Embed Size (px)
Citation preview
Fluorinated Electrolyte for 5-V Li-Ion Chemistry
Vehicle Technologies Program Annual Merit Review and Peer Evaluation Meeting
Washington, D.C. June 16-20, 2014
Project ID #: ES218
This presentation does not contain any proprietary, confidential, or otherwise restricted information
Zhengcheng Zhang (PI) Libo Hu, Zheng Xue and Chi-Cheung Su
Kang Xu (US ARL, Co-PI), Xiao-Qing Yang (BNL, Co-PI)
Argonne National Laboratory
Project Overview
2
Timeline Barriers
Budget Partners
• Project start date: Oct. 1, 2013 • Project end date: Sept. 30, 2015 • Percent complete: 25%
• Battery life: conventional organic carbonate electrolytes oxidatively decompose at high potential (> 4.5 V vs Li+/Li )
• Battery performance: poor oxidation stability of the electrolyte limits the battery energy density
• Battery Abuse: safety concern associated with high vapor pressure, flammability and reactivity
• Total project funding - 100% DOE funding • Funding received in FY14: $362 K • Funding for FY15: $338 K
• U.S. Army Research Lab (collaborator) • Brookhaven National Laboratory (collaborator) • University of Rhode Island (interaction) • Jet Propulsion Laboratory (interaction) • Dr. Larry Curtiss – Theoretical modeling • Project Lead - Argonne National Laboratory
3 3
Project Objective To develop advanced electrolyte materials that can significantly improve the electrochemical
performance without sacrificing the safety of lithium-ion battery of high voltage high energy cathode materials to enable large-scale, cost competitive production of the next generation of electric-drive vehicles.
To develop electrolyte materials that can tolerate high charging voltage (>5.0 V vs Li+/Li) with high compatibility with anode material providing stable cycling performance for high voltage cathode including 5-V LiNi0.5Mn1.5O4 (LNMO) cathode and high energy LMR-NMC cathode recently developed for high energy high power lithium-ion battery for PHEV and EV applications.
FY14’s objective is to identify and screen several high voltage electrolyte candidates including fluorinated carbonates with the aid of quantum chemistry modeling and electrochemical methods and to investigate the cell performance of the formulated electrolytes in LNMO/LTO and LNMO/graphite chemistries.
Voltage
Cathode
Anode
Advanced Electrolytes Enabling High Voltage and High Capacity Cathode Materials
Capacity
4
R&D groups all over the world work on improving electrode materials in order to maximize both energy and power density of Li batteries. High voltage cathode (Li[MMn]2O4, M=Ni, Cr, Cu) and high capacity layered cathode (Li[NiMnCo]O2) red-ox potentials approach 5.0 V and 4.6 V vs Li+/Li. Conventional alkyl carbonates/LiPF6 tend to be oxidized around 4.5 V. Development of high voltage electrolyte is urgent and challenging.
Our overall approach for high voltage electrolyte research is to first design, synthesize and characterize fluorinated solvent candidates with the aid of quantum chemistry calculation; then screen the electrochemical properties of the synthesized solvents and validate their oxidation stability using 5-V LiNi0.5Mn1.5O4 /Li or LiNi0.5Mn1.5O4 /LTO cells. Tailored electrolyte additive will be developed in combination with the main fluorinated electrolyte to enable the high voltage cathodes coupled with graphite and silicon anode.
Fluorinated High Voltage Electrolyte research will be integrated with high voltage/capacity cathode projects in DOE ABR and BATT program. Various new fluorinated solvent systems including fluorinated carbonates, fluorinated sulfones, and fluorinated esters. Synergy effect of electrolyte containing hybrid solvents will also be explored to enable the high energy high power lithium-ion battery for PHEV and EV applications.
