A Quantum Leap Forward for Li-Ion Battery Cathodes

Transcription

1 A Quantum Leap Forward for Li-Ion Battery Cathodes Josh Thomas Ångström Advanced Battery Centre, Uppsala University, Sweden. GCEP Research Symposium: Energy Research Five Years and Beyond Stanford University, 1-3 October 2008

2 See also: Poster #22: Kinson Kam, Anti Liivat et al. Poster #27: David Ensling et al. (esp. XPS/PES studies)

3 GCEP is GCEP... so a few ZEV-based perspectives on the general situation at some other spot on the industrially developed GLOBE: Sweden looks to CA for a lead yet tends to have all the answers USA can make things happen..... Sweden can only follow Swedes share the same who and our car industry is US owned killed the electric vehicle? suspicions USA is capitalist where Sweden is socialist - we look to private enterprise fluorishes the State to make the moves Sweden is more optimistic - but also more utopic and less realistic

4 A future vision borrowed from Toyota.... ALL future vehicle concepts will need (better) batteries!

5 This is where our project comes in... Title: A Quantum Leap Forward for Li-Ion Battery Cathodes A relatively uncontroversial consensus roadmap ahead towards ZEVs: ICE HEV PHEV FC-PHEV (P-)EV Time-scale? cf. Joan Ogden (Day 1) An estimate I saw somewhere a few months back:... the last totally ICE vehicle will roll of the production lines in give or take a year. But why especially better cathodes?

6 Where are batteries today after two centuries? Alessandro Volta, 1799 (Cu/Zn) 1839 (Fuel cell) 1859 Pb-acid 1899 Ni-Cd (Swedish!) 1973 Li-metal 1975 Ni-MH 1979 Li-polymer Li-ion: Sony 1990 Li-ion polymer: 2000 The battery industry is clearly very conservative

7 The rechargable Li-ion (polymer) battery good rechargability = the ability to extract and reinsert an optimal amount of Li many times (1000 s) from the active particles without significant loss in capacity

8 Better batteries Better materials Safer anodes We can upscale with what we have today More stable electrolytes The CATHODE is holding back the upscaling of the technology! Higher capacity ( performance ) Higher power ( performance ) Safer ( performance ) Longer lifetime ( performance ) Lower cost ( market ) More abundant materials ( market / environment ) Non-toxic ( environment )

9 A lower-cost EV-cathode material? Today s most common mobile phone/laptop material uses: Li(Co,Ni)O 2, LiNi 1-y-z Co y Al z O 2 Larger batteries demand lower-cost cathode materials the obvious candidate some Fe-based material: - LiFePO 4 (A123, etc.) - Li 2 FeSiO 4 (our focus) ( Fe- and Si-oxides make up >10% of the Earth s crust)

10 Cathode materials: a comparison a) Layered: LiCoO 2 -> Li 0.5 CoO 2 : ~3.9V, ~140 mah/g b) Spinels: LiMn 2 O 4 -> Mn 2 O 4 : ~3V or ~4.0V, 148 mah/g Instabilitility! Solution: doping c) Olivines: LiFePO 4 -> FePO 4, ~3.5V, ~170 mah/g d) Orthosilicates Li 2 FeSiO 4 -> LiFeSiO 4, ~2.85V, ~ 170 mah/g Poor el. conductivity! Solutions: - doping - coating - nano -sizing

11 Our new Fe-based cathode material Li 2 FeSiO 4 an orthosilicate Li 2 Fe 2+ SiO 4 LiFe 3+ SiO 4 + Li + + e - E= 2.85 V vs. Li + /Li (low!!) Q= 169 mah g -1 Advantages: Potentially lower cost, abundant, non-toxic High cycling efficiency High stability Li/Fe ordering!

12 Strategies for improvement... How can we extract >1 Li to give a higher capacity at a higher voltage? e.g., for x = 0.5 Li Li (Fe2+ (Fe2+ 1-x 1-x Mn2+ Mn2+ x )SiO x )SiO 4 4 Li+ Li+ 1-x 1-x (Fe3+ (Fe3+ 1-x Mn 1-x Mn x )SiO x )SiO (1+x)Li+ (1+x)Li+ + (1+x)e (1+x)e - -

13 Our strategy for improving performance A close interplay between DFT calculations and experiment! Compositional variation: transition-metal substitution, polyanion substitution, by-stander ions to give structural support, etc. DFT calculations: gives indications of how to optimize electrochemical performance Synthesis: e.g., nanostructures, coating,... Materials screening and characterization: - structure - surface properties (SEI layer?) - electrochemical performance

14 Nanostructures e.g., solution-based synthesis: 18 h Precipitation of metal nitrate salts Similar redox behaviour as the bulk material Wide range of particle sizes can be tuned 6 h 4 h 2 h Varying synthesis conditions

15 Strategy to compensate for the low electronic conductivity of Li 2 FeSiO 4 type materials nano-painting with an electronic conductor Conductive enough to distribute e - evenly Thin enough to allow the passage of Li + carbon coated particles

16 Nano-painting further improves particle contact... Carbon coating 1nm thick, transparent to ions Structured carbon black nanoparticles Electrolyte

17 Going beyond Li 2 FeSiO 4... e.g., substitution V 169mAh/g 3+ Start material: Li 2 FeSiO 4 Li 2 Fe SiO 4 Li Fe SiO 4 We want to make this box as small as possible! Inert Li! Fe 2+ /Fe 3+ only! Inert polyanion! Transition-metal substitution ~2.85V, ~3.8V 2+/2+ 3+/3+ 169mAh/g Polyanion substitution ~2.2V mAh/g Li 2 Mn x Fe 1-x SiO 4 Kinetics! Reversibility! >4.3V (169x) mah/g 4+/3+ Li Mn x SiO 4 Fe 1-x Li 2 Fe VO 4 ~3.4V 160mAh/g Li Fe VO 4 Li 1-x Mn x Fe 1-x SiO 4 Fe VO 4

