. The simulations indicate the plate-like shape of the θ' precipitates comes about from the competition of the elastic strain energy and the interfacial energy. Moreover, the high aspect ratio of θ' precipitates is induced by the shear strain and interfacial energy anisotropy. It is shown that the stress field of pre-existing dislocation may result in a series of paralleled θ' precipitates forming along the dislocation line. Finally, the results of the multiscale simulations are compared with experimental data on Al-4 wt. % Cu alloys that were aged at 190C for up to 30 hours.
12:15 PM - CM2.4.03
Modeling Fission Gas Bubble Evolution in Nuclear Fuels with the Phase-Field Method
Larry Aagesen 1 , Karim Ahmed 1 , Daniel Schwen 1 , Yongfeng Zhang 1
1 Fuels Modeling and Simulation, Idaho National Laboratory, Idaho Falls, Idaho, United States
Show AbstractDuring operation of nuclear reactors, gaseous by-products of fission events accumulate within the fuel elements. These fission gases precipitate to form bubbles when concentration becomes sufficiently high. The accumulation of fission gases and formation of bubbles causes nuclear fuels to swell, potentially leading to mechanical interaction between the fuel elements and cladding. Metallic fuels typically show large amounts of swelling. Additionally, a portion of the accumulated gas within the bubbles is eventually released to the space between the fuel and cladding, increasing pressure and potentially affecting thermal conductivity. Thus, understanding the dynamics of fission gas bubble evolution is of vital importance in predicting the performance of nuclear fuels over core lifetime. We have implemented a phase-field model of fission gas bubble evolution within the MARMOT code. The model accounts for the bulk free energies of the gas and bubble phases, interfacial energy between phases, and stresses caused by the gas pressure exerted by the bubbles on the surrounding fuel matrix. The model is applied to metallic (U3Si2) and oxide (UO2) fuels. A strategy for parameterizing an engineering-scale model in the BISON fuel performance code based on lower length scale simulation results is discussed.
12:30 PM - CM2.4.04
Coupling Radiation Damage to Phase Field Microstructure Evolution and Thermal Transport
Daniel Schwen 1 , Sebastian Schunert 1 , Larry Aagesen 1
1 , Idaho National Laboratory, Idaho Falls, Idaho, United States
Show AbstractModeling reactor materials in operation is a challenging multi physics and multi scale problem. The accumulated radiation damage, thermal and mechanical loading, constant driving of the system by irradiation, and chemical changes experienced during the operational lifetime have a strong influences on the material properties both directly and indirectly by impacting the microstructural evolution. In fuels this can cause strong feedback through the modified the neutron transport characteristics of the material requiring careful thermal analysis on a microstructural level. In structural materials one consequence can be the introduction of non-equilibrium steady states and strong modifications of the kinetics.
Transient radiation conditions, as exhibited under reactor accident conditions, require a constant reevaluation of the radiation environment in time. Microstructural evolution requires a spatially resolved radiation damage treatment. To deal with both the variation of the irraddiation time and in space we have developed a coupling between neutronics calculations and microstructure evolution and heat transport through concurrent binary collision Monte Carlo (BCMC) simulations. The neutron transport simulation informs the primary recoil distribution using the neutron reaction data to obtain a spatially resolved recoil energy and mass distribution. The BCMC simulation translates this into defect production, ballistic mixing, and thermal energy deposition, by performing full cascade simulations on the simulation microstructure.
We will demonstrate this novel coupling capability on multiphase systems driven by irradiation. These systems can experience transitions between patterned, phase separated, and mixed states as a function of the radiation intensity. The thermal analysis capability will be demonstrated through simulations of dispersed micro structured nuclear fuel, a system of particular interest in INL’s transient test reactor (TREAT).
12:45 PM - CM2.4.05
Microstructure Evolution of Powder Materials during Solid State Sintering—A Phase Field Study
Sudipta Biswas 2 , Daniel Schwen 1 , Vikas Tomar 2
2 School of Aeronautics and Astronautics, Purdue University, West Lafayette, Indiana, United States, 1 Fuel Modeling and Simulation, Idaho National Laboratory, Idaho Falls, Idaho, United States
Show AbstractSintering is an advanced manufacturing technique for developing new materials in a cost effective manner. During the process small powder particles consolidate into a solid polycrystalline structure by reducing the void fraction and over all surface area. Densification during sintering is governed by several physical processes such as diffusion, grain boundary migration, plastic deformation etc. In addition, during the sintering process particles can freely translate and rotate as rigid body into adjacent void places. Initial contact behavior between particles plays an important role in facilitating the rigid body motion of particles. In order to understand the densification mechanisms and microstructural evolution during sintering consideration of rigid body motion is of immense importance here. Phase field modeling is a vastly used computational method to observe microstructural changes. However, applying phase field modeling to simulate sintering process has been challenging due to unavailability of appropriate approach for including rigid body motion in phase field. Rigid body motion introduces additional integral terms to the traditional phase field equations that requires enhancement of the conventional computational method. Current work focuses on capturing the microstructural changes during the sintering process using phase field modeling. Finite element based MOOSE framework has been used for implementing the model with special emphasis on enhancing the numerical methods for solving rigid body motion of particles along with phase field. Introduced additional capability not only helps in capturing the sintering process effectively but also improves the numerical convergence of the problem reducing the over all computational cost. It is observed that, as the applied load exceeds the contact force between particles, powder compaction is primarily governed by surface and grain boundary diffusion. Additionally, grain boundary migration leads to grain growth. Due to the possibility of particle rotation the grain orientation changes during the process. Current work also proposes the model for tracking grain orientation during the process. Final equilibrium configuration of particles and grain morphology depends on initial stacking of the powder particles, process conditions, loading rate and total sintering time.
