본문 바로가기 주메뉴 바로가기
국회도서관 홈으로 정보검색 소장정보 검색

결과 내 검색

동의어 포함

목차보기

Title Page

Abstract

Contents

Ⅰ. Introduction 12

1.1. Lithium-ion batteries as conformers of energy demand increase 12

1.2. Needs for high energy density cathode 14

1.3. Mn-based cation-disordered rocksalt as next-generation cathode materials 15

1.4. Short-range order in DRX structure 17

1.5. Experimental backgrounds 18

1.5.1. Inferior electrical conductivity of DRX shackle its promising performance 18

1.5.2. Electrode engineering with multi-walled carbon nanotube 20

1.6. Computational perspectives 23

1.6.1. Density functional theory 23

1.6.2. Cluster expansion 24

1.6.3. Markov chain Monte Carlo simulation 25

1.6.4. Computational implementation of short-range order 25

Ⅱ. Computational details 26

2.1. Preparing crystal structures 26

2.2. Density functional theory calculation 26

2.3. Cluster expansion 27

2.4. Markov chain Monte Carlo simulation 27

2.5. Mn connectivity analysis 27

Ⅲ. Results and discussion 28

3.1. Defining electron transfer channel 28

3.1.1. Band structure of DRX 28

3.1.2. Polaronic charge transfer in DRX 30

3.2. Training cluster expansion models 33

3.2.1. Training data preparation 33

3.2.2. Fitting cluster expansion Hamiltonian 35

3.2.3. Minimizing overfitting by regularization 37

3.3. Constructing canonical ensemble 39

3.3.1. Markov chain Monte Carlo sampling 39

3.3.2. Temperature hyperparameter optimization 41

3.4. Mn percolation channel analysis 43

3.4.1. Automated analysis by python codes 43

3.4.2. Physical quantities demonstrating Mn percolation. 45

3.4.3. Mn percolation properties of SRO-applied DRXs 48

3.5. Ambivalent effect of Mn content 51

Ⅳ. Conclusion 53

Ⅴ. Reference 54

List of Tables

Table 1. The electrical conductivity of DRX materials (LMOF, LMTO, and LLF) and conventional layered cathode material (NCM811) measured by the DC polarization method. 19

Table 2. The first discharge energy density of LMOF electrodes with various concentrations, LMOF:CB:PVDF=70:20:10, 80:10:10, and 90:5:5 in weight percent. 21

Table 3. The first discharge energy density of LMOF electrodes with various concentrations, LMOF:MWCNT:PVDF=70:20:10, 90:5:%, 92:4:4, 94:3:3, and 96:2:2 in weight percent. 22

Table 4. RMSE value of trained CE model for training data and test data in Li₁.₂Mn₀.₄Ti₀.₄O₂ DRX. 36

Table 5. Optimized hyperparameter α value and RMSE value of CE model corrected by L₁- regularization in three DRX compositions, Li₁.₁Mn₀.₇Ti₀.₂O₂, Li₁.₂Mn₀.₄Ti₀.₄O₂, and Li₁.₃Mn₀.₁Ti₀.₆O₂. 38

Table 6. Electrical conductivity of Li₁.₁Mn₀.₇Ti₀.₂O₂, Li₁.₂Mn₀.₄Ti₀.₄O₂, and Li₁.₃Mn₀.₁Ti₀.₆O₂ DRXs measured by DC polarization method. 50

Table 7. Electrical conductivity of Li₁.₁Mn₀.₈Nb₀.₁O₂, Li₁.₂Mn₀.₆Nb₀.₂O₂, and Li₁.₃Mn₀.₄Nb₀.₃O₂ DRXs measured by DC polarization method. 50

List of Figures

Figure 1. Annual energy demand for rechargeable batteries in GWh and its prediction for the near future. 13

Figure 2. Cost (blue bar) and natural abundance (orange bar) of selected 4d, and 5d transition metal elements usually found in commercialized cathode materials and DRX materials. 16

Figure 3. Average discharge voltage and gravimetric capacity of various layered oxide and DRX oxide, oxyfluoride cathodes. Dashed lines represent gravimetric energy density. NMCA stands for Li-Ni-Mn-... 16

