Performance analysis of Lithium-ion-batteries: status and prospects

Transcription

1 Performance analysis of Lithium-ion-batteries: status and prospects DPG conference Erlangen March 218 Ellen Ivers-Tiffée, Philipp Braun, Michael Weiss Karlsruhe Institute of Technology (KIT), Germany KIT The Research University in the Helmholtz Association

2 DPG_Erlangen_E_Ivers-Tiffée.ppt, Slide: 2, Motivation Ragone Diagram 1 3 high energy high energy & high power Li-ion Duration W grav / Wh kg lead acid supercaps Ni-MH high power measurement P grav / W kg -1 Performance Ragone diagram on cell level

3 The Lithium-Ion Cell Resistance Contributions DPG_Erlangen_E_Ivers-Tiffée.ppt, Slide: 3, anode and electrolyte separator and electrolyte cathode and electrolyte e- Li + + copper e - aluminium 2-1 µm 2-4 µm 2-1 µm load

4 The Lithium-Ion Cell overpotential DPG_Erlangen_E_Ivers-Tiffée.ppt, Slide: 4, copper e- diffusion e - transport Li + + charge transfer Li + transport charge transfer diffusion e - e - transport aluminium 2-1 µm 2-4 µm 2-1 µm Loss processes U max Open Circuit Potential OCV = ϕ cathode - ϕ anode Transport (Li +, e - ) Charge transfer (Li + ) Diffusion (Li ) overpotential η I discharge i R i U cell / V U min working voltage (discharge) Q / As

5 The Lithium-Ion Cell energy- and power density DPG_Erlangen_E_Ivers-Tiffée.ppt, Slide: 5, e - transport Li + + copper e- diffusion charge transfer Li + transport charge transfer diffusion e - e - transport aluminium 2-1 µm 2-4 µm 2-1 µm Energy- and Power-density W P grav grav energy m t cell energy m OCV R I dq cell discharge discharge cell discharge cell i m i OCV R I dq t i i m discharge energy U dq U max U cell / V U min Open Circuit Potential OCV = ϕ cathode - ϕ anode Q / As

6 Characterization of Lithium-Ion Cells Electrochemical Impedance Spectroscopy (EIS) DPG_Erlangen_E_Ivers-Tiffée.ppt, Slide: 6, Characteristic time constants of different loss mechanisms in Lithium-ion cells: degradation formation diffusion electrochemical reactions charge transport frequency 1 nhz 1 µhz 1 mhz 1 Hz 1 khz 1 MHz f 1 a 1 a 1 M 1 d 1 h 1 min 1 s 1 ms 1 µs time Electrochemical Impedance Spectroscopy (EIS)

7 Characterization of Lithium-Ion Cells Electrochemical Impedance Spectroscopy (EIS) DPG_Erlangen_E_Ivers-Tiffée.ppt, Slide: 7, AC Small Signal Perturbation: Frequency Response Analyzer: Nyquist Plot: ut () it () Uˆ t Î ˆ ( ) j ( ) Z( ) e Uˆ( I ) Z '( ) jz ''( ) -Z 1 2 t 1 2 Z synthetic battery-like spectrum System Requirements: causality perturbation temperature T T end T start t t EIS t linearity time invariance T start! T end Kramers-Kronig Validity Test: Z Z ' Z ( ') ( ) d ' 2 Im Re 2 2 ' Z ( ') ( ) d ' 2 Im Im 2 2 ' M. Schönleber, D. Klotz, and E. Ivers-Tiffée, A Method for Improving the Robustness of linear Kramers-Kronig Validity Tests, Electrochim. Acta, vol. 131, pp. 2 27, 214.

8 Characterization of Lithium-Ion Cells Electrochemical Impedance Spectroscopy (EIS) DPG_Erlangen_E_Ivers-Tiffée.ppt, Slide: 8, Impedance Analysis: Z'' / cm What is the physical origin of these contributions? R ohmic Rpolarization Z' / cm 2

9 Characterization of Lithium-Ion Cells Electrochemical Impedance Spectroscopy (EIS) DPG_Erlangen_E_Ivers-Tiffée.ppt, Slide: 9, Temperature variation at SoC 8 %: State-of-charge (SoC) variation at 25 C: C 3 C 2 C 1 C C -4 % 25% 5% 75% 1% Z'' / cm² -6-4 R polarization Z'' / cm² -2 R polarization -2 R ohmic Z' / cm² Strong dependency of R ohmic and R polarization on temperature Z' / cm² Strong dependency of R polarization on SoC How to predict the performance of lithium-ion cells?