CM2.5: New Developments in Dislocation Dynamics
Session Chairs
Wednesday PM, April 19, 2017
PCC North, 100 Level, Room 125 B
2:30 PM - *CM2.5.01
ResiduAl Stress Predictions in Nanocrystalline Thin Films
Marisol Koslowski 1
1 , Purdue University, West Lafayette, Indiana, United States
Show AbstractThere has been significant progress in the understanding of the mechanisms that drive local residual stresses in thin metal films. From the size-dependence of film and grains to the effects of grain size and stress field distributions to the stress relaxation mechanisms that lead to failure of thin films.
The response of thin metal films to inhomogeneous residual stresses includes yielding, diffusional and dislocation-mediated creep, grain boundary sliding, cracking, delamination, surface roughening, recrystallization, grain growth, stress induced void growth, whisker and hillock formation. On the other hand, the particular combination of processes that dominates residual stress build up depend on thickness, crystallographic texture, adhesion to and orientation relationships with the substrate, temperature and loading history.
Dislocation dynamics simulations are performed in nanocrystalline copper thin films subjected to cyclic loading. During unloading different plastic recovery processes that are related to formation of dislocation structures. Furthermore, grain size inhomogeneity increases the amount of plastic strain recovered after the thin film is unloaded. It is also observed that the dislocations that are incorporated into grain boundaries during loading remain locked and do not show reverse motion during unloading. The results are used to quantify the distribution of residual stresses incorporating the effect of microstructure and loading history and provide metrics to relate residual stress with texture and grain size distribution that may be incorporated into continuum level simulations.
3:00 PM - *CM2.5.02
Challenges in Using Spectral Methods for Computing Stress Fields from Dislocations
Richard LeSar 2 , John Graham 2 , Anthony Rollett 1
2 Materials Science and Engineering, Iowa State University, Ames, Iowa, United States, 1 Materials Science and Engineering, Carnegie Mellon University, Pittsburgh, Pennsylvania, United States
Show AbstractWe have developed a Fast-Fourier Transform (FFT) approach to discrete dislocation dynamics (DDD) simulations to overcome the large computational costs in DDD that comes from the calculation of long-range interaction forces between dislocation segments and to easily incorporate the effects of anisotropic elasticity. Our long-range goal is to implement the FFT-DDD method within an FFT-based polycrystal plasticity method. In this talk, we will present the basic formalism of the FFT-DDD method, in which dislocations are represented by the plastic distortion tensor and the stresses from the dislocations are calculated on an FFT grid. The stresses acting on individual dislocations can then be calculated, from which the dynamics of their motion can be determined. We will discuss the challenges inherent in this approach that arise from the discontinuous nature of the plastic distortion tensor and consequent oscillations in the Fourier transforms. We will show how we have overcome some of these issues using various methods taken from the computer graphics community.
4:30 PM - CM2.5.03
Modelling Plastic Deformation in Micro- and Nano- Samples Using the Discrete-Continuum Model
Riccardo Gatti 1
1 , LEM UMR 104 CNRS-ONERA, Chatillon France
Show AbstractPlastic deformation of crystalline materials is the result of the collective movement of dislocations, in response of their mutual interactions, external applied loading and interactions with boundaries such as free surfaces, interfaces or grain boundaries. The dislocation microstructures emerging from such dynamics are intrinsically heterogeneous and the way they affect the mechanical properties is a puzzling problem. A reliable tool to model crystal plasticity in micro- and nano-materials is the Discrete-Continuum Model (DCM).
The DCM is based on a coupling between 3D Dislocation Dynamics (DD) simulations and Finite Element (FE) method. In particular, the DD code is in charge of the dislocation microstructure evolution while displacement field and boundary conditions (including surfaces and interfaces) are handled by the FE code. The DCM has been significantly improved during the last years [1]. It is now possible to handle problems with very large number of dislocations when accounting for the influence of complex boundary conditions.
Here, the DCM is used to investigate plastic deformation of Ni and Cu micro-samples and SiGe nanostructures. Those two studies emphasize the DCM strong ability to investigate plasticity in micro- and nano-objects. In addition, the attractive capability of running DD simulations making use anisotropic elasticity is highlighted.
