재료의 상변태 거동과 미세조직

시뮬레이션 심포지엄

일 시 : 2016년 3월 31일(목) - 4월 1일(금)

장 소 : 한국기계연구원 부설 재료연구소

주최 : 대한금속재료학회 전산재료과학분과위원회

ICAMS, Ruhr-University Bochum

한국기계연구원 부설 재료연구소

주관 : 한국기계연구원 부설 재료연구소 재료설계분석연구실

Symposium on Simulation of Phase Transformation

and Microstructure Evolution of Materials

March 31 – April 1, 2016

Co-organized by

Committee of Computational Materials Science Korean Institute of Metals and Materials

ICAMS, Ruhr-University Bochum Korea Institute of Materials Science

Schedule : Day 1, 2016. 3. 31 (Thu)

Time Speaker Title

13:30 ~ 14:00 Registration

14:00 ~ 15:00 Ingo Steinbach ICAMS, RUB*

Lecture : Introduction to phase-field modeling

15:00 ~ 15:30 Coffee break / Install a program package

15:30 ~ 17:30

Oleg Shchyglo R. Darvishi Kamachali

Alexander Monas ICAMS, RUB*

OpenPhase The open source phase field simulation library

17:30 ~ 18:00 Q & A

* Interdisciplinary Center for Advanced Materials Simulation, Ruhr-University Bochum

Tutorial: OpenPhase - the open source phase field simulation library

O. Shchyglo, R. D. Kamachali, A. Monas Email: oleg.shchyglo@rub.de

We present an open source phase field simulation library “OpenPhase” which is

based on a multi-phase field multi-component model of Steinbach et al.[1]. The library is intended for high quality quantitative simulations of processes that involve structural and phase transformations. It contains modules that allow simultaneously solving the multi-phase field equations, equations for advection and diffusion of different chemical components, Navier-Stokes equation for a liquid phase flow including the interaction with solid structures and large strain elasto-plastic problem. The modular structure of the library allows for easy extension and modification by adding new modules. Several simulation examples related to the metallurgical processes will be presented and explained.

The library is distributed in the form of an open and free of charge source code which is available at www.OpenPhase.de.

[1] I. Steinbach; Modelling and Simulation in Materials Science and Engineering, 17 (2009) 073001.

Schedule : Day 2, 2016. 4. 1 (Fri)

Time Speaker Title

09:00~09:20 Registration

09:20~09:30 Opening address

09:30~10:00 Ingo Steinbach ICAMS, RUB

Full-field simulation of solidification, coarsening and forming of polycrystals

10:00~10:30 Seong Gyoon Kim Kunsan Nat’l Univ.

Thermodynamic properties of phase-field models for grain boundary segregation

10:30~11:00 Munekazu Ohno Hokkaido Univ.

Quantitative phase-field modeling and its application to investigation of dendritic growth in alloys

11:00~11:20 Coffee break

11:20~11:50 Nong-moon Hwang

Seoul Nat’l Univ.

Comparison between 3-dimensional Monte Carlo simulations and real microstructures evolved during

secondary recrystallization of metals

11:50~12:20 Byeong-Joo Lee POSTECH

Some issues in phase-field modeling and simulations for grain boundary or surface segregation

12:20~13:30 Lunch / Registration

13:30~14:00 R.D. Kamachali ICAMS, RUB

Phase-field and mean field study of grain growth and recrystallization

14:00~14:30 Kunok Chang KAERI

Phase-field modeling of ODS particle behavior in the metallic system

14:30~15:00 Christian Schwarze ICAMS, RUB

Phase-field simulation of mechano-chemical coupling during precipitation in aluminum alloys

15:00~15:30 Yoon Suk Choi Pusan Nat’l Univ.

3D real microstructure-based crystal plasticity FEM simulations for a polycrystalline Ni deformed in tension

15:30~15:50 Coffee break

15:50~16:20 Pil-Ryung Cha Kookmin Univ.

Phase-field study on the formation of lath martensite

16:20~16:50 Monas Alexander

ICAMS, RUB Phase-field simulation of Mg-Al alloy solidification

16:50~17:20 Yongwoo Kwon

Hongik Univ.

