서브메뉴
검색
Next-Generation Battery Modeling, Simulation, and Development
Next-Generation Battery Modeling, Simulation, and Development
상세정보
- 자료유형
- 학위논문 서양
- 최종처리일시
- 20260202105229
- ISBN
- 9798291567081
- DDC
- 620
- 저자명
- Gao, Tianhan.
- 서명/저자
- Next-Generation Battery Modeling, Simulation, and Development
- 발행사항
- [Sl] : University of Michigan, 2025
- 발행사항
- Ann Arbor : ProQuest Dissertations & Theses, 2025
- 형태사항
- 173 p
- 주기사항
- Source: Dissertations Abstracts International, Volume: 87-03, Section: B.
- 주기사항
- Advisor: Lu, Wei.
- 학위논문주기
- Thesis (Ph.D.)--University of Michigan, 2025.
- 초록/해제
- 요약To advance the development of next-generation lithium-ion and lithium-metal batteries, it is essential to comprehensively understand the aging and degradation, such as lithium dendrite nucleation and growth in lithium-metal electrodes, thermal degradation of liquid electrolytes, side reactions at electrode surfaces, and crack propagation at both the electrode and particle levels. Gaining insight into these phenomena is critical for guiding the design of novel battery architectures and optimizing operational protocols, ultimately enabling improvements in energy density, lifespan, and operational safety. In parallel, the development of fast, robust, and physics informed battery models is crucial, such models are not only fundamental for accurately capturing multi-physical interactions within batteries, but also for seamless integration into machine learning and optimization frameworks, thereby accelerating the innovation cycle for high energy-density and lifetime batteries. This dissertation primarily focuses on several critical aspects of battery research: the development of piezoelectric technology for suppressing dendrite formation in lithium metal batteries, the analysis of electrolyte thermal degradation, the formulation of novel reduced-order physics-based electrochemical models, and the proposal of an optimized thick-electrode microstructural design. These contributions aim to enhance battery performance, longevity, and computational efficiency, driving progress toward the next generation of high-performance energy storage systems.A primary challenge in the development of next-generation lithium-metal batteries is ensuring operational safety. Among the critical degradation mechanisms, lithium dendrite formation poses a significant threat. These dendrites can grow through the electrolyte and penetrate the separator, potentially leading to internal short circuits and thermal runaway. This safety risk represents a major obstacle to the commercialization of lithium-metal batteries, particularly under fast-charging conditions, where dendrite growth is often accelerated. We firstly show a piezoelectric mechanism that effectively suppresses dendrite growth, using a compliant piezoelectric film as a separator or coating. By proposing a theory that couples electrochemistry and piezoelectricity, we quantify the suppression effect and growth morphology. We find that the dendrite-suppression capability is over 5x106 stronger than the limit of mechanical blocking by any separators or solid state electrolytes. Surprisingly, the mechanism ensures depositing to a flat surface even if the initial substrate surface has significant protrusions, suggesting its robustness and effectiveness against manufacturing defects. We further develop a theory for piezoelectric thin film that couples the fields of electrochemistry, piezoelectricity and thin film mechanics. Such a fundamental framework is expected to help analyze various new phenomena and material innovation that involves electrochemistry and thin film piezo electricity. We also develop a theory for bulk porous piezoelectric medium integrating electrochemistry, piezoelectricity and mechanics. A piezoelectric over-potential is derived, which revealsafundamental relation to surface charge density, dielectric property of the medium, electrolyte concentration and diffusivity, and the reaction coefficient. The simulations show that piezoelectric medium suppresses electrodeposition on any protrusion, leading to a flat, dendrite free surface.We not only investigate the degradation of dendrite evolution for lithium-metal batteries, but also investigate the aging mechanism of a lithium-ion battery. Specifically, the electrolyte thermal decomposition during usage is one of the degradation mechanisms that can significantly influence the electrolyte ionic diffusivity and conductivity, which further significantly affects the power density and usable energy density. Understanding the degradation mechanism and its effect on ionic diffusivity is important for both battery design optimization to provide superior performance with a long cycle life and for better battery management during usage to extend the battery life. We quantitively predict the ionic diffusivity of key electrolytes and their degradation, including DMC-LiPF6, EMC-LiPF6 and DEC-LiPF6, with classical and ReaxFF molecular dynamics simulations. The effect of temperature, salt concentration and degree of thermal degradation on electrolyte ionic diffusivity are identified. DMC-LiPF6 shows the highest thermal stability, while DEC-LiPF6 shows the lowest thermal stability. Simulations show that the diffusion coefficients of cations and anions decrease with thermal degradation.Next, we develop and employ reliable and robust battery models capable of accurately predicting performance and degradation, which is essential to accelerate the iterative development of next-generation lithium-ion and lithium-metal batteries, thereby reducing reliance on extensive physical testing. These models also play a critical role when integrated into optimization frameworks for the design of advanced battery architecture, including novel structural and geometrical configurations. However, the most widely adopted physics-based models, such as the P2D model, require solving complex partial differential equations (PDEs), typically through