[1] O. Jamond, R. Gatti, A. Roos, B.Devincre. International Journal of Plasticity, 80 19-37 (2016)
4:45 PM - CM2.5.04
A FFT-Based Formulation for Efficient Mechanical Fields Computation in Isotropic and Anisotropic Periodic Discrete Dislocation Dynamic
Laurent Caplungo 1
1 , Los Alamos National Laboratory, Los Alamos, New Mexico, United States
Show AbstractIn this study we propose a novel full-field approach based on the fast Fourier transform (FFT) technique to compute mechanical fields in periodic discrete dislocation dynamics (DDD) simulations for anisotropic materials: the DDD-FFT approach. By coupling the FFT-based approach to the discrete continuous model, the present approach benefits from the high computational efficiency of the FFT algorithm, while allowing for a discrete representation of dislocation lines. It is demonstrated that the computational time associated with the new DDD-FFT approach is significantly lower than that of current DDD approaches when large number of dislocation segments are involved for isotropic and anisotropic elasticity, respectively. Furthermore, for fine Fourier grids, the treatment of anisotropic elasticity comes at a similar computational cost to that of isotropic simulation. Thus, the proposed approach paves the way towards achieving scale transition from DDD to mesoscale plasticity, especially due to the method's ability to incorporate inhomogeneous elasticity. Demonstration of such capability is made by consdiering the case of dislocation precipitate interactions in bcc metals.
CM2.6: Discrete Dislocation Dynamics I
Session Chairs
Wednesday PM, April 19, 2017
PCC North, 100 Level, Room 125 B
5:00 PM - *CM2.6.01
Precipitates Strengthening Using Dislocation Dynamics
Sylvie Aubry 1
1 , Lawrence Livermore National Laboratory, Livermore, California, United States
Show AbstractPredicting the influence of precipitate characteristics on the
strength of selected aluminum-lithium alloys for a variety of heat
treatment conditions has important applications for the aircraft
industry. Currently, aircraft hybrid fan blades are made of
titanium. Replacing these blades by aluminum-lithium alloys has been
shown to lead to great savings in fuel consumption. The ability to
model and predict the mechanical response of these alloys as a
function of material processing parameters allows optimization of
current forgings and future designs.
As part of a multiscale approach the dislocation dynamics method has
been used to predict the strength of materials. An extension of the
dislocation dynamics method is presented. It takes precipitates into
account. Interactions of dislocations defect with ellipsoidal
precipitates is modeled. Large scale simulations of theta prime and T
precipitates are presented. The stress/strain response as a function
of precipitate characteristics are shown and explained.
Lawrence Livermore National Laboratory is operated by Lawrence
Livermore National Security, LLC, for the U.S. Department of Energy,
National Nuclear Security Administration under Contract
DE-AC52-07NA27344.
5:30 PM - *CM2.6.02
OptiDis—Hybrid MPI/OpenMP Parallelism for Large Scale Dislocation Dynamics Simulations
Laurent Dupuy 1 , Arnaud Durocher 1 2 , Pierre Blanchard 2 , Olivier Coulaud 2
1 , CEA Saclay, Gif sur Yvette France, 2 HiePACS, INRIA, Bordeaux France
Show AbstractDislocation Dynamics (DD) is a powerful tool to understand crystalline materials behavior at a mesoscopic level for multiscale modeling. Massively parallel codes based on message passing paradigm (MPI) have been proposed to simulate several millions of dislocation segments over thousands of cores. The use of hybrid parallelism (MPI + OpenMP) is necessary to take advantage of modern supercomputers composed of multicore computation nodes. OptiDis aims to implement hybrid parallelism for DD simulations.
Computing elastic interaction forces between dislocation segments is the most resource consuming step of the simulation. Hierarchical algorithms such as the Fast Multipole Method (FMM) can be used to accelerate and parallelize this task. Using the ScalFMM library, specific far-field approximations were developed to reduce the complexity of the computation, and hybrid parallelism has been implemented to distribute the work among computation nodes.
New data structures and algorithms using hybrid parallelism have been developed to be more efficient with current hierarchical architectures. FMM based hierarchical domain decomposition allows the distribution of the dislocation segments among the computation nodes and cache-conscious data structures ensure fast intra-node multi-threaded computation.
CM2.7: Poster Session
Session Chairs
Thursday AM, April 20, 2017
Sheraton, Third Level, Phoenix Ballroom
9:00 PM - CM2.7.01
A Phase-Field Model for the Development of Surface Morphology during Wet Chemical Etching
Jin-Ru Miao 1 , Kun-Dar Li 1
1 Department of Materials Science, National University of Tainan, Tainan Taiwan
Show AbstractBy virtue of the unique advantages, such as the low cost, easy-to-control and flexible procedures, a wet chemical etching process has been widely used in many fields, e.g. microelectromechanical systems and semiconductor industry. In order to manufacture a specific morphology suitable for the functional devices, all the processing parameters, such as the species, concentration and temperature of the etchant, need to be well controlled. In this study, a chemical etching model based on phase field approach was established to simulate the evolution of surface morphology during the etching process. Following the concept of an atom-vacancy complex system, the variable of vacancy content in the matrix was adopted to characterize the variation of the structure during the chemical etching process. The mechanisms of surface profile formation were connected to the diffusion and the chemical reaction of the etchant and substrate atoms. Different etching rate and crystalline system were also applied to investigate their influence on the formation of morphological structures during processing. At a low etching rate, the surface morphology was inclined to be smooth and a slow erosion movement was revealed. While a raised etching rate was considered, the surface became roughened and the characteristic morphology would be produced. Furthermore, under different crystalline systems various specific surface morphologies could be formed, including nano-pyramids, nano-ripples and nano-pillars. With the results of the simulations, a better understanding in the formation mechanism of surface morphology would be helpful to improve the technique of wet chemical etching for advanced applications.