Approach
5
LiNi0.5Mn1.5O4/1.2M LiPF6 EC/EMC (3/7 wt%) / Graphite, cycled at RT and 55oC
Technical Accomplishments and Progress 5-V Ni/Mn Spinel LiNi0.5Mn1.5O4: the Challenge Instability of the cathode surface in contact with electrolyte at 4.8 V, especially at high T
4.8V
4.9V
4.9V
4.8V
55oC
4.9V
4.9V with a CV step
4.9V with a CV step
6
LiNi0.5Mn1.5O4
Cycled at 55oC
Pristine
OO
O
O O
O
P
F
FF
F
FFLi 1.2M LiPF6 EC/EMC 3/7 in weight + +
Thick layer of electrolyte deposition on both electrode surfaces
Graphite (A12)
5-V Ni/Mn Spinel LiNi0.5Mn1.5O4: the Challenge LNMO and graphite electrode surface morphology change when cycled with electrolyte at high T
charge
discharge
55oC
55oC
7
Electrolyte Design for High Voltage Cathode Materials LUMO
HOMO
µa
µc
Ecell
Anode
Cathode
LUMO
HOMO
µa (Li, C or Si)
Ecell = 4.3 ev
SEI
Ideal Unknown Electrolytes
LUMO
HOMOµc
SEI
Ecell = 4.7 ev
Anode
Cathode µc (LiMO2)
Anode
Cathode
µa (Li, C or Si)
Cathode Passivation Additive
LUMO
HOMOµc (LiNi0.5Mn1.5O4)
SEI
Ecell = > 4.7 ev
Anode
Cathode
µa (Li, C or Si)
SEI
New Solvents with Intrinsic Stability
SOA Carbonate Electrolytes
µa (LTO)
Ecell = 3.2 ev Ecell = 3.2 ev
µa (LTO)
Ewindow Ewindow
EwindowEwindow
Combined Approach for this Electrolyte Project
8
Introduction of fluorine (F) and/or fluorinated alkyl groups (Rf) to organic solvents including cyclic and linear carbonates, sulfones, cyclic and linear esters, and ethers to increase the voltage stability of electrolyte.
Molecular engineering to obtain partially or per-fluorinated solvents to afford an improved oxidation stability.
O
O
Rf
OO
O
RfO
Rf'O
ORf
ORf'
O
RfO
Rf'Rf
RfS
Rf'
O O
F-cyclic ester F-ester F-ether
F-cyclic carbonate
F-linear carbonate F-sulfone
ORf'
O
ORf
ORf'
O
Rf
F-ester
F-linear carbonate
Molecular Engineering Towards Intrinsic Oxidation Stability
9
DFT Calculation to Predict the Oxidation Potentials of Fluorinated Molecules Prior Organic Synthesis
Electron-withdrawing groups of -F and -Rf groups lower the energy level of the HOMO, thus increase the theoretical oxidation stability of the F-compounds.
DFT calculation indicates fluorinated molecules generally have much higher theoretical oxidation potentials than their non-fluorinated counterparts.
10
OO
O
OH
1) NaH (1.2 equiv) DMF, 0 °C, 20 min
2) CF3CH2I (1.1 equiv) DMF, 0 °C−rt, 60 h
OO
O
O CF3
TFP-PC-E (Tetrafluoropropyl-Propylene Carbonate-Ether)
CO2 (gas, 1 atm.)CH3P(C6H5)3I (5 mol%)
1-methoxy-2-propanol (solvent)rt, 4 d
OO
F2C
CF2HO
F2C
CHF2
OO
O
1) Reaction proceeded cleanly to 80% conversion after day 1, then gradually increased to >97% on day 4 (GC-MS without internal standard calibration);
2) Crude product was obtained after removal of the solvent and dried by 4Å molecular sieves;
3) Vacuum distillation (90
C/0.3 mmHg) twice gave pure product (>99.8% by GC-MS), ~45% yield;
4) Pure product was further characterized by 1H NMR, 13C NMR, FT-IR, and K-F titration (~80 ppm H2O content). Reaction setup Vacuum distillation-purification
CO2 inlet
Bubbler
Reaction mixture
Synthesis of Fluorinated Carbonates: Cyclic Carbonates
11
145 (m/z)
87 (m/z)
OO
O
H2CO
F2C
CHF2
S.M. (0.02%)
Ukn. 1 (0.06%) Ukn. 2 (0.12%)
OF2C
CHF2
OO
O
ab,cd,e
f
g,h
af
g,h
b,c
d,e
TFP-PC-E (Tetrafluoropropyl-Propylene Carbonate-Ether)
OO
F2C
CF2HO
F2C
CHF2
OCO
O
H HH
Spectroscopic Identification of Tetrafluoropropyl-Propylene Carbonate-Ether
O
R
CH3P(Ph)3I (5 mol%)
1MP, rt, 48 h+ CO2 (1 atm.) O
O
O
R
OF2C
CHF2
OO
O
CF3
OO
O
CH2FO
O
O
OF2C
CF2
OO
OF2C
CF2
H
CFO
O
OCF3
CF3
Other cyclic carbonate targets
12
O Cl
OR Rf OH
Et3N
O O
OR Rf
Rf OH = F3C OH F3CCF2
OH F3C OH
CF3
(R = Me, Et)
O O
OH3C
H2C
CF3 O O
OH2C
H2C
CF3H3C O O
OH3C
H2C
CF2
CF3
DMC
F-EMC
O O
OH3C
H2C
CF3
ab
b a
• GC-MS analysis: purity > 99.8%, (carbonate vs. DCM and TFE)
• NMR analysis: DCM and TFE is trace and cannot be accurately integrated. The amount of dimethyl carbonate is around 0.5 mol %, and the amount of F-DEC can be disregarded.