10 Modelling of Lithium-Ion Cells DPG_Erlangen_E_Ivers-Tiffée.ppt, Slide: 1, Behavior Model Advanced Equivalent Circuit Model Multiphysics Model - Simple Equivalent Circuit Model: no physical meaning + simple parameterization + very short computation time Finite Element Method: + physically correct - challenging parameterization - long computation time c Li / c Li,max 1 R R 1 Transmission Line Models: OCV C diff I cell (t) C 1 U cell (t) + physically motivated + feasible parameterization + short computation time

11 Modeling of Lithium-Ion Cells Transmission Line Model (TLM) electrode thickness Microstructure parameters: Volume fractions Current collector electrolyte Tortuosity Active Surface Area Particle Size Diffusion Length ionic path A ion A tortuosity (pore) volume fraction of pores electrode area charge transfer Z CT V a CT AM l a V CT AM specific charge transfer resistance active surface area per unit volume electrode volume solid-state diffusion 2 j l 1 ldiff I Diff ZDiff D Diff C D I Finite-Space Warburg Impedance diff 1 l Diff D C Diff diffusion length diffusion coefficient differential capacity [1] J. Bisquert, G. Garcia-Belmontea, F. Fabregat-Santiagoa, and A. Compteb, Anomalous transport effects in the impedance of porous film electrodes, Electrochemistry Commun., vol. 1, no. 9, pp , DPG_Erlangen_E_Ivers-Tiffée.ppt, Slide: 11,

12 Characterization of Lithium-Ion Cells Microstructure: Anode DPG_Erlangen_E_Ivers-Tiffée.ppt, Slide: 12, Anode: Graphite X-ray nano-tomography: X-ray camera sample X-ray source electrode thickness l electrode = 9 µm volume fraction ε graphite =.75 ε pore =.25 tortuosity pore = 5.12 active surface area a graphite =.314 µm -1 particle size d graphite,vol-av = 12.7 µm M. Ender, J. Joos, A. Weber, and E. Ivers-Tiffée, Anode microstructures from high-energy and high-power lithium-ion cylindrical cells obtained by X-ray nano-tomography, J. Power Sources, vol. 269, pp , 214.

13 Characterization of Lithium-Ion Cells Microstructure: Cathode DPG_Erlangen_E_Ivers-Tiffée.ppt, Slide: 13, Focused ion beam (FIB) tomography: Cathode: NCA/LCO Blend electron beam Ga + ion beam e - Ga + sample LiNi x Co y Al 1-x-y O 2 = NCA LiCoO 2 = LCO M. Ender, J. Joos, T. Carraro, and E. Ivers-Tiffée, Three-dimensional reconstruction of a composite cathode for lithium-ion cells, Electrochem. commun., vol. 13, no. 2, pp , 211. M. Ender, J. Joos, T. Carraro, and E. Ivers-Tiffée, Quantitative Characterization of LiFePO 4 Cathodes Reconstructed by FIB/SEM Tomography, J. Electrochem. Soc., vol. 159, no. 7, pp. A972 A98, Jan electrode thickness l electrode = 75 µm volume fraction ε AM =.57 ε carbon =.17 ε pore =.26 tortuosity pore = 4.29 active surface area a AM =.73 µm -1 particle size d AM,vol-av = 4.6 µm

14 Characterization of Lithium-Ion Cells Parametrization of transition line models DPG_Erlangen_E_Ivers-Tiffée.ppt, Slide: 14, e - conductivity impedance measurement -2 microstructure Z'' / cm Z' / cm 2 electronic conductivity e transmission line model (TLM) ion l volume fractions tortuosity electrode thickness a V ps active surface area particle size M. Ender, A. Weber, and E. Ivers-Tiffée, A novel method for measuring the effective conductivity and the contact resistance of porous electrodes for lithium-ion batteries, Electrochem. commun., vol. 34, pp , 213. M. Ender, J. Joos, T. Carraro, and E. Ivers-Tiffée, Quantitative Characterization of LiFePO4 Cathodes Reconstructed by FIB/SEM Tomography, J. Electrochem. Soc., vol. 159, no. 7, pp. A972 A98, Jan conclusion Electrochemical and microstructure parameters for anode and cathode are determined