finite element or finite volume methods. These approaches demand substantial computational resources and involve extensive numerical iterations, making large-scale or real time simulations impractical. Therefore, there is an urgent need to develop computational strategies that significantly accelerate simulation speed while preserving the predictive accuracy of physics-based battery models. We then develop physical-based, reduced-order electrochemical models that are much faster than the pseudo2D (P2D) model, while providing high accuracy even under the challenging conditions of high C-rate and strong polarization of lithium ion concentration and potential. In particular, an innovative weak form of equations are developed by using shape functions, which reduces the fully coupled electrochemical and transport equations to ordinary differential equations, and provides self-consistent solutions for the evolution of polynomial coefficients. Results show that the models, named as revised single-particle model (RSPM) and fast-calculating P2D model (FCP2D), give reliable prediction of battery operations, including under dynamic driving profiles. They can calculate battery parameters, such as terminal voltage, over-potential, interfacial current density, lithium-ion concentration distribution, and electrolyte potential distribution with a relative error less than2%. Applicable for moderately high C-rates (below 2.5 C), the RSPM is up to more than 33timesfaster than the P2D model. The FCP2D is applicable for high C-rates (above 2.5 C) and is about 8 times faster than the P2D model.Furthermore, we leverage our developed battery physical-based model with machine learning algorithm to optimize the battery micro-structure to promote cell performance, specifically for lithium-ion batteries with thick electrodes, which is highly effective in increasing the specific energy of a battery cell, but the associated increase in transport distance causes a major barrier for fast charging, which can further increase the mechanical degradation within the cell. We introduce a bio-inspired electrolyte channel design into thick electrodes to improve cell performance, especially under fast charging conditions, and reduce the electrode-level stress to reduce the mechanical degradation. Machine learning by deep artificial neural network (DNN) isdeveloped to relate the geometrical parameters of channels to the overall cell performance. Integrating machine learning with the Markov chain Monte Carlo gradient descent optimization, we demonstrate that the complicated multivariable channel geometry optimization problem can be efficiently solved.
- 일반주제명
- Engineering
- 일반주제명
- Energy
- 일반주제명
- Electrical engineering
- 일반주제명
- Mechanical engineering
- 기타저자
- University of Michigan Mechanical Engineering
- 기본자료저록
- Dissertations Abstracts International. 87-03B.
- 전자적 위치 및 접속
- 로그인 후 원문을 볼 수 있습니다.
MARC
008260126s2025 us c eng d■001000017359876
■00520260202105229
■006m o d
■007cr#unu||||||||
■020 ▼a9798291567081
■035 ▼a(MiAaPQ)AAI32271877
■035 ▼a(MiAaPQ)umichrackham006497
■040 ▼aMiAaPQ▼cMiAaPQ
■0820 ▼a620
■1001 ▼aGao, Tianhan.
■24510▼aNext-Generation Battery Modeling, Simulation, and Development
■260 ▼a[Sl]▼bUniversity of Michigan▼c2025
■260 1▼aAnn Arbor▼bProQuest Dissertations & Theses▼c2025
■300 ▼a173 p
■500 ▼aSource: Dissertations Abstracts International, Volume: 87-03, Section: B.
■500 ▼aAdvisor: Lu, Wei.
■5021 ▼aThesis (Ph.D.)--University of Michigan, 2025.
■520 ▼aTo advance the development of next-generation lithium-ion and lithium-metal batteries, it is essential to comprehensively understand the aging and degradation, such as lithium dendrite nucleation and growth in lithium-metal electrodes, thermal degradation of liquid electrolytes, side reactions at electrode surfaces, and crack propagation at both the electrode and particle levels. Gaining insight into these phenomena is critical for guiding the design of novel battery architectures and optimizing operational protocols, ultimately enabling improvements in energy density, lifespan, and operational safety. In parallel, the development of fast, robust, and physics informed battery models is crucial, such models are not only fundamental for accurately capturing multi-physical interactions within batteries, but also for seamless integration into machine learning and optimization frameworks, thereby accelerating the innovation cycle for high energy-density and lifetime batteries. This dissertation primarily focuses on several critical aspects of battery research: the development of piezoelectric technology for suppressing dendrite formation in lithium metal batteries, the analysis of electrolyte thermal degradation, the formulation of novel reduced-order physics-based electrochemical models, and the proposal of an optimized thick-electrode microstructural design. These contributions aim to enhance battery performance, longevity, and computational efficiency, driving progress toward the next generation of high-performance energy storage systems.A primary challenge in the development of next-generation lithium-metal batteries is ensuring operational safety. Among the critical degradation mechanisms, lithium dendrite formation poses a significant threat. These dendrites can grow through the electrolyte and penetrate the separator, potentially leading to internal short circuits and thermal runaway. This safety risk represents a major obstacle to the commercialization of lithium-metal batteries, particularly under fast-charging conditions, where dendrite growth is often accelerated. We firstly show a piezoelectric mechanism that effectively suppresses dendrite growth, using a compliant piezoelectric film as a separator or coating. By proposing a theory that couples electrochemistry and piezoelectricity, we quantify the suppression effect and growth morphology. We find that the dendrite-suppression capability is over 5x106 stronger than the limit of mechanical blocking by any separators