Low Cost, Long Cycle Life, Li-ion Batteries for Stationary Applications

Daiwon Choi, Wei Wang, Vish V. Viswanathan and Gary Z. Yang

Pacific Northwest National Laboratory902 Battelle Boulevard P. O. Box 999, Richland, WA 99352, USA

Department of Energy, Washington DCNov. 2, 2010

Funded by the Energy Storage Systems Program of the U.S. Department Of Energy through Pacific Northwest National Laboratories

Introduction

Investigate the Li-ion battery for stationary energy storage to compliment the renewable energy resources such as solar/ wind power and load leveling for grid integration.

Energy storage unit for possible community application ~kWh level.

Li-ion battery energy storage with effective thermal management to improve its cycle and calendar life.

Design Li-ion battery pack suitable for large scale energy storage system.

Objectives

Screen possible electrode materials and its combinations suitable for large scale Li-ion battery.

Characterization of inherent heat generation of possible electrode candidates and its combinations.

Cost analyses of the full cell battery per energy density.

Improve the cycle life of Li-ion battery >1000cycles at 1C rate in coin cell configuration.

Cost Analyses

($/kWh) Based on EV battery configuration.

Li4Ti5O12 anode is more expensive due to lower specific capacity 175mAh/g over graphite (372mAh/g).

LiFePO4 cathode in full cell is slightly expensive due to lower potential 3.45V vs. Li+/Li.

Overall cost analyses depends on safety and cycle life of the full cell battery.

LFP : LiFePO4 LMO: LiMn2O4 NCA: LiNixCoyAlzO2 NCM: LiNixCoyMnzO2 LTO : Li4Ti5O12

Figure 1. Cost per kWh on various cathode / anodecombination Li-ion battery (40kWh).1-3

Ref) 1 Entek projected cost 2 TIAX estimate

3 Estimate based on numbers provided by Andrew Jansen at ANL

0

100

200

300

400

500

600

700

Figure 2. Entropy measurements on various electrode materials vs. state-of-charge (SOC).1

Entropy changes at various state of charge (SOC) have been screened for selection of various electrode combinations with effective heat management and safety (thermal runaway).

LiFePO4 cathode and Li4Ti5O12 or TiO2 anode combination show the low internal heat generation, due to minimum crystal structural changes “zero strain” during the Li insertion/extraction.

LiFePO4, Li4Ti5O12 and TiO2 redox potentials lie within stable window of conventional electrolyte where SEI layer formation is prevented.

Entropy vs. State of Charge (SOC)

Ref) 1 V.V. Viswanathan, D. Choi, D. Wang, W. Xu, S.Towne, R.E. Williford, J.-G. Zhang, J. Liu, Z. Yang, Journal of Power Sources,195 (2010) 3720–3729

Entropy Measurements

Figure 3. Average of absolute value of computed full cell entropy change over the 0~100% SOC range. FC LFP/LTO-NEI and FC LCO/G-L (last 2 columns) correspond to average of measured entropy change for full cells.

Li4Ti5O12

LiFePO4

Figure 4. SEM micrographs of LiFePO4 and Li4Ti5O12 electrode materials from different commercial sources(Company A, B, and C).

A

B

CB

C

Microstructures of LiFePO4 & Li4Ti5O12

LiFePO4 from company B and C was similar. Li4Ti5O12 from company C was not as good as those from company A and B.

Ketjen Black modified LiFePO4 by milling shows much improved rate performance.

Effective carbon mixing with electrode material and preparation step is critical for achieving the high rate and cycling stability

Figure 5. Electrochemical cycling at various C rates (a) LiFePO4 and Ketjen blackmodified LiFePO4, (b) Li4Ti5O12 (company A and C) and voltage profiles ofcharge/discharge at various C rates (c) LiFePO4 and Ketjen black modifiedLiFePO4, (d) Li4Ti5O12 (company A and C).

0 5 10 15 20 25 30 350

50

100

150

200

0 50 100 150 2002.0

2.5

3.0

3.5

4.0

4.5

0 50 100 150 2001.0

1.5

2.0

2.5

0 10 20 300

50

100

150

200

Volta

ge V

(vs.

Li)

Cycle number

10C

(a)5C2C

CC/2C/10

Charge-LFPKB Disharge-LFPKB Charge-LFP Discharge-LFP

Spec

ific

capa

city

(mAh

/g)

Cycle number

C/5 5C2CCC/2C/10 C/5

(d)

Specific capacity (mAh/g)

(c)LFP

C/10 1C 10C

LFPKB C/10 1C 10C

Specific capacity (mAh/g)

LTO (C) C/10 1C 5C

LTO (A) C/10 5C

(b)

Charge- LTO (A) Discharge-LTO (A) Charge-LTO (C) Discharge-LTO (C)

Cathode & Anode Half-Cell Rate Analyses

LiFePO4-Li4Ti5O12 Full Cell Evaluation

0 5 10 15 20 25 30 350

50

100

150

200

0 50 100 150 2001.0

1.5

2.0

2.5

10 100 100010

100

1000

0 200 400 600 800 1000 12000

25

50

75

100

125

150

175

200

(a)

5C

10C

2C

1C1/2C1/2C1/5C

1/10C

Charge LFPKB-LTODischarge LFPKB-LTOCharge LFP-LTODischarge LFP-LTO

Spec

ific

capa

city

(mAh

/g)

Cycle Number

(d)(c)

Specific Capacity (mAh/g)

Volta

ge

LFP-LTO C/10 1C 10C

LFPKB-LTO C/10 1C 10C

(b)

Pow

er d

ensi

ty (W

/Kg)

Energy Density (Wh/Kg)

LFP-LTO LFPKB-LTO LTO LFP LFPKB

Spec

ific

capa

city

(mAh

/g)

Cycle Number

LFPKB-LTO Charge Discharge

LFP-LTO Charge Discharge

Ketjen Black modified LiFePO4 by milling shows improvement in rate performance.

