Brief introductionUnlocking nanoworld mysteries

Multiscale modeling and simulation of materials

We are performing multiscale modeling and simulation of various materials, mainly focusing on their mechanical behaviors such as deformation and fracture but also investigating other physical properties. Our simulation methods include the first-principles density functional theory, all-atom and coarse-grained molecular dynamics, the phase field method and the finite element method.

Research topicsAtomistic modeling simulation of anode materials of solid oxide fuel cells

Solid oxide fuel cells (SOFCs) are one of the most promising energy sources in the next generation due to their high efficiency. We target anode materials of SOFCs and perform classical molecular dynamics simulations using a reaction force field (ReaxFF) based on first-principles calculations, to reveal the mechanisms of chemical reacion at triple phase boundary (TPB) in the power generation process and reduction of cermet materials in the nanostructure formation process.

• "Atomic structure observations and reaction dynamics simulations on triple phase boundaries in solid-oxide fuel cells"
  Shu-Sheng Liu, Leton C. Saha, Albert Iskandarov, Takayoshi Ishimoto, Tomokazu Yamamoto, Yoshitaka Umeno, Syo Matsumura and Michihisa Koyama
  (Communications Chemistry 2 (2019) 48, doi:10.1038/s42004-019-0148-x)

Development of new multiscale simulation method for nano-micro fatigue problem (CREST project)

Fatigue of materials, which often occurs in metals, is a phenomenon that arises when the materials are under cyclic load, leading to crack initiation, crack propagation and fracture. Fatigue has long been an important issue in the field of mechanical engineering as fatigue is one of major causes of materials' failure in various industrial products. Because of its industrial importance and also scientific profoundness, the fatigue problem has attracted a number of research efforts. Nevertheless, microscopic mechanisms of fatigue crack initiation have been little understood with various hypotheses remaining to be confirmed by clear evidence.

With the initiative of a JST-CREST program for 'Nanomechanics' research area, we are working on development of a novel multiscale simulation method to elucidate fatigue mechanisms at the nano and micrometer scales. Utilizing machine-learning, the new method combines deductive (bottom-up) and inductive (top-down) multiscale approaches, by which one can perform multiscale simulations overcoming drawbacks of both the conventional schemes.

Currently we are focusing on numerical simulation of evolution and interaction of dislocations under stress based on the phase-field modeling initiated by Khachaturyan et al. and later developed by Wang et al. In addition, molecular dynamics simulation of dislocation motion under cyclic load is being performed with the aim to clarify the mechanism of slip deformation triggered by defects and surfaces.

We are also working on a test implementation of the DIMS concept. To start with, we adopted the reaction-diffusion model of dislocations based on the Walgraef-Aifantis method to observe the formation of dislocation patterns at the micrometer scale.

Modeling and simulation of flexoelectricity in nanomaterials (KAKENHI project)

Flexoelectricity is electric polarization due to the gradient of strain field. In contrast to ordinary piezoelectricity, which is electric polarization that changes due to change in strain, flexoelectricity has been almost ignored for practical application because its effect in macroscopic materials is quite marginal. Flexoelectricity in nanomaterials, however, is drawing attention as the effect can be non-trivial in nanomaterials where very large gradient of strain field may be realized easily. Since the nanoscale mechanisms of flexoelctricity has been little understood, investigations from various aspects are being awaited.

Flexoelectricity in nanomaterials can exhibit unexpected behaviors owing to very large magnitude of gradient in strain disribution and substantial fraction of surface area to volume. Fully understanding the nanoscale mechanisms of flexoelectricity may contribute to realization of novel nanodevices with outstanding functions.

Our group is starting investigations of flexoelectricity in nanomaterials by means of numerical simulations based on the atomistic and the phase field models, with the aim to elucidate fundamental mechanisms of flexoelectricity. We have successfully developed in-house simulation codes for phase field model simulations, which are combined with graphic libraries and toolkits (e.g. GLUT; OpenGL Utility Toolkit) for visualization.

