Li-ion battery health diagnosis based on electrochemical model

김정수 포항공과대학교 일반대학원 2022 국내박사

Secondary batteries are widely leveraged in the rapidly growing electric vehicle (EV) and energy-storage system (ESS) markets. Among the battery types, Li-ion batteries are extensively used, owing to their advantages of high energy and power densities, low self-discharge rate, and lack of memory effect. However, they tend to be susceptible to thermal runaways caused by undesirable operations, because their thermal stability is inherently lower than other batteries. Moreover, as Li-ion batteries become more widespread, the number of associated accidents also increases significantly. To use such delicate batteries safely and efficiently, it is essential to accurately estimate their health states. This thesis deals with several methods for diagnosing the health states of Li-ion batteries from various angles based on an electrochemical model, especially a pseudo-2-dimensional (P2D) model, which precisely expresses the dynamics of a Li-ion battery. First, a data-driven state-of-health (SOH) estimation scheme for Li-ion batteries with reference performance test (RPT)-reduced experimental data is proposed. In this work, it is proved that the capacity of a Li-ion battery can be effectively estimated using only scarce data acquired through minimal number of RPT performing, unlike most of existing studies have employed abundant reference data. Second, some parameters of the P2D model for effective and practical diagnosis of Li-ion batteries are selected as aging parameters and identified. In this study, some major parameters of the P2D model were identified through a genetic algorithm (GA). Based on the results, the parameters suitable for diagnosing the battery health state were selected as aging parameters, and their physical meanings were considered. Lastly, a novel P2D model parameter identification method, genetic algorithm and neural network cooperative optimization (GANCO), is proposed. Many existing studies that employ meta-heuristic methods to identify model parameters did not efficiently utilize data generated during optimization. In this work, an 1-dimensional convolutional neural network (1D CNN) periodically trains the sample data obtained from previous iterations to learn thedynamics between the known input current and the corresponding simulated voltage. Then the trained network recommends highly probable parameter candidates to the GA. In such manner, an 1D CNN and a GA cooperate and complement each other in GANCO.

Ion conduction and ionic speciation of Li-ion battery (LIB) electrolytes

Hyejin Lee DGIST 2021 국내박사

본 논문은 리튬이온 배터리의 성능 개선을 위해 전해액의 용액 구조를 분광학적으로 분석하여 리튬이온의 전달 기작을 분석하였습니다. 최근 리튬 이온 배터리에서 부각되고 있는 필수 요소인 급속충전, 고출력 배터리 기술은 주로 배터리 전해질의 리튬이온 전달 현상과 중요하게 연관되어 있습니다. 현재의 출력 특성 및 충전 속도의 한계를 극복하기 위해서 전해질에서 원활한 리튬 이온 전달이 중요한 역할을 담당합니다. 하지만 리튬이온과 음이온, 용매와의 상호작용을 분석하고 이에 따른 배터리 성능에 대한 연구는 아직 초기 단계입니다. 용매의 특성 및 염의 종류 및 농도는 전해질의 용액 구조 -양이온 주변의 배위- 를 결정하며, 리튬 이온의 전달, 전극과 전해질 계면 거동에 영향을 줍니다. 따라서, 본 연구에서는 분광학적 방법을 이용해 전해질의 용액 구조와 리튬 이온의 전달 기작을 분석하였습니다. 첫번째로, 전해질로 자주 사용되는 카보네이트 계 전해질 중, DMC기반 용액에서 특별히 높게 나타나는 이온전도도를 용액 구조 분석 및 DFT 계산을 이용하여 분석하였습니다. 다이메틸 카보네이트(DMC)와 다이에틸 카보네이트 (DEC)는 비슷한 물리 화학적 성질에도 불구하고, 그들의 리튬 염 용액의 이온 전도도는 큰 차이를 보이며, 그 차이의 원인은 설명된 적이 없었습니다. 본 논문에서는 라만 분광법, 유전 완화 분광법(DRS) 및 Pulsed Field Gradient NMR (PFG-NMR)를 사용하여 0.1-3.0 M LiPF6의 에틸렌 카보네이트(EC)와 DMC 및 DEC 혼합 전해질의 용액 구조와 전도 거동의 상관 관계를 연구하였습니다. 우리는 극성 DMC 컨포머가 DEC 컨포머보다 쉽게 형성되어 DMC 전해질에서 높은 리튬 염 해리 및 높은 이온 전도도를 이끌어내는 것을 확인하였습니다. 또한, 컨포머의 형성은 Li 염 농도가 증가하고, EC가 일정 분율 섞였을 때도 같은 효과가 있다는 것을 확인하였습니다. 둘째, 다이메틸 설폭사이드 (DMSO)기반의 용액에서 나타나는 이온전도도의 특징을 라만 분광법, DRS 및 PFG-NMR을 이용하여 분석하였습니다. 일반적으로 염의 해리도가 높은 LiPF6 염은 LiBF4 보다 높은 이온전도도를 나타내지만, DMSO에서는 반대 순서의 이온전도도 경향을 나타냅니다. 반면, 해리성이 강한 DMSO 용액에서는 염이 동등한 수준으로 높게 해리되어, 두 염의 해리도 차이는 근소했습니다. 이는 DMSO 용매가 가진 높은 유전율과 더불어 높은 donor number 가 염을 크게 해리 시키기 때문입니다. 따라서, DMSO용액에서는 염의 해리도 보다 점도가 이온전도도를 결정하는 더 큰 요소로 작용하게 된 것입니다. 셋째, 니트릴(Nitrile) 기반 용액에서의 리튬 호핑현상을 유도하여, 고출력 및 급속충전을 위한 기능성 전해질을 개발하였습니다. 보통의 리튬이온전달 기작은 vehicular mode를 통해 전달됩니다. 이것은 리튬이 용매화 된 상태로 이동하는 것으로, 리튬이 리간드 없이 호핑하여 전달되는 hopping mode 보다 전지의 출력 특성이 좋지 않습니다. 본 연구에서는 monodentate 용매에서 리튬 간 거리가 가깝도록 용액 구조를 조절 하여 저농도부터 호핑 기작을 유도하였습니다. 지금까지 호핑 기작은 bidentate 용매를 고용했을 때 고농도부터 발현된다는 최근 보고와는 다르게, 용액 구조를 조절하여 monodentate 용매를 고용하여 저농도부터 호핑 현상이 나타나는 것을 규명하였습니다. A highly conductive electrolyte is an essential requirement for state-of-the-art rechargeable batteries. However, the delicate interplay between ionic conduction and the microscopic solution structure of battery electrolytes remains elusive. For this, it is imperative to understand the ion–ion and ion–solvent interactions governing the ion transport behavior of the electrolyte solutions to enhance the performance limit of electro-chemical devices. To meet the expectation of the advanced battery systems, it is mandatory to develop in-novative functional electrolytes ensuring stable, fast operation of LIBs. The rate capability of LIBs heavily depends on the mass transport rate in the electrolytes, which is largely dependent on ion-ion and ion-solvent interactions. Nevertheless, our understanding of the solution structure and its relation to ion transport behav-ior still remains at a premature stage. Therefore, solution structures and their ion conduction behaviors have been studied in this work. Firstly, the most widely used carbonate-based electrolytes were investigated. Despite similar physico-chemical properties of dimethyl carbonate (DMC) and diethyl carbonate (DEC), the ionic conductivity of their Li-salt solutions displays large disparity, of which origin remains elusive. We herein investigate the corre-lation of the conduction behavior with the solution structure of DMC and DEC electrolytes and their ethylene carbonate (EC) mixtures of 0.1–3.0 M LiPF6 using Raman spectroscopy, dielectric relaxation spectroscopy (DRS), and pulsed-field gradient NMR (PFG-NMR) spectroscopy. We reveal that polar DMC conformer is formed more readily than DEC conformer, leading to enhanced Li-salt dissociation and thus high ionic con-ductivity in DMC solutions. It is also confirmed that the conformational isomerism is facilitated the most in DMC/EC electrolytes. Secondly, the ionic speciation and conduction behavior in dimethyl sulfoxide (DMSO) electrolytes of 0.1–2.0 M LiPF6 and LiBF4 were examined using Raman spectroscopy, DRS, and PFG-NMR spectroscopy. This study elucidated the impact of solvent characteristics on ionic conductivity. Notably, LiPF6-DMSO, which had the highest dissociation in DMSO solvent, showed the lowest conductivity, but LiBF4-DMSO exhib-ited higher ionic conductivity with low salt dissociation. A DMSO, a solvent with a large donor number (DN), ion motilities become a dominant factor in ion conductivity than the salt dissociation. In addition, in most of the cases, lithium ions were transported by the vehicular mode, only except for the highly concentrated LiBF4-DMSO. These results provide insight into the solution structure and the interaction of the salt/solvent. Lastly, hopping conduction was intentionally induced by engineering the solution structure. High power performance becomes important for LIBs, so the fast Li-ion transport in electrolytes is in urgent need. To date, the ionic conduction in non-aqueous electrolytes has been described by the vehicle-type migration mechanism. However, recent studies suggest that ion-hopping can emerge in highly concentrated solutions. The adequate solvents having multiple coordinating sites, which establishes the so-called solvent-bridge struc-ture, have been one of the necessities for hopping conduction. However, this study reports the presence of Li-ion hopping conduction even in dilute solutions with monodentate acetonitrile (AN). We investigate the con-duction behavior and ionic speciation of AN solutions of 0.1 – 6.0 M lithium bis(fluorosulfonyl) imide (LiFSI) and lithium trifluoroacetate (LiTFA), representative dissociative and associative salts, using PFG-NMR, Raman spectroscopy, and DRS. PFG-NMR measurements show that the self-diffusion coefficient of Li-ion is faster than that of the anion in LiTFA-AN, whereas the opposite trend is observed in LiFSI-AN. Ra-man and DRS analysis elucidates that the faster Li-ion diffusion in LiTFA-AN is attributed to the Li-ion hop-ping conduction. Li-ion hop-ping is enabled by the unique anion-bridge structure driven by the highly associa-tive nature of LiTFA. The implication of the hopping conduction is demonstrated for LiFePO4 symmetric cells. The 4 M LiTFA-AN exhibits better rate performance than the 1 M LiTFA-AN despite lower ionic con-ductivity and high viscosity. This study challenges the conventional wisdom that associative Li-salts are un-suitable for battery electrolytes, and encourages to revisit numerous Li-salts unexploited so far due to their associative nature. Keywords: Li-ion battery (LIB) electrolyte, Solution structure, Ionic speciation, Ionic conductivity, conduc-tion mechanism