18 The manganese-doped silicate? Li 2 Mn x Fe 1-x SiO 4 Mn Distortion in MnO 4 Mn = 0.1 lithiated delithiated undoped Substitution of Fe by Mn facilitates a >1 Li + transfer (Mn 2+ Mn 3+ Mn 4+ ) Similar X-ray diffraction patterns to undoped Li 2 FeSiO 4 First redox reaction occurs at > 4V (Mn 2+ Mn 3+ ) during 1st cycling

19 Polyanion substitution in Li 2 FeSiO 4? DFT modelling of substitutions: extra redox-activity VO 4 3- /VO 4 4- at SiO 4 4- V 4+ /V 5+ Fe 2+ /Fe 3+ Li 2 FeVO 4 160mAh/g 2.2V Li 1 FeVO 4 160mAh/g 3.1V - Complete delithiation! - Less strain (bonds: d Si-O <d V-O <d Fe-O ) - Higher σ el (smaller bandgap) But... - Volume expansion: ~10% Li 0 FeVO 4 - Low V 4+ /V 5+ voltage: 2.2V

20 Stability???? XPS/PES surface analysis of Li 2 FeSiO 4 We must probe the stability ( lifetime ) of the material for different electrolyte systems e.g., a comparative study of performance for two commonly used salts: LiTFSI vs. LiPF 6

21 Why is this critical? Surface chemistry plays a crucial role in defining the cycling behaviour of Li-ion batteries - we can improve cycling performance if we understand the cathode-electrolyte interface (= the SEI layer)

22 Electrolyte systems studied... Structure O Cut-off voltage [V] Conc. LiTFSI F 3 C S N S CF M F O O Li + O LiPF P M F F F F F Li + LiTDI M NC F LiPDI M NC O O B LiBOB M O NC NC O Li + N N Li + N N Li + O O CF 3 C CF 3 O O Solvents: EC : DEC 2:1 Name Lithium Corrosion on bis(trifluoromethane Al current collector -sulfonyl)imide lithium Sensitive hexafluorophosphat to H 2 O e Lithium trifluoromethyl dicyanoimidazole New! Lithium pentafuoroethyl dicyanoimidazole Lithium bis-oxalato Studied on the anode side borate

23 Electrochemical performance a - charge capacity b - discharge capacity 110 LiTFSI_1 a LiPF 6 LiTFSI_1 b Capacity [mah/g] LiTFSI_2 a LiTFSI_2 b LiPF6_1 a LiPF6_1 b Capacity [mah/g] Cycle Number Cycle Number LiTDI LiTFSI LiPF6_2 a LiPF6_2 b LiTDI_1 a LiTDI_1 b LiTDI_2 a LiTDI_2 b Capacity [mah/g] LiBOB LiPDI LiPDI_1 a LiPDI_1 b LiPDI_2 a LiPDI_2 b LiBOB_1 a LiBOB_1 b LiBOB_2 a LiBOB_2 b Cycle Number

24 Comparative study: LiTFSI vs. LiPF 6 Li 2 FeSiO 4 /C cycled with two electrolytes: (a) 1M LiTFSI (2:1) EC:DEC (b) 1M LiPF 6 (2:1) EC:DEC Measured with XPS on both lithiated and delithiated precycled electrodes. Safe transfer from cell to XPS analysis: Ar glovebox Transfer vessel S PES analysis hν e- S

25 Comparative study: LiTFSI vs. LiPF 6 C 1s charged discharged Intensity C1s LiFTSI -CF 3 a b LiTFSI -CO 3 -CH C-C -COH -COOH binding energy [ev] Solvent reaction: EC polymer Intensity C1s LiPF6 c d LiPF 6 -CO 3 -CH C-C -COH -COOH binding energy [ev] Si 2p charged discharged Intensity Si2p LiTFSI a b LiTFSI binding energy [ev] Intensity Si2p F 1s charged discharged LiPF6 -SiO -SiO 4 4 c LiPF d -Si-F binding energy [ev] Intensity F1s LiFTSI a b LiTFSI -CF 3 +binder LiF binding energy [ev] Intensity F1s LiPF6 c d LiPF 6 LiF LiPF 6 +binder +P-F-O binding energy [ev] Salt degradation: PF 6 P-F-O LiF formation Corrosion of electrode: Si-F bonds

26 SEI models deduced (schematic) PEO LiTFSI Li 2 CO 3 C/Li 2 FeSiO 4 LiPF 6 /Li x PF y /Li x PO y F z PEO LiF C/Li 2 FeSiO 4 EC (C 3 O 3 H 4 ) Polymer e.g., PEO LiPF 6 / Li x PF y / Li x PO y F z LiTFSI/TFSI - Li 2 CO 3 LiF Free surface? PEO LiTFSI electrolyte LiPF 6 electrolyte Decomposition of the solvent Degradation of the salt/electrode

27 Conclusions A combination of DFT and selective synthesis is proving a most fruitful route towards understanding how the full promise of the Li 2 FeSiO 4 system can be realized in a lower-priced Li-ion battery cathode material XPS/PES is helping us tune the optimal electrolyte to use for a given cathode in a battery

28 Thank you - GCEP!