Multiphysics of structural bifurcation in nanostrucures (KAKENHI project)

Nanometer-sized materials can exhibit peculiar deformation and fracture behavior that would not appear in macroscale materials. One of the reasons is that nanostructures are susceptible to atomic-level structures. For example, slight difference in atomic structure at the site of deformation initiation can essentially affect deformation and fracture behaviors. This points out a vast area of physics to be unveiled and also suggests potential applications; e.g., controlling deformation behavior by tuning atomistic structures.

Our group is starting atomistic model analyses of structural bifurcation in nanostructures using ASIA (Atomistic Structural Instability Analysis) method that we have developed thus far. Our previous studies have revealed that latent instability modes compete with each other and become activated to exhibit resulting unstable deformation. With this approach we are invstigating the mechanism of various deformation modes in nanostructures, trying to figure out relevant factors to transition between brittle and ductile fractures.

One of interesting scientific questions in the nanoscale mechanisms of crystal slips. State-of-the-art experiments observe twin boundary formation due to sequential slips occuring adjoining slip planes as well as scattered slips resulting in no twin boundary. To reveal the mechanism and identify factors determining whether twin boundaries are formed, we have been conducting extensive atomistic model simulations of deformation of nanostructures. Our simulation results suggest unexpectedly large contribution of nano-scale structure (such as surface steps) on criteria of dislocation initiation from surfaces.

Multiscale simulation of structural polymer materials and composites

As the application of plastic composite materials to automobiles and airplanes is getting more and more accelerated these days, it is urged to establish guidelines for designing polymer materials possessing both high strength and toughness. We perform a 'bottom-up type' multi-scale simulation aided by the coarse-grained molecular dynamics method with the aim to obtain constitutive laws of materials by numerical simulations. We also carry out finite-element method (FEM) simulations of crack propagation problems to build a model that reproduce experimental results and to clarify the mechanism of crack propagation.

• "Velocity mode transition of dynamic crack propagation in hyperviscoelastic materials: A continuum model study" A. Kubo and Y. Umeno
  (Scientific Reports 7 (2017) 42305 doi:10.1038/srep42305)

• "Construction of master yield stress curves for polycarbonate: a coarse-grained molecular dynamics study"
  A. Kubo, J.-M. Albina and Y. Umeno
  (Polymer 177 (2019) 84-90, doi:10.1016/j.polymer.2019.05.045)

Molecular simulation for lubrication mechanism of metal surfaces with hetero-nano structure

Experiments have demonstrated peculiar lubrication properties of steel having hetero-nano structures (i.e., ultrafine grain structures). With the aim to eucidate its mechanism, we perform coarse-grained molecular dynamics simulations of the system consisting of hetero-nano structured metal surfaces and lubricant molecules. We are tackling this problem from the viewpoint of polymer brush structures formed on the metal surfaces and their relation to grain structures.

• "Coarse-grained molecular dynamics simulations of boundary lubrication on nanostructured metal surfaces"
  J.-M. Albina, A. Kubo, Y. Shiihara and Y. Umeno
  (Tribology Letters 68 (2020) 49 (9 pages), doi:10.1007/s11249-020-1276-2)

Analysis of fracture in environmental barrier coatings for ceramics

We perform finite element method (FEM) simulations of environmental barrier coating (EBC) for SiC-based ceramics used in aircraft engines. We calculate energy release rates of crack propagation under practical mechanical and thermal conditions in order to evaluate the reliability of EBC.

