Integrated Lithium-Ion Battery Model Encompassing Multi ... · Major Activities and...

NREL is a national laboratory of the U.S. Department of Energy, Office of Energy Efficiency and Renewable Energy, operated by the Alliance for Sustainable Energy, LLC.

Integrated Lithium-Ion Battery Model Encompassing Multi-Physics in Varied Scales: An Integrated Computer Simulation Tool for Design and Development of EDV Batteries

Gi-Heon Kim*, Kandler Smith, Kyu-Jin Lee, Shriram Santhanagopalan, Ahmad Pesaran

*: [email protected]

The 11th InternationalAdvanced Automotive Battery Conference

January 24 – 28, 2011, Pasadena, CaliforniaInternational Battery Modeling Workshop

NREL/PR-5400-50248

Innovation for Our Energy Future

CAEBATProgram Coordinator NREL

Task 1:Open Architecture

Development(ORNL Lead)

Task 2Pack Model

Development(NREL Lead)

Task 4Electrode/Material

Models(LBNL Lead)

Major Activities and Responsibilities

Task 3Cell Model

Development(NREL Lead)

DOE’s CAEBAT Program

2

• To integrate the accomplishments of battery modeling activities in national lab programs and make them accessible as design tools for industry

• To shorten time and cost for design and development of EDV battery systems

Industry/University Participation (RFP)


Contents

3

1. Introduction to the NREL’s MSMD model• Multiphysics multiscale lithium battery model framework

2. Model application to large Li-ion battery performance• Stacked prismatic cell response simulation

• Spiral wound cylindrical cell response simulation

3. Model application to large Li-ion battery degradation• Large tab-less cylindrical cell degradation simulation

4. Model application to large Li-ion battery safety• Multiphysics internal short circuit simulation

5. Summary


Performance, Durability and Safety

4

Physics of Li-Ion Battery Systems in Different Length Scales

Li diffusion in solid phaseInterface physicsParticle deformation & fatigueStructural stability

Charge balance and transportElectrical network in composite electrodesLi transport in electrolyte phase

Electronic potential &current distributionHeat generation and transferElectrolyte wettingPressure distribution

Atomic Scale

Particle Scale

Electrode Scale Cell Scale

System ScaleSystem operating conditionsEnvironmental conditionsControl strategy

Module ScaleThermal/electricalinter-cell configurationThermal managementSafety control

Thermodynamic propertiesLattice stabilityMaterial-level kinetic barrierTransport properties


Charge Conservation

r

( ) Ues −−= φφη

Porous Electrode Performance Model

5

( ) qTktTcp ′′′+∇⋅∇=∂∂ρ

• Pioneered by Newman group (Doyle, Fuller, and Newman 1993)

• Captures lithium diffusion dynamics and charge transfer kinetics

• Predicts current/voltage response of a battery• Provides design guide for thermodynamics,

kinetics, and transport across electrodes

eeeffDee

effss

effes

Li cTUTUjq φκφφκφφσφφ ∇⋅∇+∇⋅∇+∇⋅∇+

∂∂

+−−=′′′ ln

Charge Transfer Kinetics at Reaction Sites

Species Conservation

Energy Conservation • Difficult to resolve heat and electron current transport in large cell systems


Mesoscale Modeling Approach

6

Wang and Sastry, JES, 2007

• Model addresses correlation of composition, morphology, and processing conditions by resolving mesoscale geometry

• Captures mesoscale geometry impact on transport properties of composite electrodes

• Computationally expensive

Liu and Siddique, 218th ECS , 2010

Micro-Structure Reconstruction

Computational domain generated by quasi-random reconstruction process

P.R. Shearinga et. al, Electrochemistry Communication, 2010

X-Ray Tomography (Nano CT)