수명열화와 연동하는 step 충전을 통한 mobile 기기용 Li-Ion battery의 life cycle 개선방안

김성훈 高麗大學校 大學院 2012 국내석사

This paper is a study of the Li-Ion Battery SOH(State of Health) Major acceleration Factor and How to extend Li-Ion Smart battery Lifecycle Characteristic in View of Electrical Power Engineering. Li-Ion battery is Popular and Promising Energy source especially to Mobile IT Solution and EV/HEV etc. but Battery is a dynamic and nonlinear system with lots of chemical reactions. So it is a important that the internal characteristics and circuit parameters are changed according to the State of Health(SOH) and the Various operating Conditions. Therefore, without the specific modeling result of battery, it is difficult to use this smart & clean Energy source by effectively. The usefulness and improvement of the proposed step Charge algorithms are accurately verified with experiments on Li-Ion battery Packs. This work can be contributed to the improvement of Li-Ion Battery Life Cycle Characteristic and suspected that give more Highly values to a lot of products that use a Li-Ion Battery Solutions.

Unveiling Nanoscale Electrochemical Redox Pathways in Li-Ion Battery Cathodes : Investigation through Real-Time X-ray Imaging and Computational Simulation

구본호 서울대학교 대학원 2024 국내박사

요약(국문초록) 이차전지는 전기 에너지와 화학 에너지를 가역적으로 변환하는 대표적인 전기화학 장치이다. 그중 특히 리튬이온 배터리는 높은 에너지 밀도와 변환 가역성을 가져 모빌리티 시대를 열 수 있는 발판을 마련하였다.* 이 뛰어난 전기화학적 가역성의 근본적인 원리는 활물질 내 나노 스케일에서의 리튬 전달과 이에 따른 국소적인 산화·환원을 기반으로 하고 있다. 예를 들어, 충·방전 시 발생하는 전기화학적 비가역성의 증가(예: 높은 방전 속도에서 용량 감소 또는 배터리 열화로 인한 전압 강하)는 많은 경우 활물질의 국소적인 리튬 이온/전자 표면전달 반응 또는 리튬 확산 반응의 저항 증가로부터 기인한다. 따라서, 배터리의 핵심적 작동 원리를 이해하고 성능을 더욱 향상시키기 위해서는 리튬 삽입 및 이와 관련된 산화·환원 반응의 나노 스케일 경로에 대한 심층적인 이해가 필수적이다. 그러나 배터리 내 나노 스케일에서 산화·환원 반응에 관한 연구는 그 중요성에도 불구하고 현재까지 매우 제한적으로 이루어져 왔다. 이는 근본적으로 경원소 이온인 리튬 이온을 직접 관측하고 충·방전 중 나노미터 스케일에서의 실시간 반응을 직접 확인하는 것이 실험적으로 매우 어렵기 때문이며, 배터리를 이루고 있는 구성요소들의 내재적 불균일성으로 인해 국소적인 반응에 대한 분석이 더욱 제한되기 때문이다. 이러한 측면에서 본 연구자는 배터리 단일 입자 내 국소적인 리튬 이온 전달 및 산화·환원 반응을 실시간 이미징하기 위한 플랫폼을 구축하였고, 이를 대표적인 양극소재(인산철, 과리튬 망간 층상계 물질)에 적용하여 연구를 수행하였다. 또한, 각 물질에 대한 나노 스케일 반응 관측 결과와 전기 화학적 충·방전 반응의 상관관계에 대해 심층적으로 분석하였다. 1장에서는 리튬 이온 배터리 시스템에서 이온의 삽입/탈리 반응경로 및 이 반응에서 (속도 결정 단계)-(전기화학적 과전압) 관계에 대해 간략히 소개한다. 또한 이 연구에서 주로 사용된 관측방법인 실시간 주사 투과형 X-선 현미경법 (scanning transmission X-ray microscopy, STXM)과 관측된 실시간 반응 정보를 해석하기 위해 사용된 상-장(phase-field) 모델 및 밀도 범함수 이론(density functional theory, DFT) 계산에 관해 간단히 소개한다. 2장에서는 상 분리 전극에서 고속 충전에 비해 고속 방전에서 급격하게 증가하는 불분명한 전기화학적 과전압의 근본적인 원인에 대해 규명하였다. 이 연구에서는 넓은 범위의 충·방전 속도(0.07~7C)에서 [100]방향 LixFePO4(LFP) 판상형 단일입자(~1 µm 크기)에 대한 실시간 STXM 실험을 수행하였다. 관측 결과, 전극에 가해준 전류의 방향과 크기에 따라 입자 내 나노 스케일 단위의 표면 및 내부 상(phase) 형성이 매우 다르게 변한다는 것을 관측하였고, 이를 통해 기존 연구에서 제시되었던 여러 LFP 상 변화 모델들을 성공적으로 통합할 수 있었다. 더욱 중요한 점은, 이 실험에서 고속 방전과 충전이 활물질 표면에 각각 동역학적으로 과리튬(Li-rich) 및 리튬결핍(Li-poor) 영역을 형성한다는 사실을 명확하게 관측하였다. 이 과리튬/ 리튬결핍 영역은 리튬 삽입/탈리 반응에 대해 각각 음성/양성 피드백 반응을 만들어 내기 때문에 과전압을 증가/감소시키는 원인이 된다. 따라서 이 연구를 통해 인산철 배터리의 고속 방전과 충전 사이의 비대칭성적인 전기화학적 과전압 및 율속 특성의 원인을 규명할 수 있었다. 3장에서는 LFP 입자 내 균열 형성 여부에 따른 실시간 리튬 경로를 관측하였으며, 이를 통해 나노 스케일에서의 리튬 전달 경로가 어떻게 국소적인 응력 및 균열 형성에 기여할 수 있는지 규명하였다. 충·방전 중 활물질 내 리튬 전달이 고르지 않으면, 상 분포가 불균일해져 내부 응력이 집중되고 이에 쉽게 균열이 발생할 수 있다. 그러나 국소적인 리튬 농도 측정의 실험적 어려움으로 인해, 동역학적 상황의 리튬 분포와 균열 형성 사이의 관계는 제한적으로 이해되고 있었다. 이 연구에서는 균열이 존재하는/존재하지 않는 [100]방향 LFP 입자에 대한 실시간 STXM 실험을 수행하였다. 그 결과 충·방전 중 활물질 내 균열이 리튬의 삽입 탈리 반응을 기존의 (010) 반응 표면이 아닌, 균열로 인해 새롭게 형성된 표면에서 먼저 유도하는 것을 명확하게 관측하였다. 또한, 충전 과정에서 리튬부족 상이 균열이 있는 입자 중심부 영역에 우선적으로 생성되면서 동시에 실시간으로 균열이 벌어지는 현상을 직접 관측하였다. 상-장 시뮬레이션을 통해 입자 중심부에 형성된 리튬부족 상이 균열 부분에 매우 큰 인장력을 생성하여 균열을 벌어지게 했음을 확인하였으며, 이 인장력이 잠재적으로 균열을 더욱 크게 만들 수 있는 원인이 될 수 있음을 뒷받침하였다. 나아가 전기화학적 사이클링과 전자현미경을 통해 균열을 완화할 수 있는 충·방전 사이클 프로토콜을 제안하였다. 4장에서는 과리튬 망간 층상계(Li1.1Ni0.23Mn0.44Co0.23O2, LMR-NMC) 전극에서 전기 화학적 충·방전 사이클에 따라 보여지는 방전 전압의 강하(fading) 현상의 원인에 대해 규명하였다. LMR-NMC 전극에서 보여지는 이러한 방전 전압 강하는 종종 활물질 내 전이 금속 이동(migration) 또는 산소 발생(loss)으로 설명된다. 그러나 이러한 물질 내 구조변화는 충전에 비해 유독 방전 과정에서 두드러지게 보이는 비대칭적인 전압 강하를 설명하는 데 한계를 가진다. 이 연구에서는 열화된 LMR-NMC 단일 입자에 대해 실시간 가열 STXM실험(60°C)을 수행하여 시간에 따른 원소 별 산화수 변화를 측정하였고, 그 결과 입자 내 환원된 Ni에서 산화된 O로의 나노 스케일 금속-리간드 전하 전달(metal-to-ligand charge transfer, MLCT) 현상을 명확하게 이미징하였다. (즉, 가열 후 Ni은 산화, O은 환원) 이 결과는 열화된 LMR-NMC 물질 방전 과정에서, 물질 내 느려진 MLCT로 인해 Ni가 산소를 대신해 먼저 환원되었다는 것을 말해주며, 이것이 관측한 방전전압 강하에 크게 기여한다는 사실을 입증하였다. 또한, 추가 실험 및 DFT 계산을 통해, 열화된 LMR-NMC 내에 형성된 산소 동공(oxygen vacancy)이 Ni와 O 사이의 MLCT를 방해하여 방전전압 강하를 만들어 낼 수 있는 가능성을 제시하였다. 따라서, 이 연구를 통해 LMR-NMC 전극에서 관측된 비대칭 방전 전압 강하의 원인을 규명하고 이에 대한 원자 수준의 이해를 제시할 수 있었다. 이 연구에서는 에너지 저장 물질 내 나노 스케일 화학 정보를 실시간 이미징 할 수 있는 플랫폼을 성공적으로 개발했으며, 이를 통해 기존 연구에서는 관찰할 수 없었던 불균일한 리튬 전달과 이로 인한 국소적인 산화·환원 반응을 관측해낼 수 있었다. 나아가 충전-방전 과정에서 이러한 나노 스케일에서의 산화·환원 현상과 실제 배터리 시스템의 전기화학 간의 중요한 상관관계를 확인할 수 있었다. 본 연구자는 상-장 시뮬레이션 및 DFT 계산을 수행하여 관측된 동역학적 산화·환원에 대해 포괄적 및 이론적 설명을 제시할 수 있었다. 이 연구에서 소개한 바와 같이, 나노 스케일에서의 이온 삽입 반응에 대한 지속적이고 고도화된 분석은 에너지 저장에 대한 우리들의 근본적인 이해를 심화시킬 것이며, 리튬이온 배터리를 넘어 차세대 에너지 저장에 대한 새로운 패러다임의 발견을 가져올 것이라 기대한다. 주요어: 리튬이온 배터리, 양극 활물질, 전기화학적 반응, 실시간 X-선 흡광 이미징, 나노스케일 산화환원·반응, 컴퓨터 시뮬레이션 학 번: 2018-21067 Abstract A rechargeable, or so-called secondary, battery is an energy storage device that operates on the principle of reversible conversion between electrical and chemical energy. The Li-ion battery, in particular, excels in energy density and reversibility, making a significant contribution to the mobility era by enabling electrical devices with long lifetimes.