• "Crack initiation criteria in EBC under thermal stress"
  Emi Kawai, Hideki Kakisawa, Atsushi Kubo, Norio Yamaguchi, Taishi Yokoi, Takashi Akatsu, Satoshi Kitaoka and Yoshitaka Umeno
  (Coatings 9 (2019) 697 (27 pages), doi:10.3390/coatings9110697)

Ideal strength of solids

The ideal strength is defined as the maximum stress that a perfect crystal can attain when the crystal undergoes uniform (ideal) deformation. While macroscopic materials possess far smaller strength than the ideal strength due to defects being sources of fracture, nano-scale materials with few or no defects can be closer the ideal strength. The ideal strength is also a key index to understand elementary process of plastic deformation in crystals. For example, the critical shear stress for dislocation nulcreation in pristine crystals is close to the ideal shear strength. Our group has been evaluating ideal strengths of various crystals by means of density functional method calculations with the aim to reveal fundamental mechanical properties of crystals.

Ideal shear strength of crystals

When dislocations are nucleated, crystal lattices in the vicinity of the nucleation site undergoes shear deformation. The nucleation condition should be closely related with the crystal strength of ideal shear deformation, i.e., the ideal shear strength. We have performed first-principles density functional theory calculations of Si, SiC and GaN to obtain their ideal shear strengths. For SiC and GaN, we also took into account the effect of superimposed normal stress components on the shear strength, which provides useful information to understand crystals' behavior under real situation.

• Materials Science and Engineering: B, 88-1(2002), pp.79-84
  
• Modelling and Simulation in Materials Science and Engineering, 15 (2007), pp.27-37

Effect of normal stress on ideal shear strength

In real situation, crystals undergo various combinations of stress components. For example, shear stress on slip planes together with compression is exerted on crystal lattices in dislocation nucleation in a nano-indentation test. It is known that, in metal, the critical shear stress (ideal shear strength) is usually increased by compression. Our group has examined the effect of normal stress on ideal shear strength of covalent crystals. We have revealed that, in covalent crystals, the response of shear strength to normal stress depends on the type of crystal, meaning that the ideal shear strength is decreased by compression in some covalent crystals, in contrast to the phenomenon in metal.

• Physical Review B, 77 (2008), art. 100101R
  
• Journal of Physics: Condensed Matter, Vol. 23 (2011), art. 385401

Tensile strength of Si nanofilms

To evaluate the effect of ideal (100) surfaces on strength, first-principles calculations of tension in silicon nanofilms were performed.

• Physical Review B, 72 (2005), art. 165431

Ideal shear strength under finite temperatures

In reality, materials deform at finite (non-zero) temperatures. It is therefore important to evaluate how temperature affects ideal strength of materials because we need to understand whether discrepancy in critical stress of deformation between experiment and theory stems from the effect of defects or temperature. This study investigates the effect of temperature on strength by molecular dynamics simulations of ideal crystals having no defects at finite temperatures.

• Physical Review B, Vol. 84 (2011), art. 224118

Atomistic structural instability analysis

In the elementary process of plastic deformation, sliding of atomic arrangement occurs due to, e.g., dislocation motion and twin formation. It is therefore necessary to consider the process at the atomistic level in order to fully understand the characteristics of material deformation. Such phenomena like dislocation emission/motion can be interpreted as the outcome of structural instability of local atomic arrangements. In such a context, we have been trying to reveal the mechanism of emergence of structural instability at the atomistic level.

We have proposed an original scheme called "Atomistic Structural Instability Analysis; ASIA" that can rigorously evaluate the structural instability in arbitrary atomic structures. With this approach we are making efforts towards the essence of plastic deformation.

• Computational Materials Science, Vol.29-4 (2004), pp.499-510
  
• Materials Science and Engineering: A, Vol.379/1-2 (2004), pp.229-233
  
• Materials Science and Engineering: A, 462/1-2 (2007), pp.450-455
  
• Physical Review B, 80 (2010), art. 104108 (11 pp)
  
• Key Engineering Materials, Vols.592-593 (2014), pp.39-42

Ferroelectric capacitors and piezoelectric materials

Ferroelectric materials exhibit spontaneous electric polarization with charged atomic positions deviated from ideal (symmetric) lattice sites and can be applied for microscopic memory devices . For example, an ultrathin ferroelectric film sandwiched with electrode layers can work as a non-volatile memory device, which is a promising low-energy information storage. We provide theoretical estimations of fundamental properties of such nanostructures by first-principles calculations, including the critical thickness of ferroelectric instability, which is useful information for design.

