서브메뉴
검색
Investigation of the Capability of Molecular Dynamics Simulation in Interface Evolution: Uncertainties and Applications
Investigation of the Capability of Molecular Dynamics Simulation in Interface Evolution: Uncertainties and Applications
상세정보
- 자료유형
- 학위논문 서양
- 최종처리일시
- 20260202105223
- ISBN
- 9798291566466
- DDC
- 620.11
- 저자명
- Lyu, Meizhong.
- 서명/저자
- Investigation of the Capability of Molecular Dynamics Simulation in Interface Evolution: Uncertainties and Applications
- 발행사항
- [Sl] : University of Michigan, 2025
- 발행사항
- Ann Arbor : ProQuest Dissertations & Theses, 2025
- 형태사항
- 135 p
- 주기사항
- Source: Dissertations Abstracts International, Volume: 87-03, Section: B.
- 주기사항
- Advisor: Holm, Elizabeth Ann.
- 학위논문주기
- Thesis (Ph.D.)--University of Michigan, 2025.
- 초록/해제
- 요약Molecular dynamics (MD) simulations serve as an indispensable complement to experimental research in materials science, offering unique capabilities to verify experimental observations, access to experimentally inaccessible quantities, and provide precise control over system parameters. This dissertation demonstrates the multifaceted applications of MD simulations across three distinct but interconnected research domains.The first study investigated uncertainty quantification and propagation in ensemble MD simulations of nickel grain growth. Aleatoric uncertainty was introduced through randomly assigned initial atomic velocities to individual atoms. This uncertainty propagated from the atomic scale to the microstructural scale during simulation evolution, and resulting in the differentiation of individual grain evolution trajectories across different simulation replicas. Visualization analysis of these parallel computational results revealed atomic-level velocity effects on microstructural evolution that are typically inaccessible through conventional experimental characterization methods. This study showed that even small differences in initial atomic velocities can lead to statistically significant divergences in microstructural evolution. It highlighted the necessity of ensemble simulations to accurately capture microstructural behavior.The second project developed a novel methodology for directly converting voxel-based experimental microstructural data into atomic-scale initial configurations for MD simulations. This approach preserves microstructural fidelity while accommodating different lattice types and multiple scaling requirements. Unlike experimental samples that inevitably contain impurities, MD simulations provide an idealized environment of pure materials without impurity interference. Comparison with experimental nickel polycrystal data demonstrated the method's capability to capture characteristic grain growth behavior. This chapter established a bridge between experimental and atomic-resolution simulations, successfully reproducing grain boundary characteristics observed in real materials. It further confirmed that the absence of a strong correlation between grain boundary migration velocity and curvature in polycrystals is unrelated to impurities, but instead arises from network-level effects.The third investigation addresses the challenging problem of calculating solid-liquid interfacial energy at non-melting temperatures using the energy conserving orientational force. While interfacial energy is crucial for solidification processes, it is experimentally difficult to measure, traditional MD approaches are limited to melting point calculations. This work employs simultaneous equation solving to determine both interfacial energy and mobility across supercooling temperature ranges. The results revealed a non-monotonic temperature dependence of interfacial energy and a positive temperature dependence of interfacial mobility. The application of synthetic driving forces enabled MD simulations under previously inaccessible conditions and provided new insights into the temperature dependence of solid-liquid interfacial properties.These studies illustrate the multifaceted capability of MD simulations in connecting atomic-scale phenomena to macroscopic properties, delivering mechanistic insights, and achieving systematic parameter control beyond the reach of conventional experimental approaches.
- 일반주제명
- Materials science
- 일반주제명
- Molecular physics
- 일반주제명
- Applied physics
- 일반주제명
- Physical chemistry
- 일반주제명
- Thermodynamics
- 키워드
- Grain boundary
- 키워드
- Equilibrium
- 기타저자
- University of Michigan Materials Science and Engineering
- 기본자료저록
- Dissertations Abstracts International. 87-03B.
- 전자적 위치 및 접속
- 로그인 후 원문을 볼 수 있습니다.
MARC
008260126s2025 us c eng d■001000017359845
■00520260202105223
■006m o d
■007cr#unu||||||||
■020 ▼a9798291566466
■035 ▼a(MiAaPQ)AAI32271827
■035 ▼a(MiAaPQ)umichrackham006387
■040 ▼aMiAaPQ▼cMiAaPQ
■0820 ▼a620.11
■1001 ▼aLyu, Meizhong.