• Karl-Fischer titration: < 20 ppm moisture.
Synthesis of Fluorinated Carbonates: Linear Carbonates
P
F
FF
F
FFLi
EC(2) EMC(6 or 5) F-EPE(2 or 3) LiPF6
F-AEC(2) EMC(6) F-EPE(2) LiPF6
EC(2) F-EMC(6) F-EPE(2) LiPF6
E1 E2
E3 E4
E6
P
F
FF
F
FFLi
P
F
FF
F
FFLi
F2HCCF2
O
F2C
CF2H
F2HCCF2
O
F2C
CF2HOO
O
CFCF3
CF3
F2HCCF2
O
F2C
CF2HO
OO CF3
+ + +
+
+
+ +
+
P
F
FF
F
FFLiOO
O
CFCF3
CF3F2HC
CF2
O
F2C
CF2H
O
OO CF3+ + +
F-AEC(2) F-EMC(6) F-EPE (2) LiPF6
E5
+
O O
O
O O
O
F-AEC(1) EC(1) EMC(6) F-EPE(2) LiPF6
P
F
FF
F
FFLiF2HCCF2
O
F2C
CF2HOO
O
CFCF3
CF3+ + +
O O
O
+
OO
O
OO
O
OO
O
Formulation of Fluorinated Carbonates-Based Electrolytes
14
Constant voltage charge curve of Li/LiNi0.5Mn1.5O4 half cells maintained at 5.0 V, 5.1 V, 5.2 V, and 5.3 V for 10 hours at RT.
Electrochemical Oxidation Stability of Fluorinated Carbonate Electrolytes - Floating Test
The leakage current tested with the active electrode material LNMO is similar to the results obtained from the cyclic voltammetry using three electrode electrochemical cells.
Three-electrode electrochemical cell with Pt as working electrode and Li as reference and counter electrode.
Formulations with fluorinated solvent, especially the all-fluorinated formulation E5, shows much higher oxidation stability (lower leakage current) than the conventional electrolyte Gen 2.
0 100 200 300 400 500 6000.000
0.010
0.020
0.030
0.040
0.050
0.060
0.070
0.080
E1~E6E3
Gen 2
5.7 V
Time (s)
15
High Temperature Performance of Fluorinated Electrolyte LNMO/LTO LNMO/Graphite
Electrolytes with F-AEC as the only cyclic carbonate (E3 and E5) have SEI formation issue on graphitic anode.
55oC, 3.45-2.0V
E5
E6
RT, 3.45-2.0VRT, 4.9-3.5V
55oC, 4.9-3.5V
E5E3
16
Fluorinated Carbonate Electrolyte with Graphite SEI Formation Additives
OO
O
F
FEC
55 ºC
55 ºC
OF2C
CHF2
OO
O
TFP-PC-E
OO
O
VC
OB
O F
F
O
O
Li
LiDFOB
SEI formation additives works moderatelywell with E5 electrolyte , but has minimaleffect on E6 electrolyte in which ethylenecarbonate (EC) was used.
17
New Generation of Fluorinated Carbonate Electrolyte for LNMO/Graphite Cells
F-EC(3) F-EMC(5) F-EPE(2) LiPF6 (1.0 M) LiDFOB (1%)
HVE 1 P
F
FF
F
FFLiF2HCCF2
O
F2C
CF2HO
OO CF3+ + + OO
O
F+ O
BO F
F
O
O
Li
The new formulation HVE 1 shows tremendous compatibility with graphite surface as indicated by the improvement in LNMO/graphite cells compared with Gen 2 electrolyte, especially at 55 ºC.
LNMO/graphite cellCycled between 3.5-4.9 V,C-rate: C/3,Tested at RTand 55 ºC, respectively.
RT
55oC HVE 1
Gen 2
18
Electrode Morphology after Cycling with HVE 1 at 55 oC
18
Cycled Electrode @ 55oC, 100th cycle
Cycled Electrode @ 55oC, 100th cycle
Covered in thick organic residues.