• Physical Review B, 74 (2006), art. 060101R
  
• Physical Review B, 80 (2009), art. 205122 (8 pp)

Deformation of CNTs (KAKENHI project)

Electronic structures of carbon nanotubes can be modulated by the application of deformation. Such an effect can be utilized to fabricate novel nano-devices. In particular, carbon nanotubes can exhibit buckling deformation under compressive loads. Utilizing large and sudden change in the atomistic structure associated with buckling may pave the way for new nano-devices where abrupt and eminent modulation of electronic properties such as band-gap energy is expected with small input of mechanical loading. We perform molecular dynamics simulations and electronic structure calculations to clarify the mechanical and physical behaviors of carbon nanostructures.

• "Axial buckling behavior of single-walled carbon nanotubes: Atomistic structural instability analysis" Masanobu Sato, Hiroyuki Shima, Motohiro Sato and Yoshitaka Umeno (Physica E 106 (2019) pp. 319-325, doi:10.1016/j.physe.2018.05.035)
• "On the atomistic energetics of carbon nanotube collapse from AIREBO potential" Yoshitaka Umeno, Yu Yachi, Motohiro Sato and Hiroyuki Shima (Physica E 106 (2019) pp. 319-325, doi:10.1016/j.physe.2018.08.006)
• "Buckling-induced band gap modulation in zigzag carbon nanotubes" Yoshitaka Umeno, Masanobu Sato, Motohiro Sato and Hiroyuki Shima (Physical Review B 100 (2019) 155116, doi:10.1103/PhysRevB.100.155116)
• "Scaling law for the onset of surface wrinkling of multilayer tubes" Motohiro Sato, Kazusa Ishigami, Hiroyuki Kato, Yoshitaka Umeno and Hiroyuki Shima (Extreme Mechanics Letters 40 (2020) 100970 (6 pages), doi:10.1016/j.eml.2020.100970)
• "Diameter-change-induced transition in buckling modes of defective" Y. Umeno, A. Kubo, C. Wang and H. Shima (Nanomaterials, 12 (2022) 2617, doi:10.3390/nano12152617)

Magnetic materials under deformation

Magnetic materials draw a lot of attention as their applications to information storage and spintronics devices are highly expected. Since extremely large strain may arise locally due to mismatch at component interfaces, it is demanded to clarify the effect of strain (or stress) on magnetic properties. We have investigated various magnetic materials using first-principles calculations to reveal the effect of strains on magnetic properties.

• Journal of Physics: Condensed Matter 24 (2012), 245501 doi:10.1088/0953-8984/24/24/245501
  
• Journal of Materials Research, Vol. 28, No. 12 (2013), pp.1559-1566

Surface stress response to charging

Experiments revealed that nano-porous metals immersed in electrolyte liquid exhibit expansion and shrinkage under electric charge, which is considered as the effect of surface charge causing change in surface stress. We have revealed the mechanism of the phenomenon by first-principles calculations.

• Europhysics letters, Vol. 78 (2007), 13001
  
• Europhysics letters, Vol. 84 (2008), 13002 (6 pp)
  
• Physical Review B 85 (2012), 125118 (5 pp)

Stacking fault of SiC

SiC is a promising material for semiconducting power devices. A challenging problem is that various structural defects emerge in the fabrication process of the device component. Engineering techniques to reduce such structural defects are therefore expected. In such a context, we aim to provide fundamental properties of SiC crystals such as defect energies and mobility of defects. We have evaluated stacking fault energies in 3C-SiC under the environment of nitrogen inclusion using first-principles calculations. We have also simulated dislocation motion in SiC crystals under stress by classical molecular dynamics using an originally developed interatomic potential function.