NREL’s Multi-Scale Multi-Dimensional Model Approach

7

Design of Materials

Design of Electrode Architecture

Design of Transport atElectrode/Electrolyte

Design of Electron &Heat Transport Operation & Management

MSMD-µNREL

MSMD-c

ξ1

ξ2ξ3

x1

x2x3

X1

X2X3

particle domain dimension electrode domain dimension cell domain dimensionξ x X


NREL’s Multi-Scale Multi-Dimensional Model Approach

8

Design of Materials

Design of Electrode Architecture

Design of Transport atElectrode/Electrolyte

Design of Electron &Heat Transport

MSMD-µNREL

MSMD-c

ξ1

ξ2ξ3

x1

x2x3

X1

X2X3


• Introduce multiple computational domains for corresponding length scale physics

• Decouple geometries between submodel domains• Couple physics in two-way using predefined inter-domain

information exchange• Selectively resolve higher spatial resolution for smaller

characteristic length scale physics• Achieve high computational efficiency• Provide flexible & expandable modularized framework


The Model has Hierarchy StructureMSMD: Modularized Framework: Flexible & Expandable

9

ξ1

ξ2ξ3

x1

x2x3

X1

X2X3


p_module 1

p_module k

p_module 2

p_physics b

p_physics z

p_physics a

e_module 1

e_module 2

e_module k

e_physics b

e_physics z

e_physics a

c_module 1

c_module 2

c_module k

c_physics b

c_physics z

c_physics a



10





Atomic Scale

Particle Scale






Model Prediction for a Large Stacked Prismatic Cell

11

ξ1

ξ2ξ3

x1

x2x3

X1

X2X3

1D spherical particle representation model


r

1D porous electrode model 3D SPPC model

Stacked prismatic design 200 x 140 x 7.5 mm3 form factor 20 Ah PHEV10 application Single side cooling 25W/m2K 25oC

SVM

Sub-model Choice

Solution Method Choice

SVM FVM


Cell Design Evaluation

12

Case Description Lx [mm] Ly [mm] Lz [mm] Tab width [mm] Tab configuration ND Nominal design 200 140 7.5 44 Adjacent tabs CT Counter tab design 200 140 7.5 44 Counter tabs ST Small tab design 200 140 7.5 20 Adjacent tabs WS Wide stack-area design 300 140 5.0 44 Adjacent tabs

Nominal Design

Small Tab Design

Counter Tab Design

Wide Stack-area Design

ND

ST

CT

WS

Tab Location

Stack AreaTab Size


5C Discharge Voltage Response

13

• Identical discharge capacity: 18.9 Ah at 5C

• Several mV difference in discharge voltage among the compared designs

• Tendency of a few millivolts voltage difference with design change cannot be easily confirmed by testing only

0 5 10 15 202.5

3

3.5

4

discharge capacity [Ah]

volta

ge [V

]

NDCTSTWS

9.8 9.9 10 10.1 10.23.43

3.44

3.45

3.46

3.47

CT WSND ST


0 5 10 15 202.5

3

3.5

4


volta

ge [V

]

NDCTSTWS

Cell Internal SOC Imbalance

14

0

49.5

50

50.5

51

51.5

52ND

ST

CT

WS


SOC Deviation during Discharge

15

0 5 10 15 200

5

10

15

20

25

30

35


dQ/d

V [A

h/V

]

V

0 5 10 15 200

0.5

1

1.5

2

2.5

3


∆ S

OC

[%]

NDCTSTWS

SOCmax-SOCmin dQ/dV

• Results imply that Flat voltage slope would promote cell internal SOC imbalance HEV cycling at “flat section” would cause larger internal imbalance

• Modifying thermodynamics vs Optimizing electrical/thermal configuration

CT WSND ST


Thermal Response during Discharge

16

0 5 10 15 200

5

10

15

20

25

30

35

40


heat

gen

erat

ion

[W]

NDCTSTWS

0 5 10 15 2025

30

35

40

45

50

55


tem

pera

ture

(°C

)

0 5 10 15 2025

30

35

40

45

50

55


tem

pera

ture

(°C

)

0 5 10 15 2025

30

35

40

45

50

55


tem

pera

ture

(°C

)

Total Heat Generation

STCT WS

Temperature

• Similar average temperatures: ND, CT, ST• Smaller ∆T at CT• Larger ∆T at ST• Heat generation is highest with WS, but the