Figure 4. Scanning electron microscopy (SEM) images of carbon black (CB) and multi-walled carbon nanotube (MWCNT). 21

Figure 5. Voltage profiles and capacity retentions of Li₁.₆₈Mn₁.₆O₃.₇F₀.₃ (LMOF) cathodes with various electrode concentrations, LMOF:CB:PVDF=70:20:10, 80:10:10, and 90:5:5 in weight percent. 21

Figure 6. Voltage profiles and capacity retentions of Li₁.₆₈Mn₁.₆O₃.₇F₀.₃ (LMOF) cathodes with various electrode concentrations, LMOF:MWCNT:PVDF=70:20:10, 90:5:5, 92:4:4, 94:3:3, and 96:2:2 in... 22

Figure 7. Calculated orbital projected band structure of Li₁.₂Mn₀.₄Ti₀.₄O₂ DRX structure. 29

Figure 8. Mn-Mn edge-sharing and corner-sharing configuration which are compose of polaronic charge transfer in DRX structure. 32

Figure 9. Calculated energy barriers for hole polaron hopping to neighboring sites through edge- and corner-sharing configuration. 32

Figure 10. Relationship between Ewald electrostatic energy and calculated DFT energy in Li₁.₃Mn₀.₁Ti₀.₆O₂, Li₁.₂Mn₀.₄Ti₀.₄O₂, and Li₁.₁Mn₀.₇Ti₀.₂O₂ DRXs. The cross symbol stands for each... 34

Figure 11. Schematic illustrations of underfitting and overfitting. 36

Figure 12. RMSE value for training data and test data with respect to hyperparameter α in Li₁.₂Mn₀.₄Ti₀.₄O₂ DRX cluster expansion model training. The blue vertical line represents the optimized... 38

Figure 13. Flowchart of MCMC simulation. 40

Figure 14. Tetrahedral cluster occurrence of SRO-applied Li₁.₂Mn₀.₄Ti₀.₄O₂ DRX relative to random structures with respect to hyperparameter temperature. The green box remarks the closest match with... 42

Figure 15. Code output representing percolated Mn channels. 'Dead' represents isolated Mn which does not contribute to constructing electron transfer channel. 44

Figure 16. Venn diagram representing A, Bi, and U.[이미지참조] 47

Figure 17. Mn percolation probability and average content in both random and SRO-applied Li₁.₁Mn₀.₇Ti₀.₂O₂, Li₁.₂Mn₀.₄Ti₀.₄O₂, and Li₁.₃Mn₀.₁Ti₀.₆O₂ DRX structures. 49

Figure 18. Visualization of Mn percolation in SRO-applied Li₁.₁Mn₀.₇Ti₀.₂O₂, Li₁.₂Mn₀.₄Ti₀.₄O₂, and Li₁.₃Mn₀.₁Ti₀.₆O₂ DRX structures. 49

Figure 19. The average ratio of accessible and inaccessible Mn in Li₁.₁Mn₀.₈Nb₀.₁O₂, Li₁.₂Mn₀.₆Nb₀.₂O₂, and Li₁.₃Mn₀.₄Nb₀.₃O₂ DRXs. 52

Figure 20. The first charge volume change ratio of previously reported DRX cathode materials. 52

초록보기

 Mn-based cation-disordered rocksalt (DRX) materials are promising next-generation cathode materials for lithium-ion batteries owing to their sustainability, cost-effectiveness, and superior energy density. However, employing Mn-based DRX in cathode electrodes at a practical level has been hindered by their poor electrical conductivity. In this study, the factors affecting the electrical conductivity of Mn-based DRXs are investigated by cluster expansion (CE) method and Markov chain Monte Carlo (MCMC) simulations. As a result, it is revealed that higher Mn content in Mn-based DRX has an ambivalent effect, it causes higher electrical conductivity and larger volumetric strain simultaneously. To overcome this dilemma, fabricating low Mn content DRX with advanced conductive materials, for example, multi-walled carbon nanotube (MWCNT), and aliovalent substitution Cr3+, V3+ for Mn3+ are suggested. This work shed light on producing next-generation high-energy-density cathode electrodes in an industrial scope.