9:00 PM - CM2.7.02
Computer Simulation of Microstructural Evolution during Oblique Angle Deposition
Sheng-Jie Hong 1 , Yu-Wei Dong 1 , Kun-Dar Li 1
1 Department of Materials Science, National University of Tainan, Tainan Taiwan
Show AbstractOver the past decades, featured morphologies of thin films had attracted great interest in the engineering community for the purpose of specific applications in advanced technologies. Among various physical or chemical synthesis methods to produce these unique microstructures, one of the most successful approaches was the physical vapor deposition (PVD), utilizing an oblique angle deposition (OAD) technique. Usually, in the process of oblique angle deposition the substrate was placed in a slanted angle with respect to the target source. To meet the needs of practical applications, a complete understanding of formation mechanism during thin-film deposition is essential, especially for the attempt of a better control on its microstructures and morphologies. In this study, a phase field model was adopted to integrate the diffusion equation and deposition processes, including the shadowing effect, to illustrate the growth of thin films during an oblique angle deposition. From the numerical simulation results, it distinctly demonstrated that the shadowing effect and atomic diffusion were both involved in the specific mechanisms of morphological formation. In general, the atomic diffusion led to a reduction in defects of deposits and a smoothening of film surface, while the shadowing caused a preferential deposition on mounds, leading to the formation of a rough, porous, columnar microstructure. Among the multiple process parameters in an oblique angle deposition, the incident angle, temperature, and deposition rate were also played critical factors for the determination of the surface morphologies. By performing a series of numerical calculations with various parameters, the influences of the incident angle, temperature and the deposition rate on the microstructural formation of thin films were exhibited systematically. Furthermore, the quantitative information for the deposited films, such as the growth rate and the ratio of surface area to volume, was also analyzed. This numerical simulation provided the valuable insight into the underlying mechanisms of morphological formation during oblique angle depositions.
9:00 PM - CM2.7.03
3D Micromechanical Modeling of Dual Phase Steels Using the Representative Volume Element Method and Response Surface Methodology: Parametric Study
Tarek Belgasam 1 , Hussein Zbib 1
1 School of Mechanical and Materials Engineering, Washington State University, Pullman, Washington, United States
Show AbstractDual Phase (DP) Steels have been receiving worldwide attention as playing an important role in automotive industries due to their light density for weight saving and also for fuel saving and decreasing emissions. Recently, studies on developing DP steels showed that the combination of strength/ductility could be significantly improved after changing the volume fraction and grain size of phases in the microstructure depending on microstructure properties. Consequently, steel manufacturers who develop DP steels are interested in predicting the microstructure properties relationship for optimization of microstructural design. In this work, a microstructure-based approach by means of representative volume elements (RVEs) is built to study the flow behaviour of DP steels using virtual tension tests of RVE to achieve the desired mechanical properties. Microstructures with different martensite and ferrite grain size, martensite fractions, Carbon content in DP, and morphologies are studied in 3D RVE approaches as microstructure parameters. The effect of these microstructure parameters on the combination of strength/ductility of DP steels has been investigated by using Response Surface Methodology (RSM). Response surface contours were constructed for determining the optimum conditions for a required combination of strength/ductility. The verification experiment is carried out to check the validity of the developed model that predicted combination of strength/ductility.
Symposium Organizers
Enrique Martinez, Los Alamos National Laboratory
Laurent Capolungo, Georgia Institute of Technology
Daniel Schwen, Idaho National Laboratory
Aurelien Vattre, Commissariat a l'energie atomique
CM2.8: Discrete Dislocation Dynamics II
Session Chairs
Thursday AM, April 20, 2017
PCC North, 100 Level, Room 125 B
9:30 AM - *CM2.8.01
An Anisotropic Non-Singular Theory of Dislocations
Giacomo Po 1 , Markus Lazar 2 , Jaime Marian 3 , Nasr Ghoniem 3
1 Mechanical and Aerospace Engineering Department, University of California, Los Angeles, Los Angeles, California, United States, 2 Department of Physics, Darmstadt University of Technology, Darmstadt Germany, 3 Materials Science and Engineering Department, University of California Los Angeles, Los Angeles, California, United States
Show AbstractThe singular nature of the elastic fields produced by dislocations presents conceptual challenges and computational difficulties in the implementation of discrete dislocation-based models of plasticity. In this work we consider theoretical and numerical aspects of the non-singular theory of discrete dislocation loops in a particular version of Mindlin’s anisotropic gradient elasticity with up to six independent gradient parameters. The framework models anisotropic materials where there are two sources of anisotropy, namely the bulk material anisotropy and a weak non-local anisotropy relevant at the nano-scale. The Green tensor of this framework, which we derive as part of the work, is non-singular and it rapidly converges to its classical counterpart a few characteristic lengths away from the origin. Therefore, the new Green tensor can be used as a physical regularization of the classical Green tensor. The Green tensor is the basis for deriving a non-singular eigenstrain theory of defects in anisotropic materials, where the non-singular theory of dislocations is obtained as a special case. The fundamental equations of curved dislocation loops in three dimensions are given as non-singular line integrals suitable for numerical implementation using fast one-dimensional quadrature. These include expressions for the interaction energy between two dislocation loops and the line integral form of the generalized solid angle associated with dislocations having a spread core. The six characteristic length scale parameters of the framework are obtained from the components of the rank-six tensor of strain gradient coefficients of Mindlin’s theory. In turn, the components of such tensor are obtained from atomistic calculations. In particular, we show that the rank-six tensor of strain gradient coefficients has an explicit local representation in terms of the derivatives of atomistic potentials. By virtue of this explicit representation, the link between atomistic and the simplified theory of gradient elasticity is established, and a non-singular and parameter-free theory of dislocations in anisotropic materials is obtained. Several applications of the theory are presented.