Crystallization kinetics of nanoscale Ge2Sb2Te5 and its influence to data retention statistics of phase-change

memory

17:20~17:30 Closing remark

Full-field simulation of solidification, coarsening and forming of polycrystals

E. Borukhovich, A. Monas, M. Tegeler and I. Steinbach

Interdisciplinary Centre for Advanced Materials Simulation (ICAMS)

Ruhr-University Bochum, 44801 Bochum, Germany Email: ingo.steinbach@rub.de

The phase-field method has emerged as the method of choice for simulation of

microstructure evolution and phase-transformations in material science. It has wide applications in solidification and solid state transformations in general. Recently the method has been generalized to treat large deformation and damage in solids. A through process full-field simulation will be presented starting from solidification, homogenization heat treatment and forming and ending with the evolution of damage during large deformation. Aspects of numerical discretization, efficient numerical integration and massive parallelization will be discussed.

Thermodynamic properties of phase-field models for

grain boundary segregation

Seong Gyoon Kim1, Jae Sang Lee2 and Byeong-Joo Lee3

1 Department of Materials Science and Engineering, Kunsan National University

2 Graduate Institute of Ferrous Technology (GIFT), POSTECH 3 Department of Materials Science and Engineering, POSTECH

Email: sgkim@kunsan.ac.kr

Impurity atoms when segregated at grain boundary (GB) regions can dramatically change the physical and chemical properties of GBs. Such changes often appear to be attributed to the GB energy reduction and/or solute drag effect. Phase-field models have been utilized to clarify both the thermodynamic and kinetic effects of the GB segregation. In this study, we developed the phase-field models for GB segregation, which are the diffuse interface versions of the classical two-phase model of GB segregation. The thermodynamic state of any point in the system is represented as a mixture of a GB phase and a matrix phase. There are two choices for the thermodynamic relation between the GB phase and the matrix phase which constitute the point: the equal composition condition in the model I and the equal diffusion potential condition in the model II. Most previous PFMs for GB segregation appear to be the specific cases of the model I. We examined the thermodynamic properties of the model I and II, and compared with each other and the classical two-phase model. Although all the models resulted in the same GB composition, the GB energy and its dependency on the composition at equilibrium state appeared to be quite different from each other. In model I, there is a lower bound to the GB energy which originates from the equal composition condition. The GB energy from the model II shows no such a lower bound and it is represented as the vertical distance between the parallel tangent lines on the free energy diagram, as in the classical two-phase model. Nevertheless, the compositional dependence in the model II is quite different from that in the classical two-phase model. This originates from the different choice for the composition-independent parameter in the models: a constant gradient energy coefficient in the model II and a constant GB width in the classical two-phase model. The differences in thermodynamic properties found in this study have to be seriously taken into consideration in choosing a PFM for simulating GB motion and grain growth with segregation.

Quantitative phase-field modeling and

its application to investigation of dendritic growth in alloys

Munekazu Ohno1 , Tomohiro Takaki2 and Yasushi Shibuta3

1 Faculty of Engineering, Hokkaido University, Japan 2 Mechanical and System Engineering, Kyoto Institute of Technology, Japan

3 Department of Materials Engineering, The University of Tokyo, Japan Email: mohno@eng.hokudai.ac.jp

Understanding and controlling solidification microstructure of alloys are one of important subjects in the field of materials science and engineering, because the size and morphology of the solidified crystals and non-uniform distribution of alloying elements, viz., microsegregation determine the quality of as-cast materials. The alloy solidification is a multi-physics problem involving thermal diffusion, solute diffusion and fluid flow and so on. Hence, analyses and prediction of the solidification microstructure require these physics and their coupling to be precisely described.

The phase-field model has emerged as an effective computational tool to simulate microstructural evolution processes in multi-physics problems. Importantly, so-called quantitative phase-field model now enables us to carry out quantitative description and prediction of microstructural evolution processes [1, 2]. Combined with high-performance computational environments, it is now possible to carry out very large-scale simulation of solidification process, for instance, involving competition growth of bundles of dendrites during directional solidification [3, 4]. In this talk, our recent progress in the quantitative phase-field modeling is introduced. Also, we will show some examples where the phase-field simulations are successfully utilized to elucidate the formation processes of solidification microstructures.