* The origin of the excellent electrochemical reversibility is traced back to nanoscale Li transport and the corresponding redox reactions within the active materials. It follows that limitations of Li-ion batteries, such as reduced capacity at high discharge rates or voltage fading during battery cycling, are predominantly determined by Li-ion/electron transfer reactions at the material surfaces or Li diffusion within the host structures. An in-depth understanding of the Li insertion pathway and the associated redox reactions is therefore vital for advancing the kinetics and efficiency in Li-ion batteries and further enhance their performance. Despite their significance, exploration of these nanoscale phenomena in Li-ion batteries has been hindered by technological challenges in directly observing Li-ions and their transport, further exacerbated by inherent heterogeneities within the electrodes. In this regard, coworkers and I have developed a platform for real-time imaging of local Li transports and the corresponding redox reactions within single particles during battery cycling. This platform was applied to widely used cathode materials, including lithium iron phosphate and Li-Mn-rich layered oxide. Additionally, a thorough analysis was carried out on the correlation between the observed nanoscale redox reactions for each cathode and the electrochemical charge-discharge reactions. Chapter 1 offers a brief introduction to the Li insertion pathway, and explores the relationship between the rate-determining step for the insertion reaction and the electrochemical overpotential in Li-ion battery systems. The chapter also provides a concise overview of the research methods predominantly employed in this study: operando scanning transmission X-ray microscopy (STXM), and the phase-field simulation, and density functional theory (DFT) calculations for interpreting the real-time observations of the reactions. The research objective of Chapter 2 is to comprehend the unclear increase in electrochemical overpotential during high-rate lithiation in phase-separating electrodes, which is rarely observed during delithiation. In this study, operando STXM is applied to investigate the [100]-oriented LixFePO4 (LFP) sub-micron particles cycled over a broad range of currents (0.07 to 7C). The study revealed a strong dependence of surface and bulk phase evolution on the direction and magnitude of the applied current, successfully reconciling contradictions among existing models. More importantly, the experiments revealed that fast lithiation and delithiation currents stabilized Li-rich and -poor domains, respectively, in close proximity to the active particle surface. The Li-rich/-poor domains were identified as contributing to negative/positive self-feedback on the reaction rate, respectively, thereby increasing/decreasing electrochemical reaction overpotential. This research elucidated the origin of the asymmetry in overpotential and rate capability between fast lithiation and delithiation. Chapter 3 systematically demonstrates dynamic redirection of the Li transport pathway driven by cracks within LFP particles, and elucidates how internal chemo-mechanical stress from local Li compositions can contribute to additional crack formations. During charge-discharge process, uneven Li transport leads to nonuniform phase distribution, resulting in focused internal stress that often creates cracks within the materials. However, due to the challenges in observing dynamic Li transports, the intricate relationship between kinetic phase distribution and crack formation has remained poorly understood. Utilizing operando STXM on [100]-oriented LFP particles, it was explicitly observed that cracks within the particles decisively redirect lithiation/delithiation pathways, developing at the newly formed crack-surfaces rather than at the original (010) reaction surfaces. Moreover, ‘real-time crack opening,’ driven by the formation of Li-poor domains at the cracked-core region, was directly observed within the single particle during the charge process. Phase-field simulations supported the idea that significant tensile stress at the crack region is generated by Li-poor domains, opening up the crack and potentially governing further crack propagation. This study highlights how crack formation can redirect Li transport pathways and internal stress at the nanoscale, proposing cycling protocols to suppress crack formations for enhanced battery performance. Chapter 4 explores the origins of ambiguous discharge voltage fading in the Li1.1Ni0.23Mn0.44Co0.23O2 (LMR-NMC) electrodes over electrochemical cycles, which is less apparent in its charge process. This decay is often evaluated by transition metal (TM) migration or oxygen loss in the material. However, it encounters challenges in explaining the asymmetric voltage drop during discharge process. In this study, in-situ heating STXM experiments at 60°C were conducted on degraded LMR-NMC single particles to observe the evolution of elements’ oxidation states during the relaxation upon heating. The results clearly visualised the nanoscale metal-to-ligand charge transfer (MLCT) phenomenon from reduced Ni to oxidized O within the particles at the 120th discharged state. This provides evidence that, during the discharging process, Ni within the degraded LMR-NMC material is kinetically reduced first instead of O due to the impeded MLCT, contributing to the source of the severe discharge voltage fading. Furthermore, ex-situ experiments and DFT calculations support the hypothesis that MLCT between Ni and O is possibly impeded by oxygen vacancies formed within the materials. This research elucidates the origin of the asymmetric discharge voltage fading and provides atomic-level insight into LMR-NMC electrodes. In this study, coworkers and I successfully developed a real-time nanoscale chemical imaging platform for energy storage materials, enabling the exploration of heterogeneous redox reactions resulting from uneven Li transports, which could not be observed in previous studies. Additionally, meaningful connections were established between these nanoscale redox processes and the electrochemical behavior during charge and discharge cycles of bulk battery systems. Furthermore, comprehensive and theoretical interpretations of the observed noble kinetics were provided through phase-field simulations and DFT calculations. Ongoing advancements in nanoscale analyses, as introduced in this study, are expected to deepen humanity’s fundamental understanding of energy storage materials, potentially leading to paradigm-shifting discoveries for next-generation energy storage beyond Li-ion batteries. Keyword: Li-ion battery, cathode materials, electrochemical response, in-situ X-ray absorption imaging, nanoscale redox reactions, computational simulation Student Number: 2018-21067