10:00 AM - *CM2.8.02
Advanced Time Integration Algorithms for Dislocation Dynamics Simulations of Work Hardening
Rayn Sills 2 , Amin Aghaei 1 , Nicolas Bertin 1 , Wei Cai 1
2 , Sandia National Laboratories, Livermore, California, United States, 1 , Stanford University, Stanford, California, United States
Show AbstractEfficient time integration is a necessity for dislocation dynamics simulations of work hardening to achieve experimentally relevant strains. An efficient time integration scheme using a high order explicit method with time step subcycling and a newly-developed collision detection algorithm are evaluated. First, time integrator performance is examined for an annihilating Frank–Read source, showing the effects of dislocation line collision. The integrator with subcycling is found to significantly out-perform other integration schemes. The performance of the time integration and collision detection algorithms is then tested in a work hardening simulation. The new algorithms show a 100-fold speed-up relative to traditional schemes. Subcycling is shown to improve efficiency significantly while maintaining an accurate solution, and the new collision algorithm allows an arbitrarily large time step size without missing collisions. We also discuss the potential of applying asynchronous integrators, in which every pairwise interaction has a different time step, for dislocation dynamics simulation.
This work was supported by Sandia National Laboratories (R.B.S.) and by the U.S. Department of Energy, Office of Basic Energy Sciences, Division of Materials Sciences and Engineering under Award No. DE-SC0010412 (W.C.). Sandia National Laboratories is a multi-program laboratory managed and operated by Sandia Corporation, a wholly owned subsidiary of Lockheed Martin Corporation, for the U.S. Department of Energy's National Nuclear Security Administration under contract DE-AC04-94AL85000.
10:30 AM - *CM2.8.03
Extending the Length and Time Scale of Discrete Dislocation Dynamics by Exploiting Hot-Spot Detection and Grain-Based Parallelization
Daniel Weygand 1 , Markus Stricker 1
1 , Karlsruhe Institute of Technology, Karlsruhe Germany
Show AbstractThe discrete dislocation dynamics approach allows to study the plastic response of samples in the micrometer range of single [1-4] or polycrystalline structures [5,6]. The method allowed to identify new mechanisms, e.g. the multiplication of dislocations by the glissile reaction [4]. Furthermore, due to the use of the plastic displacements, both plastic strain and spin information is locally available. This allows to determine lattice rotations and local misorientation gradients, e.g. in imposed stress/strain gradients or due to grain boundaries. This enables to link the results of a simulations directly to experimental measurements by EBSD or µLaue diffraction.
Nevertheless the method is limited to rather small systems and short times due to the computational resources needed. In the current contribution, an analysis of the dislocation microstructure is presented, which allows to improve the detection of hot-spots requiring more frequent recalculations, extending the current incremental scheme [1]. Furthermore, to improve scalability for polycrystalline samples, the superposition approach of O’Day and Curtin [7] is followed.
[1] D. Weygand, L.H. Friedman, E. Van der Giessen, A. Needleman, Model. Simul. Mater. Sci. Eng. 10 (2002) 437.
[2] J. Senger, D. Weygand, C. Motz, P. Gumbsch, O. Kraft, Acta Mater. 59 (2011) 2937.
[3] P.D. Ispánovity, Á. Hegyi, I. Groma, G. Györgyi, K. Ratter, D. Weygand, Acta Mater. 61 (2013) 6234.
[4] M. Stricker, D. Weygand, Acta Mater. 99 (2015) 130.
[5] F. Šiška, D. Weygand, S. Forest, P. Gumbsch, Comput. Mater. Sci. 45 (2009) 793.
[6] M. Stricker, J. Gagel, S. Schmitt, K. Schulz, D. Weygand, P. Gumbsch, Meccanica (2015).
[7] M.P. O’Day, W.A. Curtin, J. Appl. Mech. 71 (2005) 805.
CM2.9: Discrete Dislocations to Plasticity
Session Chairs
Thursday PM, April 20, 2017
PCC North, 100 Level, Room 125 B
11:15 AM - *CM2.9.01
Influence of Dislocation Cross-Slip and Collinear Annihilation on the Evolution of Dislocation Density and Strain Hardening
Benoit Devincre 1
1 , LEM, CNRS-Onera, Chatillon France
Show AbstractIn all crystalline materials, strain hardening occurs because of an increase of the dislocation density. When the elementary mechanisms controlling dislocation storage, i.e. dislocation dipole and junction formation and dislocation-interface interaction are extensively studied, their counterpart contributing to a decrease of the dislocation density receive much less attention. In this study, the two main mechanisms promoting stain softening in bulk materials are investigated with the help of DD simulations. Therefore, the relative contributions of dislocation cross-slip and collinear annihilation to the plasticity of FCC single crystal are systematically studied. Comparison is made between single-slip and multi-slip deformation mode. The impact of these two mechanisms with increasing dislocation density is also presented.