No change compared with pristine.
Pristine graphite
Pristine LNMO
Gen 2 graphite HVE 1 graphite
Gen 2 LNMO HVE 1 LNMO
Pristine Electrode
19
3500 3000 2500 2000 1500 1000 50030
40
50
60
70
80
90
100
Pristine Graphite Cycled Graphite in HVE 1 (4.9 V at RT) Cycled Graphite in HVE 1 (4.9 V at 55oC)
Tra
nsm
itta
nce
(%)
Wave number (cm-1)
(d)
3500 3000 2500 2000 1500 1000 50030
40
50
60
70
80
90
100
Pristine Graphite Cycled Graphite in Gen2 (4.9 V at RT) Cycled Graphite in Gen2 (4.9 V at 55oC)
Tra
nsm
itta
nce
(%)
Wave number (cm-1)
(c)
3500 3000 2500 2000 1500 1000 50030
40
50
60
70
80
90
100
Pristine LNMO Cycled LNMO in Gen2 (4.9 V at RT) Cycled LNMO in Gen2 (4.9 V at 55oC)
Tra
nsm
itta
nce
(%)
(a)
3500 3000 2500 2000 1500 1000 50030
40
50
60
70
80
90
100
Pristine LNMO Cycled LNMO in HVE 3 (4.9 V at RT) Cycled LNMO in HVE 3 (4.9 V at 55oC)
Tra
nsm
itta
nce
(%)
(b)
FT-IR of both cycled LNMO cathode and graphite anode showed less decomposition when HVE 1 electrolyte was used at RT and 55°C.
Cycled Electrodes Harvested from LNMO/Graphite Cells
FT-IR spectra of pristine and harvested cathodes from cycled LNMO/graphite cells with (a) Gen 2 and (b) HVE electrolytes and harvested anodes from cycled LNMO/graphite cells with (c) Gen 2 and (d) HVE electrolytes. All cells cycled at RT and 55 ⁰C.
20
GC-MS of harvested electrolyte from cells cycled at 55 ºC.
New Formulation of Fluorinated Carbonate Electrolyte in High Voltage LNMO/graphite System
Gen 2 showed a large amount of varieties of decomposition products in the harvested electrolyte, while no side products were observed in the cycled HVE 1 electrolyte by GC-MS.
transesterification
Gen 2 HVE 1
21
CondiGon: 3-‐cycle formaGon; cutoff voltage: 3.5-‐4.9 V; C/10; at room temperature.
Self Discharge Test: charge to 4.9 V at C/10 at 55 °C, then rest at the same temperature (55°C) and monitor the voltage change. Data points are taken every 5 minutes.
Self-Discharge Performance of HVE Electrolyte with LNMO/Graphite (A12) Cells
In LNMO/graphite coin cells, HVE electrolyte has improved storage stability than Gen 2 at high temperature.
22
Charging
Li+
Lithiated graphite
Discharge
Partially lithiated graphite
Lithiated LNMO
Li+
Li metal will discharge first
Graphite LNMO
Li
DelithiatedLNMO
Li metal or SLMP
Press and contact with electrolyte
Lithiated graphite
Li metallithiation
Anode Graphite
Lithiated graphite
Graphite anode
Spacer
LNMO cathode
Separator
Spacer
PP gasketWave washer & cap
Case
LiLiLi Li
(a) (b)
(c) (d)
Lithium Reservoir to Compensate Active Lithium Loss
(a) LNMO/graphite cell assembly with incorporated lithium metal; (b) lithium metal working mechanism at the formation cycles; (c) prelithiation of graphite anode using electrochemical method; (d) direct shorting of graphite anode.
55oC
HVE 1 With Li reservoir
HVE 1
Gen2
HVE 1
Gen2
23
Reversibility of the charge and discharge profiles of LNMO/graphite cells using HVE 1 electrolyte with lithium reservoir was greatly improved. (Cut-off voltage: 3.5-4.9 V at C/3 rate at 55 ºC)
Voltage Profiles of LNMO/Graphite Full Cells with HEV1 and HVE1 + Li Reservoir Cycled at 55 oC
HVE 1 charge
HVE 1 discharge
HVE 1 + Li charge
HVE 1 + Li discharge
• 0.5 M LiPF6 in EC/DMC (50:50)
25
JP-F 10
No apparent effect of solvent ratio on 19F-chemical shift
• ~0.10 ppm variations (negligible compared with 17O-chemical shift changes of carbonyl group ~20 ppm)
• Dependence within experimental error Hence, there is little anion-solvent interaction
With increase salt concentration, 19F-chemical shift changes down-field, still with small variations (~0.19 ppm)
However, coupling constant between P-F (JP-F) reveals interesting solvent preference information
• With increasing EC concentration JP-F decreases
• DMC presence interferes the P-F bonding more than EC
Hence, PF6 slightly prefer DMC than EC.