Evaluating the Private and External Costs and Benefits of Select Large-Scale Li-Ion Battery Energy Storage Applications

Porzio, Jason Edward University of California, Berkeley ProQuest Disser 2024 해외박사(DDOD)

Lithium-ion (Li-ion) batteries have experienced a massive rise in popularity since their initial commercial introduction in 1991. Their implementation into several economic sectors has been instrumental in achieving large-scale electrification, with the eventual goal of sector-wide decarbonization. However, there has been little consensus on how to report the impacts associated with Li-ion batteries, and the standard of modeling the use-phase of Li-ion technologies often relies on broad assumptions, particularly with the future prices of Li-ion batteries. Additionally, there is little consensus on whether the integration of Li-ion technologies provides net positive impacts in several sectors. My research aims to provide recommendations for life-cycle assessments (LCA) on Li-ion technologies with the intent of helping future studies be more interpretable, representative, and impactful, as well as critically examine the assumptions used to forecast Li-ion prices. I then employ LCA and technoeconomic analysis (TEA) to model the climate, human health, and economic impacts of Li-ion technologies serving in peaker replacement and heavy-duty long-haul freight roles. The results from these studies show that the relative net impact of using Li-ion batteries in these roles can be positive or negative depending on several factors. Greater details of these studies are provided below.Life-cycle Assessment Consideration for Batteries and Battery MaterialsRechargeable batteries are necessary for the decarbonization of the energy systems, but life-cycle environmental impact assessments have not achieved consensus on the environmental impacts of producing these batteries. Nonetheless, life cycle assessment (LCA) is a powerful tool to inform the development of better-performing batteries with reduced environmental burden. This review explores common practices in lithium-ion battery LCAs and makes recommendations for how future studies can be more interpretable, representative, and impactful. First, LCAs should focus analyses of resource depletion on long-term trends toward more energy and resource-intensive material extraction and processing rather than treating known reserves as a fixed quantity being depleted. Second, future studies should account for extraction and processing operations that deviate from industry best-practices and may be responsible for an outsized share of sector-wide impacts, such as artisanal cobalt mining. Third, LCAs should explore at least 2-3 battery manufacturing facility scales to capture size- and throughput-dependent impacts such as dry room conditioning and solvent recovery. Finally, future LCAs must transition away from kg of battery mass as a functional unit and instead make use of kWh of storage capacity and kWh of lifetime energy throughput.Temporal Variations in Learning Rates of Li-ion Technologies: Insights for Price Forecasting and Policy through Segmented Regression AnalysisSince their initial development in 1991, Li-ion cell prices have decreased by over 97%. However, decades of lithium-ion battery cost reductions are often represented by a single learning rate in an experience curve. Learning rates are not inherently constant, however, and changes in learning rates can have dramatic impacts on cost forecasts and subsequent policy and investment decisions. This analysis is the first study to employ segmented regression to describe how learning rates have historically changed for lithium-ion technologies in different periods of time. Additionally, the distinctions between cost and price data are highlighted to emphasize the value of allowing learning rates to vary over time when performing experience curves for lithium-ion batteries. This analysis identifies past changes in the learning rate of lithium-ion cells, modules, and installations: for lithium-ion cells, the learning rate was 4% through 1997, 34% through 2003, and 24.4% onward. This dynamic learning behavior is explained as periods of market development, shakeout, and stabilization respectively. By allowing greater flexibility in the experience curve, a secondary shakeout period emerges from 2013 onward, with a learning rate of 40.9%. While this secondary shakeout has less statistical significance, we find that it aligns well with the growth of Li-ion markets and may emerge as significant as more data becomes available. Modules and installed costs follow a similar trend, with low learning (6-8% for 4-6 years) followed by an acceleration to 31-37%. The importance of capturing these historical variances is highlighted by demonstrating the impact of varying learning rates on forecasted lithium-ion cell prices through scenario analysis. We observe that price forecasts are much more sensitive to the uncertainty in learning rate compared to the uncertainty in technology deployment. Utilizing multiple learning rates from a segmented experience curve can enhance future Li-ion technology price projections, improving both price forecasting and policy development.Private and External Costs and Benefits of Replacing High-Emitting Peaker Plants with BatteriesFalling costs of Li-ion batteries have made them attractive for grid-scale energy storage applications. Energy storage will become increasingly important as intermittent renewable generation and more frequent extreme weather events put stress on the electricity grid. Environmental groups across the United States are advocating for the replacement of the highest-emitting power plants, which run only at times of peak demand, with Li-ion battery systems. We analyze the life-cycle cost, climate, and human health impacts of replacing the 19 highest-emitting peaker plants in California with Li-ion battery energy storage systems (BESS). Our results show that designing Li-ion BESS to replace peaker plants puts them at an economic disadvantage, even if facilities are only sized to meet 95% of the original plants' load events and are free to engage in arbitrage. However, five of 19 potential replacements do achieve a positive net present value after including monetized climate and human health impacts. These BESS cycle far less than typical front-of-the-meter batteries and rely on the frequency regulation market for most of their revenue. All projects offer net air pollution benefits but increase net greenhouse gas emissions due to electricity demand during charging and upstream emissions from battery manufacturing.Private and External Costs and Benefits of Electrifying Heavy-Duty Long-Haul Trucking with Li-ion BatteriesThe electrification of long-haul heavy-duty vehicles (HDVs) is necessary for the decarbonization of the transportation sector in the United States, but there is no clear technological pathway to replace the diesel internal combustion engine enabling this transport mode. Li-ion batteries have emerged as a popular candidate when exploring options to electrify HDVs, largely due to the rapidly growing popularity of Li-ion battery passenger electric vehicles and decreasing Li-ion battery prices. While many studies point to the climate and human health benefits that will arise from replacing diesel HDVs with Li-ion HDVs, other studies claim that technological limitations will make Li-ion HDVs economically inviable for long-haul freight. We use life-cycle assessment and technoeconomic analysis to model the total ownership cost, climate, and human health impacts associated with replacing a diesel Class 8 truck performing long-haul freight with a Li-ion Class 8 truck in the United States. Our results show that when including monetized contributions to global warming potential and human health burden, Li-ion Class 8 trucks in long-haul freight have greater lifetime costs per mile than diesel Class 8 trucks due to the high price and specific energy of Li-ion batteries, as well as high costs associated with the use of charging infrastructure. Additionally, the current use of Li-ion Class 8 trucks results in marginal improvements to social impacts relative to diesel Class 8 trucks under a high renewable energy cost scenario, but worse social impacts under a low renewable cost scenario. However, by 2035, the social impacts of Li-ion Class 8 trucks are substantially less than diesel Class 8 trucks under both renewable energy cost scenarios as more renewable energy is integrated into the electricity grid.