[1] M. Ohno, Phys. Rev. E, 86 (2012), 051603.

[2] M. Ohno, T. Takaki and Y. Shibuta, Phys. Rev. E, 93 (2016), 012802.

[3] Y. Shibuta, M. Ohno and T. Takaki, JOM 67 (2015), 1793.

[4] T. Takaki, M. Ohno, T. Shimokawabe, and T. Aoki, Acta Mater., 81 (2014), 272.

Comparison between 3-dimensional Monte Carlo simulations and

real microstructures evolved during secondary recrystallization of metals

Nong-Moon Hwang1, Da-Hee Cho2 and Tae-Wook Na1

1 Seoul National University 2 Research Institute of Industrial Science and Technology

Email: nmhwang@snu.ac.kr

Secondary recrystallization, which is also called abnormal grain growth (AGG), often takes place after primary recrystallization of deformed polycrystalline materials. A famous example is the evolution of the Goss texture after secondary recrystallization of Fe-3%Si steel. As a mechanism of selective AGG of Goss grains, we suggested the sub-boundary enhanced solid-state wetting, which agrees with many microstructural features of AGG. Here, 3-dimensional (3-D) Monte Carlo simulations and real microstructure evolutions are compared. 3-D microstructures of abnormally-growing Goss grains are constructed by repeated serial polishing, etching and photographing of 100 ~ 130 slices using Robo-Met 3D after the initial stage of secondary recrystallization of Fe-3%Si steel by heating at 1050 ~ 1080oC for 5 min. The highly irregular shapes of abnormally-growing Goss grains are highly in contrast with the relatively regular shapes of matrix grains. Such highly irregular shapes come from the highly irregular migration rate of the growth front, which also produces peninsular-like grains at the growth front and island-like grains within the abnormally-growing grains. These microstructures are compared with the 3-D microstructures evolved using the 3-D Monte Carlo simulations based on sub-boundary enhanced solid-state wetting mechanism starting with the real misorientation distribution of grains after primary recrystallization of Fe-3%Si steel and considering the grain growth inhibition by precipitates.

Some issues in phase-field modeling and simulations for

grain boundary or surface segregation

Kyeong-Min Kim and Byeong-Joo Lee

Department of Materials Science and Engineering, POSTECH Email: calphad@postech.ac.kr

Phase-Field models provide the best simulation tool for the effect of grain

boundary (GB) or surface segregation of impurity atoms on microstructural evolution. Today, fundamental materials information such as the amount of solute segregation and resultant reduction of GB or surface energy can be provided by atomistic simulation techniques in a functional form of the orientation of GBs or surfaces. However, it looks difficult to reproduce the amount of solute segregation and resultant reduction of GB or surface energy, simultaneously, using the existing Phase-Field models.

In the present talk, a multiscale computational approach to develop highly value-added {100} textured steel sheets will be briefly outlined. Some difficulties met during the Phase-Field modeling and simulation, and the ways to avoid the difficulties will also be introduced, asking an advice from experts in the field of Phase-Field models.

Phase-field and mean field study of grain growth and recrystallization

R. Darvishi Kamachali and I. Steinbach

Interdisciplinary Centre for Advanced Materials Simulation (ICAMS)

Ruhr-University Bochum, 44801 Bochum, Germany Email: reza.darvishi@rub.de

Grain growth and recrystallization are common phenomena in production

processes and in-service materials. The many-body nature of these phenomena suggest multiphase-field method as a well suited efficient approach for studying these effects. On the other hand, today’s abilities for performing large-scale simulations offer possibilities for examining physical theories such as mean field approach.

In this presentation, we go through our recent phase-field studies concerning normal grain growth [1,2] and recrystallization [3] and discuss our results in terms of size, shape and texture of the grains in contrast to available mean field theories. We will show that new paradigms of self-similarity are possible in the size distribution of grains during coarsening. [1] R. Darvishi Kamachali, I. Steinbach; Acta Materialia 60 (2012) 2719.

[2] R. Darvishi Kamachali, A. Abbanandolo, K.-F. Siburg, I.Steinbach; Acta Materialia 90 (2015) 252.

[3] R. Darvishi Kamachali, S. J. Kim, I. Steinbach; Computational Materials Science 104 (2015) 193.