11:45 AM - CM2.9.02
Developing a Crystal Plasticity Model for Metallic Materials Based on the Discrete Element Method
Agnieszka Truszkowska 1 2 , Qin Yu 1 , Alex Greaney 4 , T. Matthew Evans 2 , Jamie Kruzic 3
1 School of Mechanical, Industrial, & Manufacturing Engineering, Oregon State University, Corvallis, Oregon, United States, 2 School of Civil & Construction Engineering, Oregon State University, Corvallis, Oregon, United States, 4 Department of Mechanical Engineering, University of California, Riverside, Riverside, California, United States, 3 School of Mechanical and Manufacturing Engineering, UNSW Sydney, Sydney Austria
Show AbstractMetallic material failures due to plastic and/or creep deformation occur by the emergence of necking, microvoids, and cracks at heterogeneities in the material microstructure. While many traditional deformation modeling approaches have difficulty capturing these emergent phenomena, the discrete element method (DEM) has proven very effective for the simulation of materials whose properties and response vary over multiple spatial scales, e.g., granular materials. The DEM framework inherently provides mesoscale simulation level that can be used to model macroscopic response of a microscopically diverse system. DEM naturally captures the heterogeneity and geometric frustration inherent to deformation processes. While DEM has recently been adapted successfully for modeling the fracture of brittle solids, to date it has not been used for simulating metal deformation. In this paper we present our progress in reformulating DEM to model the key elastic, plastic, and visco-plastic deformation characteristics of FCC polycrystals to create an entirely new crystal plasticity modeling methodology well-suited for the incorporation of heterogeneities and simulation of emergent phenomena.
Acknowledgments: This material is based upon work supported by the Department of Energy National Energy Technology Laboratory under Award Number DE-FE0024065.
12:00 PM - *CM2.9.03
FFT-Based Algorithms for Micromechanical Analysis of Polycrystalline Metals at the Mesoscale
Ricardo Lebensohn 1 , Reeju Pokharel 1
1 , Los Alamos National Laboratory, Los Alamos, New Mexico, United States
Show AbstractEmerging characterization methods in experimental mechanics [1] pose a challenge to modelers to devise efficient formulations to interpret and exploit the massive amount of data generated by these novel techniques. In this talk we report recent advances in Fast Fourier Transform-based micromechanical modelling [2], which can use direct input from voxelized microstructural images of polycrystalline aggregates to predict their local and effective response [3]. New FFT-based algorithms for complex constitutive behaviors of plastically deforming materials, including elasto-viscoplasticity [4], dilatational plasticity [5] and non-local plasticity [6] will be highlighted.
[1] U. Lienert et al.: “High-energy diffraction microscopy at the Advance Photon Source”. JOM 63, 70 (2011).
[2] H. Moulinec and P. Suquet: “Numerical method for computing the overall response of nonlinear composites with complex microstructure”. CMAME, 157, 69 (1998).
[3] R. Pokharel et al.: “Polycrystal plasticity: comparison between grain scale observations of deformation and simulations”, Ann. Rev. Condens. Matter Phys. 5, 317 (2014).
[4] R.A. Lebensohn, A.K. Kanjarla and P. Eisenlohr: “An elasto-viscoplastic formulation based on fast Fourier transforms for the prediction of micromechanical fields in polycrystalline materials”. IJP, 32-33, 59 (2012).
[5] R.A. Lebensohn et al.: "Modelling void growth in polycrystalline materials", Acta Mater. 61, 6918 (2013).
[6] R.A. Lebensohn and A. Needleman: “Numerical implementation of non-local polycrystal plasticity using Fast Fourier Transforms”. JMPS, in press.
12:30 PM - *CM2.9.04
Recent Progress in the Concurrent Atomistic-Continuum Method and Its Applications to Thermal Phonon Transport
Xiang Chen 1 , Weixuan Li 1 , Liming Xiong 3 , David McDowell 2 , Youping Chen 1
1 , University of Florida, Gainesville, Florida, United States, 3 , Iowa State University, Ames, Iowa, United States, 2 , Georgia Institute of Technology, Atlanta, Georgia, United States
Show Abstract
In this talk we present recent progress in the concurrent atomistic-continuum (CAC) method with regard to its applications to thermal phonon transport across grain boundaries. The CAC method is based on a unified atomistic-continuum formulation of conservation laws as an extension of Kirkwood’s statistical mechanical theory of transport processes. It differs from the Kirkwood’s formulation by a concurrent two-level description of a crystalline material as a continuous collection of material points (unit cells) while each material point possesses internal degrees of freedom that describe the movement of atoms inside each unit cell. This two-level structural description of crystalline materials leads to a concurrent atomistic and continuum method for multiscale simulation of defect dynamics and transport processes within one single theoretical framework. As a result, CAC can simulate complex crystalline materials, reproduce both acoustic and optical phonon branches, resolve details to full atomistic resolution at interfaces or in other regions of interest while coarse-graining to thousands of atoms per finite element elsewhere, and simulate dynamics on the length and time scales of interest.