리튬 이온 전지 음극물질로서의 대면적 실리콘 나노와이어 및 실리콘아산화물 나노구조 제작

유선영 포항공과대학교 일반대학원 2011 국내석사

The size of anode materials is closely related with Li-ion battery performance. In nano-size systems the distances which Li+ must diffuse are decreased. These size effects; large surface areas, and fast diffusion rates; can play a large role in improving the performance of Li-ion batteries. And the benefit of Si materials as Li-ion battery anodes is highest known theoretical charge capacity (4,200mAh/g; 4). The use of Si materials as anode in Li-ion batteries is, however, still hindered from severe volume changes, 400%, during Li insertion/extraction. It is leading to poor cycling performance. So, we use more suitable SiOx nanostructure as Li-ion battery anodes than Si nanostructure In this thesis, we introduce two kinds of growth method of large scale synthesis of Si nanowires and SiOx nanostructures for using Li-ion battery anodes. First, Si nanowires were synthesized by catalyst-assisted chemical vapor deposition. Using this method, we can obtain single-crystalline Si nanowires. We then demonstrate the fabrication of intrinsic and doped single crystalline Si nanowire arrays with various height and surface roughness through metal-assisted chemical etching. Lastly, we synthesis of SiOx/NiSix core-shell nanowires. During charge/discharge, Li2O plays a structural buffering role to minimize the mechanical stress. Furthermore, electrically conducting and electrochemically inactive NiSix nanowire contributes to enhancing the cyclic behaviors of Li-ion batteries, where core matrix act as fast electron supplier. We have achieved the highly reversible capacity over 1800 mAh/g with its stable cyclic retention.

Two-Dimensional Correlation Spectroscopy on the Cathode Surface of the Li-ion Battery during Overcharged Process : 이차원상관분광학을 이용한 과충전된 리튬이온전지의 양극활 물질 연구