or solid state electrolytes. Surprisingly, the mechanism ensures depositing to a flat surface even if the initial substrate surface has significant protrusions, suggesting its robustness and effectiveness against manufacturing defects. We further develop a theory for piezoelectric thin film that couples the fields of electrochemistry, piezoelectricity and thin film mechanics. Such a fundamental framework is expected to help analyze various new phenomena and material innovation that involves electrochemistry and thin film piezo electricity. We also develop a theory for bulk porous piezoelectric medium integrating electrochemistry, piezoelectricity and mechanics. A piezoelectric over-potential is derived, which revealsafundamental relation to surface charge density, dielectric property of the medium, electrolyte concentration and diffusivity, and the reaction coefficient. The simulations show that piezoelectric medium suppresses electrodeposition on any protrusion, leading to a flat, dendrite free surface.We not only investigate the degradation of dendrite evolution for lithium-metal batteries, but also investigate the aging mechanism of a lithium-ion battery. Specifically, the electrolyte thermal decomposition during usage is one of the degradation mechanisms that can significantly influence the electrolyte ionic diffusivity and conductivity, which further significantly affects the power density and usable energy density. Understanding the degradation mechanism and its effect on ionic diffusivity is important for both battery design optimization to provide superior performance with a long cycle life and for better battery management during usage to extend the battery life. We quantitively predict the ionic diffusivity of key electrolytes and their degradation, including DMC-LiPF6, EMC-LiPF6 and DEC-LiPF6, with classical and ReaxFF molecular dynamics simulations. The effect of temperature, salt concentration and degree of thermal degradation on electrolyte ionic diffusivity are identified. DMC-LiPF6 shows the highest thermal stability, while DEC-LiPF6 shows the lowest thermal stability. Simulations show that the diffusion coefficients of cations and anions decrease with thermal degradation.Next, we develop and employ reliable and robust battery models capable of accurately predicting performance and degradation, which is essential to accelerate the iterative development of next-generation lithium-ion and lithium-metal batteries, thereby reducing reliance on extensive physical testing. These models also play a critical role when integrated into optimization frameworks for the design of advanced battery architecture, including novel structural and geometrical configurations. However, the most widely adopted physics-based models, such as the P2D model, require solving complex partial differential equations (PDEs), typically through finite element or finite volume methods. These approaches demand substantial computational resources and involve extensive numerical iterations, making large-scale or real time simulations impractical. Therefore, there is an urgent need to develop computational strategies that significantly accelerate simulation speed while preserving the predictive accuracy of physics-based battery models. We then develop physical-based, reduced-order electrochemical models that are much faster than the pseudo2D (P2D) model, while providing high accuracy even under the challenging conditions of high C-rate and strong polarization of lithium ion concentration and potential. In particular, an innovative weak form of equations are developed by using shape functions, which reduces the fully coupled electrochemical and transport equations to ordinary differential equations, and provides self-consistent solutions for the evolution of polynomial coefficients. Results show that the models, named as revised single-particle model (RSPM) and fast-calculating P2D model (FCP2D), give reliable prediction of battery operations, including under dynamic driving profiles. They can calculate battery parameters, such as terminal voltage, over-potential, interfacial current density, lithium-ion concentration distribution, and electrolyte potential distribution with a relative error less than2%. Applicable for moderately high C-rates (below 2.5 C), the RSPM is up to more than 33timesfaster than the P2D model. The FCP2D is applicable for high C-rates (above 2.5 C) and is about 8 times faster than the P2D model.Furthermore, we leverage our developed battery physical-based model with machine learning algorithm to optimize the battery micro-structure to promote cell performance, specifically for lithium-ion batteries with thick electrodes, which is highly effective in increasing the specific energy of a battery cell, but the associated increase in transport distance causes a major barrier for fast charging, which can further increase the mechanical degradation within the cell. We introduce a bio-inspired electrolyte channel design into thick electrodes to improve cell performance, especially under fast charging conditions, and reduce the electrode-level stress to reduce the mechanical degradation. Machine learning by deep artificial neural network (DNN) isdeveloped to relate the geometrical parameters of channels to the overall cell performance. Integrating machine learning with the Markov chain Monte Carlo gradient descent optimization, we demonstrate that the complicated multivariable channel geometry optimization problem can be efficiently solved.
■590 ▼aSchool code: 0127.
■650 4▼aEngineering
■650 4▼aEnergy
■650 4▼aElectrical engineering
■650 4▼aMechanical engineering
■653 ▼aLithium-ion battery
■653 ▼aLithium-metal battery
■653 ▼aDendrite suppression
■653 ▼aElectrolyte thermal degradation
■653 ▼aFast-calculation model
■690 ▼a0537
■690 ▼a0791
■690 ▼a0544
■690 ▼a0548
■71020▼aUniversity of Michigan▼bMechanical 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=T17359876▼nKERIS▼z이 자료의 원문은 한국교육학술정보원에서 제공합니다.