Anion-Solvation: 19F-NMR
-2
2
6
10
ΔR
rms (
nm)
Upper layer
Control FEC
Control + 5 v% FEC
In-situ AFM of SEI on Model Graphite Anode
HOPG (WE, sample) at 0V vs. Li (RE/CE), 1st cycle
1
10
100
1,000
Thi
ckne
ss (
nm)
Control FEC
SEI upper layer thickness derived from F-d curves
Roughness change
Control: 1.3 M LiTFSI in EC
-2
2
6
10
ΔR
rms (
nm)
Lower layer
OCV to 0.8 V OCV to 0 V 0.8 to 0.0 V
Upper layer Lower layer
Lower layer
Lower layer
27
Collaboration: (1) U.S. Army Research Laboratory (Dr. Kang Xu, Project team member) (2) Brookhaven National Laboratory (Dr. Xiao-Qing Yang, Project team member) (3) Center of Nano-Materials at Argonne (Dr. Larry Curtiss)
Interactions: (4) University of Rhode Island (Dr. Brett Lucht) (5) Jet Propulsion Laboratory (Dr. Marshall Smart) (6) Lawrence Berkeley National Laboratory (Dr. Vincent Battaglia) (7) Cell Analysis, Modeling, and Prototyping Facility (CAMP) (Dr. Andrew Jansen) (9) Material Engineering and Research Facility (MERF) ( Dr. Gregory Krumdick) (10) Arkema (Dr. Ryan Dirkx)
Collaboration and Coordination with Other Institutions
28
During the rest of the FY14, we will continue to explore the fluorinated carbonate-based electrolytes to enable the high voltage high energy cells.
- Explore the additive effect on the new developed high voltage fluorinated electrolyte HVE 1 on graphite electrode. - Further design and synthesize new fluorinated carbonate solvents based on the recent research results. - Tailored cathode electrolyte interphase (CEI) additives to further improve the instability on the LNMO/electrolyte interphase, especially at elevated temperatures. - Expand the electrode surface diagnosis using X-ray and XPS in addition to AFM. - Scientific write-up for publication in peer-reviewed journals of the recent results.
In the year of FY15, we propose the following work in order to achieve the milestones
and the final goal of this project:
- Design and synthesis of fluorinated non-carbonate solvents as backup high voltage electrolytes. - Optimization of fluorinated electrolytes. - Fabrication and evaluation of 10 mAh pouch cells in the lab. - Delivery of twelve 10 mAh pouch cells to DOE for testing and verification.
Proposed Future Work
29
PHEV and EV batteries face many challenges including energy density, calendar life, cost, and abuse tolerance. The approach of this project to overcome the above barriers is to develop highly stable electrolyte materials that can significantly improve the high voltage cell performance without sacrificing the safety to enable large-scale, cost competitive production of the next generation of electric-drive vehicles. Argonne took a combined approach to tackle the voltage instability of electrolyte by
developing the fluorinated carbonate-based electrolytes with intrinsic stability and the passivating cathode additive to afford a stable electrode/electrolyte interphase.
Fluorinated cyclic carbonates and fluorinated linear carbonates were synthesized, characterized and purified and their electrochemical performance in high voltage cell were evaluated in LNMO/Li, LNMO/LTO and LNMO/graphite cells.
FEC-based electrolyte HVE 1 has achieved superior capacity retention at both ambient and elevated temperatures for 5 V LNMO/graphite cell. Post-test analysis showed that the fluorinated electrolyte HVE 1 is much more stable in both the liquid electrolyte phase and on the electrolyte/electrode interface.
LNMO/graphite (A12) cells with fluorinated electrolyte showed less self-discharge even at elevated temperature with 100% SOC.
Lithium compensation provides an efficient way to enhance the cycle life with LNMO/graphite cells even with conventional electrolyte; however a more stable electrolyte, like fluorinated electrolyte, in combination with an active lithium reservoir is still preferred for desired performance.
New electrolyte additives were synthesized and characterized; Live-formation of SEI by F-solvent was observed by in-situ electrochemical AFM.
Summary