박연주 강원대학교 대학원 2013 국내박사

리튬이온전지는 현재 가장 널리 사용되고 있는 에너지원으로서 본 물질에 대한 연구가 활발히 진행되고 있으며, 계속적으로 증가하는 추세이다. 특히 기존 리튬이온전지보다 전기화학적 반응이 뛰어나고 용량이 커지는 방향으로 많은 노력을 기울이고 있다. 리튬이온전지의 성능 향상과 더불어 안정성의 문제도 함께 거론 되고 있으며, 이 또한 매우 중요한 과제로 남아있다. 따라서 안정성 향상에 대한 많은 연구가 최근 진행되고 있다. 본 연구에서는 리튬이온전지의 안정성의 문제로 대두되고 있는 과충전 상태의 전극 표면에서 일어나는 반응을 다양한 분광학을 사용하여 연구하였다. 본 연구에서는 기존에 양극활 물질로 많이 사용되고 있는 LiCoO2 전이금속산화물과 LiCoO2의 문제점을 극복하기 위해 대표적으로 연구되는 다성분계 전이금속산화물 (Li[Ni0.31Co0.32Mn0.28Al0.09]O2과 Li[Ni0.42Co0.18Mn0.25Al0.12]O2)에 대하여 과충전을 한 조건에서 실험을 하여 얻은 결과를 분석하였다. 분석은 Raman, 적외선 (IR), X-선 광전자 (X-ray photoelectron spectroscopy, XPS), X-선 흡수 (X-ray absorption spectroscopy, XAS) 분광학을 이용하였으며, 깊이 있는 분석을 위해 이차원상관분광학을 적용하였다. 과충전 된 LiCoO2 양극활 물질의 Raman, X-선 광전자, 이차원상관분광학을 이용한 분석 결과 전해질 자체의 분해와 전극과 전해질 반응으로 생기는 SEI (solid-electrolyte interface) 막의 구성성분이 Li2O, Co3O4, LiF, 폴리카보네이트, 카보네이트임을 알았고, SEI 필름 형성 메커니즘을 규명 할 수 있었다. 이차원 X-선 광전자 상관분광학으로부터 전해질 자체의 분해로 생기는 카보네이트가 전해질간의 고분자화반응으로 만들어지는 폴리카보네이트보다 빨리 형성된다는 것을 알 수 있었다. X-선 흡수 스펙트럼의 결과로 과충전 된 LiCoO2 양극활 물질에 형성된 SEI 막이 100 Å 이하의 두께임을 알 수 있었다. 과충전 된 다성분계 전이금속 산화물로 이루어진 양극활 물질에 대한 Raman, 적외선, X-선 광전자, 이차원상관분광학의 결과 LiCoO2 양극활 물질과 동일하게 SEI의 성분이 카보네이트와 폴리카보네이트임을 알았다. 이차원 X-선 광전자 상관분광학의 결과로부터 전해질 간의 고분자화반응으로 만들어지는 폴리카보네이트가 전해질 자체의 분해로 생기는 카보네이트보다 빨리 형성된다는 것을 알 수 있었다. Raman, 적외선 분광학의 결과로 Ni성분이 더 많이 들어간 다성분계 양극활 물질의 표면이 더 민감하다는 것을 알 수 있었으며, X-선 흡수분광학의 분석결과와 비교해보았을 때 Ni의 산화로 인해 양극 표면 반응이 달라짐을 알 수 있다. LiCoO2과 다성분계 양극활 물질에서의 반응을 비교한 결과 과충전된 LiCoO2에 경우 서는 SEI 막의 형성이 먼저 일어나며, 헥사고날 구조로 이루어진 LiCoO2 양극활물질의 구조적인 변성이 일어나고, 다성분계 전이금속 산화물의 경우 전극 표면의 산화가 먼저 일어나고, 그 다음에 SEI 막이 형성됨을 알았다. 본 결과는 전통적인 분광학적 분석 방법뿐만 아니라 이차원상관분광학을 이용하여 복잡한 전극표면 반응의 메커니즘을 명확하게 규명하였다. Li-ion battery has been widely used for electronic device and electric vehicles because of its higher density and higher capacity than the other commercial batteries (e.g., rechargeable Ni-Cd and Ni-MH batteries). LiCoO2 is generally used as a cathode material in commercial Li-ion battery. To enhance the cycle performance and stability, surface modification of LiCoO2 with metal oxides and the substitution of Al, Mg, and Zn for Co have been widely studied. However, some of reactions and processes in Li-ion battery at high voltage condition are very exothermic and can sometimes cause thermal runway and even battery explosion. Many studies to solve these problems have been reported. To understand the composition and structure of the solid- electrolyte interface (SEI) film on the cathode surface at high voltage conditions, various spectroscopic techniques (i.e., Raman, IR and X-ray spectroscopies) were applied to different battery systems in this study. Analysis of 2D Raman and 2D x-ray photoelectron spectroscopy (XPS) correlation spectra for the conventional Li-ion battery (LiCoO2/Li) system showed that organic surface components formed due to decomposition for electrolyte and reaction between electrolyte and electrode lead to structural distortion during high voltages. Results of x-ray absorption spectroscopy (XAS) also confirmed that the SEI layer was formed on the LiCoO2 electrode by decomposition of the organic solvents. The thickness of the SEI layer was less than 100 ?. For the mixed transition metal oxides (Li[Ni0.31Co0.32Mn0.28Al0.09]O2 and Li[Ni0.42Co0.18Mn0.25Al0.12]O2) systems, spectral changes of Raman spectra demonstrated that variations in the local atomic environment occurred with increasing extraction of Li+ ions. 2D XPS correlation spectra revealed that the mechanisms of SEI formation on the surface for cathodes of mixed transition metal oxides were different with that of LiCoO2. The polymerization and decomposition of the organic solvents on the mixed transition metal oxides surface taken place faster than that on the LiCoO2 surface. Analysis of Raman and IR spectra of the mixed transition metal oxides showed that the surface reaction of Li[Ni0.42Co0.18Mn0.25Al0.12]O2 was more sensitive than that of Li[Ni0.31Co0.32Mn0.28Al0.09]O2. Based on results of XPS and XAS spectra obtained during the overcharged process, it could be suggested that reactive Ni ions play an important role in the formation of the SEI layers. Furthermore, 2D XPS/Raman and 2D XAS/Raman hetero-spectral correlation spectra clearly elucidated the correlation between spectra observed the different probes, confirming their band assignments. In this study, surface reaction mechanism during overcharged process was clearly investigated at th molecular level by 2D correlation analysis. Results of this study can provide details of electrochemical reaction in cathode materials.

A Study on Lead-based Anode Materials for Li/Na-ion Batteries

Jehee Park Ulsan National Institute of Science and Technology 2022 국내박사

With the growing demand of electric transportation and stationary storage applications, Li-ion batteries are widely used because of their high energy density and long lifetime compared to other battery systems. As these batteries become commercialized, Li-containing oxide cathodes and graphite-based materials serve as a practical anode. Although the graphite in these conventional anodes offers low operating voltage, acceptable specific capacity and good electric conductivity, its limited capacity—372 mAh/g, 856 mAh/cm3—cannot satisfy the growing demands of advanced performance for electric applications. Accordingly, a variety of alloying and conversion-type materials have been considered as candidates with the potential of higher energy density. For example, Group IV metals (e.g., Si, Ge, Sn)-based materials can provide a significantly higher capacity, and Lead (Pb) is abundant, cheap and has a high volumetric capacity for Li storage (550 mAh/g, 1906 mAh/cm3). Its toxicity, however, has prevented it from being considered a serious contender as new anode material. The toxicity of Pb-based materials can be addressed with well-established closed recycling systems based on the current lead-acid industry. Furthermore, from the degeneration of lead-acid batteries by replacement with Li-ion batteries, reused Pb-based materials can be used in advanced Li-ion batteries. Herein, we demonstrate that Pb-based material are suitable candidates for Li-ion and Na-ion battery anodes and elucidate a unique energy-storage mechanism that challenges conventional alloying reactions. Based on these findings, the development strategies through simple material engineering for Pb-based anodes are proposed to improve electrochemical performance. The dissertation can be summarized as follows: In Chapter II, we demonstrate Pb anode using commercial powder was for Li-ion batteries and investigate electrochemical Li-alloying reaction by ex-situ XRD analysis. Electrochemical Li alloying reaction of Pb anode forms crystalline Li-Pb alloy compounds as intermediate phases and is different from other alloy-based anodes. A comparative study of Pb and other alloy-based anodes in Group IV provides a new understanding of the fundamental alloying reaction by material properties and emphasizes the advantages of Pb anode, resulting in improved performance with less volume change, faster Li diffusion and enhanced reversibility. In Chapter III, a new high-performance Pb-based nanocomposite anode material was developed for lithium-ion batteries. Pb@PbO-C nanocomposite was synthesized by facile high energy ball milling. Electrochemical performance tests showed excellent reversible capacity (600 mAh/g) and cycle stability (92% retention at 100th cycle). The carbon matrix buffers the large volume changes, and the nanoscale particle size can alleviate the stress. We monitored the detailed lithium storage mechanism with the wide reversible Pb redox (between Pb2+ and Pb4-) and the evolution of extended Zintl-type LiyPb structures during the electrochemical reaction through advanced characterization using synchrotron X-ray diffraction and absorption analysis. In Chapter IV, the Pb@PbO-C nanocomposite was used for Na-ion batteries with advantages of low-cost, easy fabrication process and high capacity. These batteries exhibit decent performance compared to previous works. However, their storage performance is poorer than that of Li cell even using the same materials. Compared to Li-Pb alloying phase transition, the Na storage mechanism was elucidated to the formation of Zintl clusters in the electrochemical Na-Pb alloying reaction by experiment, resulting in a limited sodiation kinetic and reversible capacity due to its ionic character and high electrical resistivity. These findings provide new insights for future research directions through understanding and development of the unique Na storage mechanism of Pb-based anodes. In Chapter V, further development strategies for Pb@PbO-C nanocomposite were suggested according to our new understanding. Although the carbon nanocomposite design resolves the issues on large volume change of Pb-based anodes, the Li/Na diffusion kinetics are still limited by the formation of ionic Zintl compounds as an intermediate phase. Therefore, Pb-based multi-metallic compounds (Pb-M-O/C) were introduced by adding other metallic elements during synthesis and obtained improved cycling performance for Li/Na-ion batteries. Furthermore, the alloy-type anodes suffer irreversible capacity loss especially at the initial cycle due to large volume change and solid electrolyte interphase (SEI) formation, which limits the commercial use by decreasing energy density of full cell. Chemical prelithiation/presodiation technique using biphenyl-based reagents was applied to compensate the Li/Na loss of Pb anode resulting in improved the coulombic efficiency. These results suggest the possibility of overcoming the limitations of high capacity Pb-based anodes and practical use in the future. Based on these promising results, this study provides an understanding of Pb-based anodes in terms of their unique Li/Na storage mechanism beyond the conventional alloying reaction and presents a new perspective of developing high-performance anode materials for Li/Na-ion batteries.