■24510▼aInvestigation of the Capability of Molecular Dynamics Simulation in Interface Evolution: Uncertainties and Applications
■260 ▼a[Sl]▼bUniversity of Michigan▼c2025
■260 1▼aAnn Arbor▼bProQuest Dissertations & Theses▼c2025
■300 ▼a135 p
■500 ▼aSource: Dissertations Abstracts International, Volume: 87-03, Section: B.
■500 ▼aAdvisor: Holm, Elizabeth Ann.
■5021 ▼aThesis (Ph.D.)--University of Michigan, 2025.
■520 ▼aMolecular dynamics (MD) simulations serve as an indispensable complement to experimental research in materials science, offering unique capabilities to verify experimental observations, access to experimentally inaccessible quantities, and provide precise control over system parameters. This dissertation demonstrates the multifaceted applications of MD simulations across three distinct but interconnected research domains.The first study investigated uncertainty quantification and propagation in ensemble MD simulations of nickel grain growth. Aleatoric uncertainty was introduced through randomly assigned initial atomic velocities to individual atoms. This uncertainty propagated from the atomic scale to the microstructural scale during simulation evolution, and resulting in the differentiation of individual grain evolution trajectories across different simulation replicas. Visualization analysis of these parallel computational results revealed atomic-level velocity effects on microstructural evolution that are typically inaccessible through conventional experimental characterization methods. This study showed that even small differences in initial atomic velocities can lead to statistically significant divergences in microstructural evolution. It highlighted the necessity of ensemble simulations to accurately capture microstructural behavior.The second project developed a novel methodology for directly converting voxel-based experimental microstructural data into atomic-scale initial configurations for MD simulations. This approach preserves microstructural fidelity while accommodating different lattice types and multiple scaling requirements. Unlike experimental samples that inevitably contain impurities, MD simulations provide an idealized environment of pure materials without impurity interference. Comparison with experimental nickel polycrystal data demonstrated the method's capability to capture characteristic grain growth behavior. This chapter established a bridge between experimental and atomic-resolution simulations, successfully reproducing grain boundary characteristics observed in real materials. It further confirmed that the absence of a strong correlation between grain boundary migration velocity and curvature in polycrystals is unrelated to impurities, but instead arises from network-level effects.The third investigation addresses the challenging problem of calculating solid-liquid interfacial energy at non-melting temperatures using the energy conserving orientational force. While interfacial energy is crucial for solidification processes, it is experimentally difficult to measure, traditional MD approaches are limited to melting point calculations. This work employs simultaneous equation solving to determine both interfacial energy and mobility across supercooling temperature ranges. The results revealed a non-monotonic temperature dependence of interfacial energy and a positive temperature dependence of interfacial mobility. The application of synthetic driving forces enabled MD simulations under previously inaccessible conditions and provided new insights into the temperature dependence of solid-liquid interfacial properties.These studies illustrate the multifaceted capability of MD simulations in connecting atomic-scale phenomena to macroscopic properties, delivering mechanistic insights, and achieving systematic parameter control beyond the reach of conventional experimental approaches.
■590 ▼aSchool code: 0127.
■650 4▼aMaterials science
■650 4▼aMolecular physics
■650 4▼aApplied physics
■650 4▼aPhysical chemistry
■650 4▼aThermodynamics
■653 ▼aComputational material science
■653 ▼aMolecular dynamics simulation
■653 ▼aInterface physics
■653 ▼aGrain boundary
■653 ▼aEquilibrium
■690 ▼a0794
■690 ▼a0215
■690 ▼a0348
■690 ▼a0609
■690 ▼a0494
■71020▼aUniversity of Michigan▼bMaterials Science and Engineering.
■7730 ▼tDissertations Abstracts International▼g87-03B.
■790 ▼a0127
■791 ▼aPh.D.
■792 ▼a2025
■793 ▼aEnglish
■85640▼uhttp://www.riss.kr/pdu/ddodLink.do?id=T17359845▼nKERIS▼z이 자료의 원문은 한국교육학술정보원에서 제공합니다.