15 Li-Ion Battery Model Homogenized 1D Model DPG_Erlangen_E_Ivers-Tiffée.ppt, Slide: 15, Advanced Equivalent Circuit Model: anode thickness separator thickness Ztotal Zanode RSep Zcathode cathode thickness Current collector χ 1 χ 1 χ 1 χ 1 χ 1 Sep χ 1 2 χ 1 2 χ 1 2 χ 1 2 χ 1 2 Current collector Zanode Zcathode Impedance: Discharge curve: Ragone Plot: calculated impedance Open Circuit Potential OCV = ϕ cathode - ϕ anode 1 3 Li-ion -Z'' / cm 2 ohmic losses charge-transfer cathode anode Z' / cm 2 diffusion U cell / V energy U dq W grav / Wh kg lead acid supercaps Ni-MH measurement model calculation 1 Q / As P grav / W kg -1

16 Model-Based Optimization DPG_Erlangen_E_Ivers-Tiffée.ppt, Slide: 16, Energy Optimization energy cathode y-axis x-axis anode dq Relevant Design Factors Example: cathode active materials 5 LCO HV Spinell NMC energy density safety fast-charging capability 4 LMO NCA potential / V 3 2 LFP power density costs LTO 1 Si graphite spec. capacity / mah g -1 cycle durability calendar life LFP NMC NCA LMO LFP: LiFePO 4 NCA: LiNi 1-x-y Co x Al y O2 NMC: LiNi 1-x-y Co x Mn y O 2 LMO: LiMn 2 O 4 HV Spinell:.3Li 2 MnO 3.7LiNi,5 Mn 1,5 O 4

17 Model-Based Optimization change parameters of state-of-the-art active materials DPG_Erlangen_E_Ivers-Tiffée.ppt, Slide: 17, Anode Cathode Passive Components High Energy Optimization Increase capacity Reduce mass Specific Capacity: graphite + 3% graphite + 15% Si (45 mah g -1 ) (35 mah g -1 ) Specific Capacity: LCO / NCA + 4% NMC811 Volume Fractions: 57% 7% + 2% NCA/LCO (145 mah g -1 ) (2 mah g -1 ) NMC811 current collectors: l cc = 2 µm - 5% l cc = 1 µm High Power Optimization Increase surface area Reduce electrode thickness Particle Size: Thickness: 9 µm 4 µm Particle Size: 12.1 µm.5 µm 4.1 µm.5 µm - 95% - 88% + Thickness: 75 µm 3 µm - 55% - 6% all high energy optimization steps

18 Model-Based Optimization change parameters of state-of-the-art active materials DPG_Erlangen_E_Ivers-Tiffée.ppt, Slide: 18, W grav / Wh kg lead acid supercaps Li-ion Ni-MH measurement high energy optimization high power optimization P grav / W kg -1 Ragone diagram on cell level

19 Model-Based Optimization change parameters of state-of-the-art active materials DPG_Erlangen_E_Ivers-Tiffée.ppt, Slide: 19, x1 2 Renault Zoe 25 Wh kg -1 W grav / Wh kg -1 2x consumer cell measurement high energy optimization high power optimization Tesla Model S BMW i3 24 Wh kg Wh kg P grav / W kg -1 Ragone diagram on cell level

20 State-of-the-arte electric vehicles comparison from cell to car DPG_Erlangen_E_Ivers-Tiffée.ppt, Slide: 2, Energy-Density: Cell-Level Energy-Density: Battery-Level Consumption W grav / Wh kg Wh kg Wh kg -1 NMC622 NCA 175 Wh kg Wh kg -1 NMC111 W grav / Wh kg consumption / Wh km Wh kg Wh kg Wh km Wh km Wh km -1 Renault Zoe Tesla Model S BMW i3 Renault Zoe Tesla Model S BMW i3 Renault Zoe Tesla Model S BMW i pouch cells cylindrical cells prismatic cells

21 DPG_Erlangen_E_Ivers-Tiffée.ppt, Slide: 21, Electric Range Prediction State-of-the-art Battery 2 km Future Prospects 8 km (calculated by volume*)

22 DPG_Erlangen_E_Ivers-Tiffée.ppt, Slide: 22, Acknowledgements: Research Partners and Team IAM-WET