Develop & evaluate materials &additives that enhance thermal & overcharge abuse

Zonghai Chen (PI) L. Zhang, C. K. Lin, Y. Qin, and K. Amine

Argonne National Laboratory May 17th, 2012

Project ID: ES035

This presentation does not contain any proprietary, confidential, or otherwise restricted information.

Overview

• Start: 10/01/2008 • End: 09/30/2014 • 60% completed

• Barriers addressed – Safety

• Total project funding – DOE - $1850K – Contractor - $ 0

• Funding received in FY10 – $470K

• Funding for FY12 – DOE - $500K

Timeline

Budget

Barriers

• Sandia National Laboratory • Superior Graphite Inc. • Hanyang University, Korea (YK Sun). • Yang Ren (APS)

Partners

3

Identify the role of each cell material/components in the

abuse characteristics of different cell chemistries.

Identify and develop more stable cell materials that will lead to more inherently abuse tolerant cell chemistries.

Secure sufficient quantities of these advanced materials (and electrodes) & supply them to SNL for validation of safety benefits in 18650 cells.

Objectives of the work

Approach

Secure materials

• Commercial source

• Partners • In-house

synthesis

Impact on safety at

component level

• Thermal analysis • Electrochemical

analysis

Validation at cell level

• SNL • Industrial

partners

Current targets: a) Safer electrode materials – cathode and anode b) Impact of surface chemistry on graphite c) Redox shuttles for overcharge protection

5

Recent Accomplishments and Progress Thermal decomposition pathway of delithiated cathodes

o Confirmed poor reproducibility of DSC data for delithiated cathodes

o Developed in situ high energy X-ray diffraction (HEXRD) and applied to

investigate the thermal decomposition of delithiated cathodes

o Identified different decomposition pathway of Li1-x(Ni1/3Mn1/3Co1/3)O2 at various conditions.

o Investigated the effect of salts (LiPF6 LiBF4, LiTFSI and Li2B12F12) as well as pure solvents on safety. (LiPF6 has negative impact on safety of cathode by reducing the onset temperature from ~310oC to about ~200oC).

Impact of ANL-2 redox shuttle on electrochemical performance o Studied cell chemistry: MCMB/LiFePO4 using 1.2 M LiPF6 in EC/EMC

(3:7, by weight) with and without 0.3 M ANL-2 as electrolyte. -The addition of 0.3 M ANL-2 slightly increases the impedance of the cell, but not major difference was observed on rate capability. -Life testing is ongoing. No major difference was observed on capacity retention. -No oral discussion about above topics due to the time limitation.

Impact of surface chemistry on graphite

o Secured nature graphite coated with amorphous carbon source from Superior Graphite.

-Heat treated the nature graphite at various temperatures from 700oC to 1080oC in glove box to introduce variation on the surface chemistry. -Physical, electrochemical, and thermal characterization of resulted materials are ongoing. -No oral discussion due to the time limitation.

Recent Accomplishments and Progress (cont’d)

Thermal response of battery components

50 100 150 200 250 300 350-1.5-1.0-0.50.0 (d) Separator

Temperature, oC

01020 (c) lithiated MCMB in Gen3E

0246

(b) delithiated NCM (Gen3P) in Gen3E

Hea

t flo

w, W

/g

-1012

(a) 1.2 M LiPF6 in EC/EMC (3:7) / Gen3E

• SEI decomposition occurs at the low temperature at about 110oC.

• Delithiated cathode generated largest amount of heat at above 200oC.

• Focus of FY12 is the thermal decomposition mechanism of delithiated cathodes.

Typical thermal response of delithiated cathodes

100 150 200 250 300 3500

1

2

3

4

5

6

7

Hea

t flo

w, W

/g

Temperature, oC

Li1-x[Ni1/3Mn1/3Co1/3]O2, 4.1V

1.2 M LiPF6 in EC/EMC (3:7, by weight)

•The onset of exothermal reaction is about 200oC. •Difficult to reproduce the data. •Sensitive to the ratio of electrode material to the added electrolyte. •Similar results were reported by Brian Barnett (TIAX) on LiCoO2. •XRD can help to study the change of cathode materials during heating.