The propagation of a heat pulse in a single crystal and across grain boundaries (GBs) is simulated and visualized by using CAC furnished with a coherent phonon pulse model. With a heat pulse that is constructed based on a Bose-Einstein distribution of phonons, this work has reproduced the phenomenon of phonon focusing in single and polycrystalline materials. Simulation results provide visual evidence that the propagation of a heat pulse in crystalline solids with or without GBs is partially ballistic and partially diffusive. In particular, there is ballistic and diffusive thermal transport co-exist, with the long-wavelength phonons exhibiting ballistic transport and the short-wavelength phonons manifesting diffusive transport via scattering. To gain a quantitative understanding of GB thermal resistance, the kinetic energy transmitted across GBs is monitored on the fly and the time-dependent energy transmission for each specimen is measured; the contribution of coherent and incoherent phonon transport to energy transmission is estimated. Simulation results reveal that the presence of GBs modifies the nature of thermal transport, with the coherent long-wavelength phonons dominating the heat conduction. It is also found that the phonon-GB interaction can result in the reconstruction of the GBs.
CM2.10: Crystal Plasticity
Session Chairs
Thursday PM, April 20, 2017
PCC North, 100 Level, Room 125 B
2:30 PM - *CM2.10.01
Crystal Plasticity Simulations on Real Data—Towards Highly Resolved 3D Microstructures
Martin Diehl 1 , Yannick Naunheim 1 , Lutz Morsdorf 1 , Dayong An 1 , Franz Roters 1 , Dierk Raabe 1
1 , Max-Planck-Institut für Eisenforschung GmbH, Düsseldorf Germany
Show AbstractCrystal plasticity (CP) simulations are a powerful tool to investigate the mechanical performance of metallic materials. There are essentially two routes to obtain the model microstructure, i.e. grain morphology and orientation: Either use measured data or create an artificial microstructure. Voronoi tessellations are an examples for the latter approach that allow to scan the parameter space of even not producible microstructures. Experimental characterization techniques to obtain input data include electron back scatter diffraction (EBSD) imaging and synchrotron measurements.
While artificial microstructures are advantageous for large simulation studies to investigate systematically the influence of selected features (Diehl et al. 2016), their complexity is typically limited in comparison to real microstructures. For example, the orientation scatter within individual grains is typically neglected as each grain is considered to have one fixed orientation. The challenges associated with the use of measured microstructures are data conversion, time consuming experimental characterization, and filtering techniques to handle e.g. missing data and measurement artifacts. Recent developments in the field of experimental characterization, however, enable the acquisition of highly resolved microstructures at high quality that are suitable for micro mechanical simulations.
Based on simulations on various materials we show the current status of simulations on real microstructures. The simulations are conducted with a spectral solver (Eisenlohr et al. 2013, Shanthraj et al. 2014) implemented in the Düsseldorf Advanced Material Simulation Kit by Roters et al. (2012). Since the spectral method operates on a regular grid, it enables an easy takeover of measured data which typically exists in the form of pixels (2D) or voxels (3D).
It is shown how the availability of highly resolved microstructures enables new insights into stress and strain partitioning in polycrystalline materials. Of special importance is the fact that techniques like 3D EBSD allow the acquisition of three dimensional data sets and that the presented simulation approach is able to handle these microstructures within moderate computation times.
References
Diehl, M. et al. (2016). Neighborhood influences on stress and strain partitioning in dual-phase microstructures. Meccanica, 51, 429–441.
Eisenlohr, P. et al. (2013). A spectral method solution to crystal elasto-viscoplasticity at finite strains. Int. J. Plast., 46, 37–53.
Shanthraj, P. et al. (2014). Numerically robust spectral methods for crystal plasticity simulations of heterogeneous materials. Int. J. Plast., 66, 31–45.
Roters, F. et al. (2012). DAMASK: the Düsseldorf Advanced MAterial Simulation Kit for studying crystal plasticity using an FE based or a spectral numerical solver. Procedia IUTAM (Vol. 3, pp. 3–10).
3:00 PM - CM2.10.02
Crystal Plasticity Simulation of Aluminum Lithium Alloy—Verification and Experiments
Ali Ramazani 1 , Veera Sundararaghavan 1
1 , University of Michigan, Ann Arbor, Michigan, United States
Show AbstractAl-Li alloy is a polycrystalline material exhibiting high anisotropy in yield and tensile/compression strength, which is a consequence of the strong crystallographic texture of the material. During deformation processes, the crystallographic slip and lattice rotation are the primary means of plastic deformation in such materials. By controlling texture evolution during deformation, it is possible to optimize the process parameters and accordingly design the optimum microstructure to achieve the tailored properties. In the current study, a crystal plasticity based model is utilized to study texture evolution of Al-Li alloy under different loading conditions. EBSD measurements were made in all surfaces (L-LT, L-ST and LT-ST) of samples in both as-received (forged specimens) and deformed conditions (compression & tension tests). Grain Orientation Spread Method is used to quantify the percent of recrystallization in as received specimens as well as the heat-treated ones at 510 C for 1 h and 4 hrs. 5% - 15% recrystallization is measured at L direction of as-received specimens. Recrystallization effect is then neglected in the developed model. The precipitates, which decorate the grain boundaries in the as-received material, are dissolved after performing heat treatment at 510 C for 1 hr. No grain growth, texture evolution, or recrystallization was observed after 510 C for up to 4 hrs. Based on the experimental results, an orientation distribution function (ODF) model is developed to simulate the texture evolution at different strains in various loading conditions. Here, the polycrystalline microstructure is represented through a finite element discretized orientation distribution function and texture evolution is modeled using ODF conservation laws via Taylor assumption. Additionally, an elasto-visco plastic crystal plasticity framework is utilized to predict crystal reorientation. Numerical results show very good agreement with the experimental texture measurements.