Effective carbon mixing with electrode material and electrode preparation is critical for achieving the high rate and cycling stability.

LiFePO4 cathode plays important role in cycling stability.

Figure 6. Electrochemical cycling at various C rates of (a) LiFePO4-Li4Ti5O12 fullcell with and without Ketjen black modified LiFePO4, (b) Ragone plot of full celland half cell, (c) voltage profiles of charge/discharge at various C rates and (d)long term cycling of LiFePO4-Li4Ti5O12 full cell with and without Ketjen blackmodification.

0 200 400 600 8000

50

100

150

200

250

300

350

400

0 200 400 600 800 1000 12000

50

100

150

200

250

300

350

400

(b)Sp

ecifi

c Ca

paci

ty (m

Ah/g

)

Discharge Charge

Anatase TiO2/Graphene

Cathode: LiFePO4

Electroyte: LiPF6 in EC:DMC

(a)Li4Ti5O12

Cycle Number

Over 1000cycles without negligible capacity fade have been achieved in coin cell.

Full cell cycling stability depends on LiFePO4 cathode stability.

Effective carbon coating on LiFePO4 cathode improves rate and cycle life.

Figure 7. Cycling performance of Li-ion batteries made from LiFePO4 cathode and (a) self-assembled anataseTiO2/graphene (5wt%) composite anode and (b) Li4Ti5O12 anode.1

Ref) 1 D. Choi, D. Wang, V. V. Viswanathan, I.-T. Bae, W. Wang, Z. Nie, J.-G. Zhang, G. L. Graff, J. Liu, Z. Yang, T. Duong, Electrochem. Commun.12 (2010) 378–381.

Full Cell Cycling Evaluation

0 500 1000 1500 20000

20406080

100120140160180200220240

Li4Ti5O12 (B) Charge Li4Ti5O12 (B) DischargeSp

ecifi

c Ca

paci

ty (m

Ah/g

)

Cycle Number

at 1C

Figure 8. Electrochemical cycling of Li4Ti5O12 and LiFePO4 cathode with various carbon (C1, C2 and superP) additives (10wt%).

LiFePO4 cathode plays important role in cycling stability.

Optimized cycling of LiFePO4 cathode using different conductive carbon.

On going test (>3500 cycles).

LiFePO4 and Li4Ti5O12 Half Cell Cycling

0 500 1000 1500 2000 2500 3000 3500

50

100

150

200

LFP + C1 Discharge LFP + C1 Charge LFP + C2 Discharge LFP + C2 Charge LFP Discharge LFP Charge

LFP from PhostechSpec

ific

Capa

city

(mAh

/g)

Cycle Number

at 1C

Commercial anatase TiO2 Anode

0

50

100

150

200

250

300

0 5 10 15 20 25 30 35 40 45 50

10wt% Ketjen black

20C

10C

5C

2.5C

1C

C/2.5C/5 20wt% Ketjen black

Spec

ific C

apac

ity (m

Ah/g

)

Cycle Number

C/10

10wt% Super P

40wt% Ketjen black

0 200 400 600 800 10000

20

40

60

80

100

120

140

160

180

200

Spec

ific

Capa

city

(mAh

/g)

Cycle Number

Discharge Charge(b) Rate : 20C

Cheapest anatase TiO2 commercially available have been tested for anode.

Working to improve the rate performance through effective carbon mixing.

Cycling stability is excellent.

Figure 9. Electrochemical rate test on (a) Ketjen black modifiedanatase TiO2 (b) extended cycling at 20C and (c) voltage profiles atvarious C-rates.

(a)

(c)

Summary

Various electrode materials have been tested for entropy (reversibleheat) measurements. LiFePO4/Li4Ti5O12 cathode/anode combination issuitable Li-ion battery for stationary energy storage.

Cost comparison on different electrode material combinations show~$600/kWh for LiFePO4/Li4Ti5O12 but life cycle cost ($/kWh/cycle) plussafety is expected to be competitive over other combinations.

Rate and cycling tests were performed on half and full cell andelectrodes are being further optimized. Cathode using LiFePO4 is criticalfor stable cycling.

Cycling performance of >1000cycles at 1C rate was achieved at R.T.

Future Tasks

Enhance cycling stability >1000 cycles of LiFePO4 using cheaper route.

Fabrication and testing of 18650 cell using LiFePO4 / Li4Ti5O12combination electrodes.

Calorimetric analyses of 18650 cell for total heat-generation.

Cost analysis on 18650 cell with cost per cycle will be reported.