Why high energy X-ray diffraction (HEXRD)? Real battery

High pressure sample

High-T furnace

Structure Defects

•Materials: Process optimization Quality control Intercalation mechanism Abuse tolerance •Battery: Nondestructive diagnosis Failure analysis Mechanical fatigue Heat generation Capacity and power fade

• Thermal decomposition of material

• Heat generation of redox shuttles

• LTO gassing

• Voltage drop of high energy cathode

2

1.0 1.2 1.4 1.6 1.8 2.0 2.2 2.4 2.6

Tem

pera

ture

, oC

50

100

150

200

250

300

350

400

In situ HEXRD setup

0 2 4 6 8

6.0 6.5 7.0 7.5 8.0

(222)

(311)

(220)(200)

(111)

(10,13)(208)

(119)(027)(025)

(024)(00,12)

(202)(116)

(10,10)

(110)

(113)

(018)

(009)(107)

(015)

(104)

(012)

(101)

Inte

nsity

2 θ

(003)

Layered oxide (Red)Iron/Copper (Blue)

λ=0.1 Anstrom(b)

? (420)

(331)

(400)(10,17)

(10,16)

(217)(125)

(214)(11,12)

(122)(02,10)(211)

(10,13)

(01,14)

(208)

?

(a) Sample

X-ray beam To detector

• Sector 11-ID-C at APS of ANL. • High energy X-ray source is critic to penetrate through the stainless steel

vessel that is used to seal the volatile solvents.

Impact of electrolyte components on thermal decomposition of Li1-x(Ni1/3Mn1/3Co1/3)O2 @ 4.1 V.

200 250 300 35005

10152025 (d) TBMPTFSI

Temperature, oC

05

10152025 (c) EMC

05

10152025 (b) EC

Hea

t flo

w, W

/g

05

1015202530

(a) EC/EMC

200 250 300 3500369

12 (h) LiPF6/TBMPTFSI

Temperature, oC

02468 (g) LiPF6/EMC

02468 (f) LiPF6/EC

Hea

t flo

w, W

/g

02468

10

(e) LiPF6/EC/EMC

• Solid line and dotted line represent two independent runs of the same charged electrode. • Solvents do show some impact. But major influence comes from the addition of LiPF6.

Structural evolution of “dry” Li1-x(Ni1/3Mn1/3Co1/3)O2 @ 4.1 V

1.2 1.3 2.5 2.6 2.7

Au

012 R

101 R

001 O

1

(a)400oC

170oC

318oC

Inte

nsity

2 θ

278oC

003 R

3.6 4.0 4.4 4.8 5.2 5.6 6.0

400oC

170oC

318oC

278oC

Inte

nsity

2 θ

(b)

• “Dry” sample: (1) half cell containing Li1.1(Ni1/3Mn1/3Co1/3)0.9O2 was charged to 4.1 V. (2) Open the cell in glove box and dry the electrode for 10 minutes. (3) Harvested about 3 mg electrode material and sealed in an in situ DSC high pressure vessel.

• “Dry” sample may contain delithiated NMC, carbon black, PVDF, and some LiPF6. • The sample was heated to 400oC with a heating rate of 5oC per minute, and XRD spectrum

was collected every 20 seconds. • The transformation of “dry” sample to O1 phase starts at about 278oC, and at about 318oC.

No further decomposition was observed up to 400oC.

Mechanism changes when it is “wet”

1.2 1.3 2.5 2.6 2.7

400oC

302oC288oC

250oCInte

nsity

2 θ

170oC

(a)

1.2 1.3 2.5 2.6 2.7

400oC

296oC

231oC

197oC

Inte

nsity

2 θ

170oC

(a)

3mg NMC + 3uL solvent 3mg NMC + 3uL electrolyte

Major difference: 3-step reaction (a) Expansion of c axis from 250oC to 288oC. (b) Conversion to O1 phase from 288oC to 302oC. (c) Disappearance of O1 phase above 302oC.