3:15 PM - *CM2.10.03
Multiscale Modeling of Plasticity Using Data Science Approaches
Surya Kalidindi 1
1 , Georgia Institute of Technology, Atlanta, Georgia, United States
Show AbstractIn recent work, we have demonstrated the viability and computational advantages of using a compact database of discrete Fourier transforms (DFTs) for facilitating crystal plasticity solutions in cubic polycrystalline materials subjected to arbitrary deformation paths. This new DFT database approach allows for compact representation and fast retrieval of crystal plasticity solutions, which is found to be able to speed up the calculations by about two orders of magnitude. In this paper, we present the successful implementation of this spectral database approach in a commercial finite element code to permit computationally efficient simulations of heterogeneous deformations using crystal plasticity theories. More specifically, the spectral database approach to crystal plasticity solutions was successfully integrated with the commercial finite element package ABAQUS through a user materials subroutine, UMAT. Details of this new crystal plasticity spectral database-FE approach are demonstrated and validated through a few example case studies for selected deformation processes on face centered and body centered cubic metals. The evolution of the underlying crystallographic texture and its associated macroscale anisotropic properties predicted from this new approach are compared against the corresponding results from the conventional crystal plasticity finite element method. It is observed that implementing the crystal plasticity spectral database in a FE code produced excellent predictions similar to the classical crystal plasticity FE method, but at a significantly faster computational speed and much lower computational cost.
4:00 PM - CM2.10.05
Modeling of Dynamic Recrystallization in Austenitic Stainless Steel 304L by Coupling a Full Field Approach in a Finite Element Framework with Mean Field Laws
Ludovic Maire 1 , Nathalie Bozzolo 1 , Charbel Moussa 1 , Marc Bernacki 1
1 , CEMEF Mines ParisTech, Sophia-Antipolis France
Show AbstractMean field (MF) models of dynamic recrystallization (DRX) emerged in the last decades with the intention to implicitly describe the microstructure by considering grains sets as spherical classes. These models have the advantage to provide accurate results in terms of macroscopic results such as recrystallized fraction or grain size but also to provide additional information in terms of grain size distribution and dislocation density distribution [1,2,3,4].
In parallel, finer approaches called full field (FF) models have emerged in the last decades. These approaches consider a complete description of the microstructure topology at the polycrystal scale [5]. A review of the most significant numerical methods can be found in [6].
Several DRX models based on a full field approach can already be found in the literature [7,8,9]. Although literature already provides a large number of papers on full-field DRX models, major drawbacks are either they are developed in 2D and/or they only consider small deformations (< 20%).
In the present work, a 3D model based on the level-set method in a FE framework is employed to model the DRX phenomenon in austenitic stainless steel 304L at large deformations. The level-set approach coupled to a remesher provides an accurate tracking of interfaces (i.e. grain boundaries) all along the simulation while mean field laws are used for the nucleation and work hardening mechanisms.
A first part of this work is dedicated to a presentation of the level-set approach and the constitutive equations of this model. That part is followed by a sensibility study concerning the choice of the initial number of grains and elements so that the model correctly describes experimental results on 304L. The subsequent part presents a comparison between an enriched DRX mean field models [2] and experimental results. Some remarks about the choice of use either a mean field or full field model will conclude this work.
References
[1] Montheillet, F., Lurdos, O., and Damamme, G. (2009). Acta Materialia, 57(5):1602–1612.
[2] Bernard, P., Bag, S., Huang, K., and Logé, R. (2011). Science and Engineering: A, 528(24):7357–7367.
[3] Cram, D. G., Zurob, H. S., Brechet, Y. J. M., and Hutchinson, C. R. (2009). Acta Materialia, 57(17):5218–5228.
[4] Maire, L., Moussa, C., Bozzolo, N., Scholtes, B., Pino Muñoz, D., Bernacki, M. (2016). Journal of Materials Science, 51(24):10970-10981.
[5] Scholtes, B., Shakoor, M., Settefrati, A., Bouchard, P.-O., Bozzolo, N., and Bernacki, M. (2015). Computational Materials Science, 109:388–398.
[6] Hallberg, H. (2011). Metals, 1(1):16–48.
[7] Mellbin, Y., Hallberg, H., and Ristinmaa, M. (2015). Modelling and Simulation in Materials Science and Engineering, 23(4):045011.
[8] Sitko, M., Pietrzyk, M., and Madej, L. (2016). Journal of Computational Science, 16:98–113.
[9] Zhao, P., Song En Low, T., Wang, Y., and Niezgoda, S. R. (2016). International Journal of Plasticity, 80:38–55.