• "Ab initio density functional theory calculation of stacking fault energy and stress in 3C-SiC"
  Y. Umeno, K. Yagi and H. Nagasawa
  (Physica Status Solidi (b) 249, No.6 (2012) pp.1229-1234, DOI: 10.1002/pssb.201147487)
  
• "Molecular dynamics study of deformation and fracture in SiC with angular dependent potential model"
  A. Kubo, S. Nagao and Y. Umeno
  (Computational Materials Science 139 (2017) pp. 89-96 doi:10.1016/j.commatsci.2017.07.023)

Construction of interatomic potentials

For realizing atomistic simulations with high accuracy, it is essential to construct reliable interatomic potential functions (which are also called "force fields"). Especially, it is demanded to construct potentials tha can reproduce multiple properties of materials. To achieve this, we need to adopt relatively complex functional forms and employ a well-organized parameter fitting scheme. As a result, constructed potentials can reproduce material properties under various local environments (e.g. surface, crystals under strain, etc.), which is called "transferability". Thus far, we have succeeded in constructing highly reliable potential functions for various systems including SiC (ADP model), YSZ (Dipole and ReaxFF models), Neodymium magnet (ADP model), etc. We are also developing a new scheme to efficiently calculate electronic density of states by means of the artificial neural network model.

• "Development of interatomic potential for Nd-Fe-B permanent magnet and evaluation of magnetic anisotropy near interface and grain boundary" A. Kubo, J. Wang and Y. Umeno
  (Modelling and Simulation in Materials Science and Engineering, Vol. 22, No.6 (2014), 065014 doi:10.1088/0965-0393/22/6/065014)
  
• "Development of a new dipole model: interatomic potential for yttria-stabilized zirconia for bulk and surface"
  A.M. Iskandarov, A. Kubo and Y. Umeno
  (Journal of Physics: Condensed Matter 27 (2015) 015005 (9pp) doi:10.1088/0953-8984/27/1/015005)
  
• "Atomistic modeling study of surface effect on oxide ion diffusion in yttria-stabilized zirconia"
  A.M. Iskandarov and Y. Umeno
  (Solid State Ionics 279 (2015) pp.46-52 doi:10.1016/j.ssi.2015.07.014)
  
• "Effect of cation dopants in zirconia on interfacial properties in nickel/zirconia systems: An atomistic modeling study"
  A. Iskandarov, Y. Ding and Y. Umeno
  (Journal of Physics: Condensed Matter 29 (2017) 045001 doi:10.1088/1361-648X/29/4/045001)
  
• "Molecular dynamics study of deformation and fracture in SiC with angular dependent potential model"
  A. Kubo, S. Nagao and Y. Umeno
  (Computational Materials Science 139 (2017) pp. 89-96 doi:10.1016/j.commatsci.2017.07.023)
  
• "Atomic structure observations and reaction dynamics simulations on triple phase boundaries in solid-oxide fuel cells"
  Shu-Sheng Liu, Leton C. Saha, Albert Iskandarov, Takayoshi Ishimoto, Tomokazu Yamamoto, Yoshitaka Umeno, Syo Matsumura and Michihisa Koyama
  (Communications Chemistry 2 (2019) 48, doi:10.1038/s42004-019-0148-x)
  
• "Prediction of Electronic Structure in Atomistic Model Using Artificial Neural Network"
  Y. Umeno and A. Kubo
  (Computational Materials Science 168 (2019) 164-171, doi:10.1016/j.commatsci.2019.06.005)
  
• "Prediction of Electronic Structure in Atomistic Model Using Artificial Neural Network"
  Y. Umeno and A. Kubo
  (Computational Materials Science 168 (2019) 164-171, doi:10.1016/j.commatsci.2019.06.005)