The onset temperature for the first reaction is reduced to 197oC with the addition of LiPF6.

Replacing LiPF6 with other lithium salts

1.2 1.3 2.5 2.6 2.7

(a)

350oC

305oC

271oC

223oC

Inte

nsity

2 θ

170oC

1.2 1.3 2.5 2.6 2.7

Inte

nsity

2 θ

350oC

307oC

270oC

215oC

170oC

(a)

LiBF4 LiTFSI

Li2B12F12 1.2 1.3 2.5 2.6 2.7

(a)350oC

302oC

270oC

213oC

Inte

nsity

2 θ

170oC

• The lithium salt shows significant impact on the first step reaction. The onset temperature increases in the order of LiPF6<LiTFSI~Li2B12F12<LiBF4.

• The onset temperature for the second reaction is 231oC for LiPF6 and 270oC for LiBF4, LiTFSI and Li2B12F12.

• The mechanism will be further investigated using density function theory.

Thermal decomposition of LiPF6

100 150 200 250 300 350

0

2

4(b) LiPF6 only

Temperature, oC

-10123

Hea

t flo

w, W

/g

(a) LiPF6/EC/EMC only •LiPF6 start to decompose at 200oC. •Electrolyte starts at about 220oC. •What’s the product of electrolyte decomposition? •What’s the impact of LiPF6? •DFT calculation is ongoing.

How about high energy cathodes? LiCoO2 High cost

Limited resource Low lithium utilization

LiMnO2 Low cost Abundant

High thermal stability Low energy density

LiNiO2 High energy density

Poor thermal stability

Concentration gradient

• Major effort has been devoted to searching high energy cathode in Ni-Mn-Co ternary system.

• Both concentration gradient cathodes and composite materials are promising for PHEV applications.

• How about the thermal stability of these materials?

Preliminary results on full concentration gradient cathode

Li1-xNi0.90Co0.10O2 Gradient cathode –Li1-xNi0.86Co0.10Mn0.04O2

• The thermal decomposition mechanism is different from that of composite NMC cathode. • Li1-xNi0.9Co0.1O2 started converting to spinel phase even at RT, and the conversion

completed at about 110oC, leading to poor capacity retention. • The newly formed phase reacted with electrolyte at about 200oC. • The gradient material didn’t convert to spinel phase at RT, and small portion of spinel

phase was observed at 150oC. The exothermal reaction between cathode and the electrolyte started at about 240oC.

1.2 1.3 1.4 2.5 2.6 2.7 2.8

178oC

Inte

nsity

2 θ

107oC

1.2 1.3 1.4 2.5 2.6 2.7 2.8

239oC

Inte

nsity

2 θ

149oC

18

Proposed Future work Due to the re-scoping of ABR program the future effort will be rebalanced

between the safety and the voltage fade of lithium-manganese-rich NMC materials.

Initiate effort to investigate the structural evolution of LMR-NMC during and after electrochemical activation using synchrotron-based in situ techniques.

Continue investigate the thermal decomposition mechanism of high energy cathodes and identify safer materials for automobile application.

Finishing the effort to identify the impact of the surface chemistry on thermal stability of SEI layer and thus cell safety.

19

Summary

•In situ high-energy X-ray diffraction technique was developed to study the structural change of delithiated cathodes during thermal decomposition in the presence of the electrolyte. •Lithium salt has significant impact on the decomposition pathway of delithiated cathodes. •Full concentration gradient cathode has shown to have advantage on the thermal stability.

Acknowledgment Support from David Howell and Peter Faguy of the U.S. Department of Energy’s Office of Vehicle Technologies Program is gratefully acknowledged. The authors also acknowledge the use of the Advanced Photon Source of Argonne National Laboratory supported by the U.S. Department of Energy, Office of Science, Office of Basic Energy Sciences.