Elucidating Electrochemical Properties of Graphite from Specific Surface Area Control

김재원 서울대학교 대학원 2020 국내박사

Li-ion batteries are promising renewable energy storage devices that convert the chemical energy of lithium into electrical energy. Li-ion batteries consist of four main components: the positive electrode mainly determines the capacity and voltage of the Li-ion battery, and the negative electrode stores Li ions and electrons during charging and sends them to an external circuit during discharge. In addition, there is a separator that prevents physical contact between the positive and negative electrodes, and an electrolyte that allows for the movement of Li ions. Unlike the other components, where various alternatives have been developed over the past decades, graphite has remained the most commonly used material in the anodes of Li-ion batteries since their commercialization in 1991. This is due to the distinct advantages of graphite, which is cheaper than other anode materials such as silicon, lithium metal, or transition metal oxides. Graphite is also used as a reference electrode in many studies, and electrochemical characteristics close to theoretical values can be realized through various methods. This thesis focuses on selecting and interpreting factors that can predict long-term electrochemical properties in advance, rather than looking at and improving the characteristics of graphite electrodes through simple post-processing. In addition, the identification of the factors affecting the electrochemical properties is intended to be a useful for both academia and industry. In Chapter 1, a brief description of each component is given, along with the basic principles of Li-ion batteries. The overall research process and the limitations of the graphite anode are also described, and the factors that affect the electrochemical properties of the graphite anode are briefly mentioned. Nitrogen adsorption analyses are also briefly reviewed, as these are the main techniques used in this thesis. In Chapter 2, the correlation between nanopore of graphite and electrochemical properties are investigated. Artificial graphite (AG) and natural graphite (NG) have different physical properties, which should have significant effects on their unique electrochemical characteristics: NG is known to have superior specific capacity and low cost, whereas AG shows superior cyclability and low-swelling characteristics. Herein, we present a straightforward strategy using triphenylphosphine to improve NG, enabling its physical properties to resemble those of AG and thereby improving its cyclability and swelling characteristics. The volume of a-few-nanometer-sized pores of NG is significantly reduced by triphenylphosphine treatment, which is similar to that of the AG. As a result, triphenylphosphine-treated NG (TPP) anode shows improved cyclability (83.6% after 300 cycles) compared to that of NG, confirming the correlation between nanopore volume and cyclability. Furthermore, the expansion of single-layer pouch cell prepared with TPP, as monitored by in situ thickness measurement, is suppressed to 9% after 55 cycles, which is superior to the pristine NG showing 14% expansion after the same number of cycles. Electrochemical impedance analyses of symmetric cells reveal that the changes of ionic and solid-electrolyte interface (SEI) resistance are accompanied by the reduction of nanopores, confirming that the formation of decreased SEI layer is a crucial factor for realizing less-swelling and highly durable anodes based on the cost-effective NG materials. In Chapter 3, the correlation between heterogeneity of graphite surface and electrochemical properties are investigated. Graphite materials for commercial Li-ion battery usually undergo special treatment to control specific parameters such as particle size, shape, and surface area to have desirable electrochemical properties. In graphite, surface structure can be clearly classified to basal and nonbasal planes. As diffusion of Li-ion through nonbasal plane is much swifter than that of the basal plane, and reactivity on each plane is different, electrochemical properties of graphite are expected to be greatly influenced by the exposed area of nonbasal plane. In order to quantify the detailed surface structure of graphite materials, local-absorption isotherms were utilized, and the analyzed structure parameters of various commercial graphite samples were correlated with the electrochemical properties of each graphite anode. Thereby, we have confirmed that the fraction of nonbasal plane and fast charging capability has strong linear relation. With the additional peeling treatment which facilely increased the surface ratio of nonbasal plane by ~20%, various commercial artificial and natural graphite exhibited predictable tendency in terms of fast charging capability and cyclability. 리튬 이온 배터리는 리튬의 화학적 에너지를 전기 에너지로 변환하는 에너지 소자로 친환경 대체에너지원의 저장 시스템 중에서 가장 활발하게 연구가 이루어 지고 있다. 리튬 이온 배터리는 크게 4개의 구성요소로 되어 있다. 배터리의 용량과 전압을 주로 결정하는 양극과 함께 외부회로로 전자를 내보내는 음극이 있으며, 리튬 이온의 이동로 역할을 하는 전해액과 양극과 음극의 물리적 접촉을 막는 분리막으로 이루어져 있다. 다른 구성 요소와 비교해서 음극의 경우 1991년 상용화 될 때부터 지금까지 대부분이 흑연이 사용되고 있다. 실리콘이나 리튬 메탈 또는 전이 금속 산화물과 같은 다른 음극재 보다 값이 싸다는 흑연이 가지고 있는 뚜렷한 장점 때문이다. 따라서 많은 연구들에서 기준이 되는 전극으로 흑연이 사용되고 있으며, 이렇게 오래된 연구 기간 만큼 원하는 전기화학 특성을 얻기 위해 기본적으로 코팅이나 도핑과 같은 방법을 통해서 이론 값과 비슷한 수준으로 활용하고 있다. 본 학위 논문에서는 앞서 언급한 단순한 후처리를 통해 흑연 전극의 특성 향상을 살펴보고 그 원인에 대해 설명하기보다는, 오랜시간이 걸리는 전기화학 특성을 확인 하기 전에 미리 특성을 예측할 수 있는 인자를 선별하고 해석하는 것에 중점을 두었다. 또한, 이런 인자들이 전기화학 특성에 영향을 미치는 원인을 함께 확인 함으로써, 학계 및 업계에서 활용가능한 지표로써 도움이 되고자 하였다. 1장에서는 리튬 이온 배터리의 기본 원리와 함께 각 성분에 대한 간단한 설명을 한다. 그 중에서도 흑연 음극에 대한 전반적인 연구 과정 및 한계점에 대해서 설명하고, 또한 흑연 음극의 전기화학 특성에 영향을 주는 인자들에 대해 간단하게 언급한다. 그리고 본 논문에서 주로 사용 되는 분석 법인 질소 흡착 분석과 전기화학 임피던스 분석에 대해서 간단하게 리뷰한다. 2장에서는 흑연의 나노 포어와 전기 화학적 특성의 상관 관계를 중점적으로살펴 보았다. 인조 흑연과 천연 흑연은 물리적 특성이 서로 달라 상이한 전기 화학적 특성을 가진다. 천연 흑연은 우수한 비 용량과 비교적 저렴한 비용이 장점이며, 인조 흑연은 우수한 사이클 특성과 낮은 팽창 특성을 나타내고 있다. 우선 값이 싼 천연흑연에 간단한 트리 페닐 포스핀 처리를 통해서, 인조 흑연과 유사한 물리적 성질을 가지게하여 사이클 특성과 팽창 특성을 향상시킬 수 있었다. 특히 천연흑연의 수 나노 미터 크기의 기공의 부피는 트리 페닐 포스핀 처리로 1/3 가량 감소하는데, 이는 인조 흑연의 나노 기공의 부피와 유사한 정도이다. 결과적으로, 트리 페닐 포스핀 처리 된 흑연 음극은 천연흑연과 비교하여 개선 된 사이클 특성 (300 사이클 후 83.6 %)을 나타내었다. 또한, 팽창 특성은 트리 페닐 포스핀 처리 된 흑연 음극의 경우 55 사이클 후에 9%로, 15% 증가하는 천연 흑연과 비교하여 우수하다. 나노 기공에 따른 전기화학 특성의 연관성을 확인 하기 위해 대칭 셀의 임피던스 분석을 진행하였고, 이온 및 고체 전해질 계면 (SEI) 저항의 변화는 나노 기공의 감소와 함께 동일하게 감소하는 것을 알 수 있었다. 3장에서는 흑연 표면상태 (흡착 에너지 분포)에 따라 질소 흡착이 달라지는 것을 기반으로 흑연 표면을 기저면과 비기저면 (칼날면과 결함)으로 분류하였다. 그리고 이를 기존에 사용되었던 다양한 흑연 물성 (격자 상수, 비표면적, 입자 크기 분포, 결정립 크기 등등)과 함께 전기화학 특성 (고속 충전 및 수명 특성)과 어떤 연관성이 있는지를 확인하였다. 이를 위해 전극 로딩과 코인 풀 셀 환경을 상용화 수준으로 준비하였고 음극 또한 상용화된 흑연과 표면처리를 진행한 흑연을 이용하였다. 흑연의 경우 비등방성 때문에 리튬이온의 확산은 기저면과 비기저면에 따라 다르게 나타나고, 따라서 비기저면의 비율이 많은 흑연일수록 고속 충전 특성과 키네틱 특성이 우수하게 나타난다. 그리고 이런 비기저면이나 나노기공이 많은 흑연 표면에서는 고체 전해질 계면층이 지속적으로 생성될 수 있기 때문에 수명 특성에는 이를 통해 흑연의 수명특성을 저하시킴을 확인하였다.