Phase-field modeling of ODS particle behavior in the metallic system

Kunok Chang and Junhyun Kwon

Korea Atomic Energy Research Institute Email: kunokchang@kaeri.re.kr

Oxide dispersion strengthened alloys (ODS) are strong candidate for the materials for the future nuclear system due to their outstanding creep and swelling resistance. Especially, since their mechanical properties and thermal stability is excellent, various Y2O3-based ODS have been intensively investigated. From the former experimental observations using HRTEM, it is known that Y2O3 particle on the α-Fe matrix present in the cuboidal shape and their morphology is heavily affected by the elastic effect. We investigated the morphological evolution of the Y2O3 in the Fe- and Ni-matrix. The effects of the particle size, the interfacial characteristic and the elastic property of the matrix on the morphological evolution of the Y2O3 have been analyzed and predicted using the phase-field method. The obtained results were compared with the experimental observations.

Phase-field simulation of mechano-chemical coupling

during precipitation in aluminum alloys

C. Schwarze, R. Darvishi Kamachali and I. Steinbach

Interdisciplinary Centre for Advanced Materials Simulation (ICAMS) Ruhr-University Bochum, 44801 Bochum, Germany

Email: christian.schwarze@rub.de

Aluminum alloys importance on the broad fields of engineering is unbroken and therefore the understanding of complex hardening processes like precipitation is fundamental for guaranteeing required mechanical properties and applications. As it is been demonstrated recently spinodal decomposition precipitation was assumed [1] and implemented into open source phase-field software, OpenPhase [2], using a recently developed dissipation model for interfaces [3,4]. Furthermore, mechano-chemical coupling [5] is taken into account which features new aspects of precipitation processes [6]. For evaluation, different interface homogenization methods for elastic properties were implemented into our software [7]. First precipitate growth simulations (metastable δ’) were done for a binary Al-Li alloy and will be extended to Al-Cu and Al-Li-Cu. Moreover, the growth behaviour and interaction of adjacent precipitates are analyzed and will be compared to experimental results. Forward-looking, other processes like coherency loss effects during phase transformation, the consideration of Onsager atomic mobilities and the influence of vacancies on the kinetic and structure of precipitates will be analyzed. [1] K.T. Kashyap, P.G. Koppad, Bull. Mater. Sci., 34 (2011) 1455.

[2] www.openphase.de

[3] I. Steinbach, L. Zhang, M. Plapp, Acta Mater., 60 (2012) 2702.

[4] I. Steinbach, Ann. Rev. Mater. Res., 43 (2013) 89.

[5] F.C. Larche, J.W. Cahn, Acta Metall., 30 (1982) 1835.

[6] R. Darvishi Kamachali, E. Borukhovich, O. Shchyglo, I. Steinbach, Philos. Mag. Letters, 93 (2013)

680.

[7] K. Ammar, B. Appolaire, G. Cailletaud, S. Forest, European Journal of Computational Mechanics,

18 (2009) 485.

3D real microstructure-based crystal plasticity FEM simulations

for a polycrystalline Ni deformed in tension

Yoon Suk Choi

School of Materials Science and Engineering, Pusan National University Email: choiys@pusan.ac.kr

A micro-tensile test system equipped with in-situ monitoring of the in-plane displacements of a surface and an electron backscattered diffraction based serial-sectioning technique were used to study the deformation (up to 2.4% axial plastic strain in tension) of a polycrystalline nickel micro-specimen. The experimental data includes the global engineering stress-engineering strain curve, the local mesoscopic in-plane displacement and strain fields, the three-dimensional microstructure of the micro-specimen reconstructed after the tensile test, and the kernel-average misorientation distribution. The crystal plasticity finite element method using elasto-viscoplastic constitutive formulations was used to simulate the global and local deformation responses of the micro-specimen. Three different boundary conditions were applied in simulation in order to study the effects of the lateral displacement (perpendicular to the loading direction) of the top and bottom faces of the specimen gage section. The simulation results were compared to the experimental results. The comparison between experiment and simulation results are discussed, based upon their implications for understanding the deformation of micro-specimens and the causes associated with uncertainties embedded in both experimental and numerical approaches. Also, the sensitivity of boundary conditions to near field and far field responses of the micro-specimen was systematically studied. Results show that the experimental methodology used in the present study allows for limited but meaningful comparisons to crystal plasticity finite element simulations of the micro-specimen under the small plastic deformation.