Experimental and numerical studies on direct oil cooling system for Li-ion battery in electric vehicle

한정우 동아대학교 대학원 2023 국내석사

The objectives of this thesis are to investigate the cooling performance characteristics of Li-ion battery pack with direct oil cooling system technology and improve the cooling performance using various fin structures. The objectives of thesis are fulfilled through experimental and numerical studies on Li-ion battery pack with direct oil cooling system using dielectric fluid. The experimental study is conducted to investigate the electrical and thermal performance characteristics of Li-ion battery pack with direct liquid cooling. The numerical study investigates the thermal and flow performance characteristics of Li-ion battery pack with direct liquid cooling considering various fin structures. The experimental study is performed to analyze the discharge and heat transfer characteristics of 18650 cylindrical Li-ion pack immersed in dielectric fluid. The discharge voltage, discharge capacity, maximum temperature, temperature difference, average temperature, heat absorbed, and heat transfer coefficient are experimentally studied under various conditions of discharge rate, inlet temperature and volume flow rate. The experimental results showed that the operating voltage and discharge capacity are decreasing with increase in the volume flow rate and decrease in the inlet temperature for all discharge rates. At the higher discharge rate of 4C, the lowest battery maximum temperatures of 60.2 °C and 44.6 °C and the highest heat transfer coefficients of 2884.25 W/m^2 K and 2290.19 W/m^2 K are reported for the highest volume flow rate of 1000 mLPM and the lowest inlet temperature of 15 °C, respectively. The feasibility of direct oil cooling system for Li-ion battery pack is assured through the experimental parametric study conducted as a preliminary investigation. To improve the cooling performance of Li-ion battery pack direct oil cooling system, the numerical study is conducted using various fin structures. The thermal and flow characteristics of battery pack with direct oil cooling system are analyzed for three fin structures of triangle, rectangle and circular and compared with base structure without fins. The numerical results shows that the triangle fin structure depict best cooling performance compared to other fin structures, however the pressure drop of triangle fin structure is higher than other fin structures. The performance evaluation criteria (PEC) is considered to suggest the best fin structure. The PEC values of 1.014, 1.003, 1.055 are evaluated at the volume flow rates of 5 LPM for circular, rectangle, triangle fin structures, respectively. The triangle fin structure is recommended as the best fin structure to achieve the optimum thermal and flow characteristics of Li-ion battery pack with direct oil cooling system. Furthermore, the influence of fin dimensions of triangle fin structure is investigated on thermal and flow characteristics. The results reveal that the maximum PEC value of 1.049 is evaluated at the volume flow rates of 5 LPM for triangle fin structure with A/B ratio of 4.304. In summary, the present work proposes parametric investigations on dielectric fluid direct oil cooling system and best fin structure with optimum thermal and flow characteristics to achieve effective thermal management of battery pack in electric vehicles 본 학위논문의 목적은 절연유체 직접 유냉 시스템이 적용된 리튬이온 배터리 팩의 냉각 성능 특성 연구 및 다양한 핀 형상 구조를 배터리팩에 적용하여 냉각 성능을 향상시키는 것이며, 논문 목적을 달성하기 위해 절연유체가 사용된 직접 유냉식 시스템에 대해 실험적 및 수치 해석적 연구가 진행되었다. 실험적 연구는 직접 유냉시스템이 적용된 리튬이온 배터리 팩의 전기 및 열적 성능 특성 연구를 진행하였으며, 수치 해석적 연구는 다양한 핀 형상이 적용된 리튬이온 배터리 팩의 열 및 유동 성능 특성 연구가 진행되었다. 실험적 연구에서 절연유체에 침지된 18650 원통형 배터리팩에 대해 방전 및 열전달 특성을 분석하기 위해 다양한 방전 속도, 유체 입구온도, 유량 조건에서 방전 전압, 방전 용량, 최고온도, 온도 차이, 평균온도, 열전달량 및 열전달계수가 분석되었다. 실험 결과로 작동 전압 및 방전 용량은 유량이 증가할수록 입구 온도가 감소할수록 모든 방전 속도에서 감소하였으며 높은 방전 속도인 4C-rat 경우에서 가장 낮은 배터리 최고온도 및 온도 차이는 60.2 °C, 44.6 °C 으로 나타났다. 열전달 계수는 가장 높은 유량 1000mLPM 및 가장 낮은 입구 온도 15 °C에서 각각 2884.25 W/m^2 K, 2290.19 W/m^2 K 으로 나타났다. 직접 유냉 시스템이 적용된 리튬이온 배터리 팩의 타당성은 실험적 파라메트릭 연구를 통해 확인하였으며 직접 유냉 시스템이 적용된 리튬이온 배터리 팩의 냉각 성능을 향상시키기 위해 다양한 핀 형상을 이용한 수치 해석적 연구가 진행되었다. 직접 유냉 시스템이 적용된 리튬이온 배터리 팩에 대해 원형, 삼각형, 사각형의 세 가지 핀 형상 구조가 적용되었으며 핀이 없는 기본 구조에 대해 열 및 유동 특성을 비교 분석하였다. 수치해석 결과로 삼각형 핀 형상 구조가 다른 핀 형상에 비해 냉각 성능이 가장 우수하였으나 압력 강하 또한 가장 높게 나타났다. 성능 평가 기준(PEC)을 사용하여 3가지 핀 중 최적 효율을 가지는 핀 형상 구조 제안하였으며, PEC값은 5 LPM 에서 원형, 사각형, 삼각형 핀 형상 구조에 대해 각각 1.014, 1.003, 1.055로 나타났으며, 삼각형 핀 형상 구조가 직접 유냉 시스템이 적용된 리튬이온 배터리 팩에 대해 최적 열 및 유동 특성을 가지는 가장 효과적인 핀 구조 현상임을 확인하였다. 또한 삼각형 핀 형상 길이에 따른 열 및 유동 특성에 대해 연구하였다. 수치해석 결과로 A/B 비율이 4.304인 경우 5LPM에서 PEC 값이 1.049로 가장 높게 나타났다. 본 연구는 전기자동차 리튬이온 배터리 팩의 효율적인 열관리를 위해 차세대 냉각 방법인 절연유체을 이용하는 직접 유냉 시스템의 파라메틱 연구 및 최적 열 및 유동 성능에 대한 최적 핀 형상 구조가 제안되었다.