Phase-field study on the formation of lath martensite

Min-Kyu Cho1, Dong-Uk Kim1, Pil-Ryung Cha1, Moon-Gi Bae2, Seung-Hyun Hong2

1 School of Advanced Materials Engineering, Kookmin University 2 Metallic Materials Research Lab, Hyundai Kia Motors Namyang Institute

Email: cprdream@kookmin.ac.kr

Lath martensite is a characteristic structure in quenched steels with a low or negligible carbon content, such as plain low-carbon steels, low-carbon and low-alloy steels, maraging steels and interstitial free (IF) steels. The transformation of a parent austenite grain into lath martensite was characterized as a grain subdivision on different length scales. On a coarse scale, the austenite grain breaks down to several packets, each containing extended parallel blocks. Each block is further subdivided by laths, which are narrow units with a width in the sub-micrometer range. In this study, we analyze the origin of lath martensite formation by using volume-conserving multi-phase field model with micro-elasticity. Although various deformation mechanisms such as Bain, Kurdjumov–Sachs (KS), Nishiyama-Wassermann (NW) and Pitch mechanisms have been proposed for cubic to tetragonal martensitic transformation, Only Bain deformation mechanism has been considered in previous phase field models. In this study, we consider all the formation mechanisms including Bain, KS, NW and Pitch deformations and reveal a plausible origin for the multi-scale nature of lath martensite.

Phase-field simulation of Mg-Al alloy solidification

A. Monas1, O. Shchyglo1, S.J. Kim2, C.D. Yim2 and I. Steinbach1

1 Interdisciplinary Centre for Advanced Materials Simulation (ICAMS)

Ruhr-University Bochum, 44801 Bochum, Germany 2 Korea Institute of Materials Science

Email: alexander.monas@rub.de

We present the results of the microstructure evolution simulation during solidification of Mg-Al alloys. The full process of solidification is simulated using the free phase-field simulation library "OpenPhase". Consecutive nucleation and growth of primary alpha-Mg phase and secondary beta-Mg phase is simulated. The influence of the main processing control parameters such as cooling rate, nucleation density and aluminum concentration on the as cast Mg alloy microstructure has been investigated. The obtained simulation results closely match experimental observations.

Using a dual scale approach, a zoomed in region of the interdendritic eutectic melt was simulated. A system consisting of a melt channel between two primary alpha-Mg dendrites is cooled below the eutectic point. A rapid coverage of the alpha-Mg dendrites by beta-Mg phase promotes the formation of divorced eutectic solidification microstructure. It consists of a mixture of tertiary alpha- and secondary beta-Mg. The formation of such a divorced eutectic microstructure is an important factor for the improvement of the corrosion resistance of the cast alloy. The obtained simulation results show remarkable agreement with experimentally observed microstructures.

The proposed simulation approach can later be used for computational design of Mg-alloys with desirable corrosion properties.

Crystallization kinetics of nanoscale Ge2Sb2Te5 and its influence to

data retention statistics of phase-change memory

Yongwoo Kwon

Department of Materials Science and Engineering, Hongik University Email: ykwon722@hongik.ac.kr

Crystallization kinetics, the change of the crystalline fraction as a function of time, has been described by a deterministic Johnson-Mehl-Avrami-Kolmogorov (JMAK) equation that yields the well-known S-curve. However, the JMAK equation is derived for an infinitely large system and thus cannot be applied to nanoscale material systems with finite size such as phase-change memory (PCM), a promising next-generation nonvolatile memory (NVM). More specifically, the crystallization directly affects the data retention time of the PCM. There are two critical size effects in the viewpoint of the data retention, longer data retention time and larger statistical spread in smaller systems. In this talk, I will quantitatively analyze such size effects by means of numerical and analytical modeling. The first half of my talk will be about a phase-field model on nucleation and growth and its application to Ge2Sb2Te5 in phase-change memory with two cell schemes, confined and mushroom cells. The latter half will be about the analytical modeling for the size effects, a kind of modification of the JMAK equation to nanoscale systems.