EOD average T is lowest

ND

Single side cooling on top surface With h = 25 W/m2K At Tamb= 25 oC


CT

WS

Temperature Imbalance at EOD

17

X(mm)

Y(m

m)

0 50 100 150 2000

50

100

X(mm)

Y(m

m)

0 50 100 150 2000

50

100X(mm)

Y(m

m)

0 50 100 150 2000

50

100

X(mm)

Y(m

m)

0 50 100 150 2000

50

100

X(mm)

Y(m

m)

0 50 100 150 2000

50

100

X(mm)

Y(m

m)

0 50 100 150 2000

50

100X(mm)

Y(m

m)

0 50 100 150 200 250 3000

50

100

X(mm)

Y(m

m)

0 50 100 150 200 250 3000

50

100

46

47

48

49

50

47

48

49

46

47

48

49

50

51

43

44

45

46

47

48

ND

ST

Tavg= 48.1oC∆T = 4.3oC




Cooling Surface

BottomSurface


Pulse Power Response Comparison

18

• HPPC at 20% SOC at 25oC initial temperature


Vehicle Use Evaluation

19

0 5 10 150

0.5

1

1.5

2

2.5

3

time [minute]

∆ S

OC

[%]

NDCTSTWS

0 5 10 150

20

40

60

80

100

time [minute]

Sta

te o

f Cha

rge

[%]

NDCTSTWS

0 5 10 15-1000

-500

0

500

1000

time [minute]

Pow

er [W

]

SOCavg SOCmax-SOCmin

Battery Power per Cell

• PHEV10 mid-size sedan • 15 minutes US06 Driving Profile• Battery power from Vehicle simulation

Thermodynamics + Cell Design + System Control


Thermal Response during Driving

20

0 5 10 150

2

4

6

8

10

12

time [minute]

∆ T

[°C

]

NDCTSTWS

0 5 10 1525

30

35

40

45

50

time [minute]

aver

age

tem

pera

ture

[°C

]

NDCTSTWS

• Similar average temperatures: ND, CT, ST• Smaller ∆T at CT• Larger ∆T at ST• WS: lower average T during CS mode drive, but significant ∆T


Cell Internal Kinetics Non-Uniformity

21

0

12.8

13

13.2

13.4

13.6

13.8

14

14.2

6.0% 2.5%

6.9% 12.7%

CT

WS

ND

ST

• Ah Throughput

• PHEV10• 15min US06


Model Prediction for a Spirally Wound Cylindrical Cell

22

ξ1

ξ2ξ3

x1

x2x3

X1

X2X3



r

1D porous electrode model 3D SWC model

Spirally wound cell design D40, H100 mm form factor 10 Ah PHEV10 application

SVM

Sub-model Choice


SVM FVM


Spirally Wound Cell (SWC) Model

Stacking process: Forming a pair between inner electrodes

current collector

current collectorSeparator

Winding process: Forming a second pair between outer electrodes

electrode

electrode

Spirally Wound Cell :

Paring Inner electrodes

Separator

electrode

electrode

Paring Outer Electrodes

• One pair of wide current collector foils• Two pairs of wide electrode layers• Complex electrical configuration

23


Model Case

24

Diameter 40mm, inner diameter 8mm, height 100 mm form factor Positive tabs on the top side, negative tabs on the bottom side 10 Ah capacity

5C constant current dischargesocini = 90%Natural convection :

hinf = 5 W/m2KTamb = 25oC Tini = 25oC

0 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 1.8 20

0.05

0.1

X [m]

]m[

Y

0 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 1.8 20

0.05

0.1

X [m]

]m[

Y

Tab locations for 5 tab case

Negative current collector

0 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 1.8 20

0.05

0.1

X [m]

]m[

Y

0 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 1.80

0.05

0.1

X [m]

]m[

Y

Tab configuration of each electrode pairs

Positive current collector

Inner electrode pair

Outer electrode pair


X [m]

Y [m

]

0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 1.8 20

0.05

0.1

122124126128

SWC Model Results

25

X [m]

Y [m

]

0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 1.8 20

0.05

0.1

0

0.01

0.02

X [m]

Y [m

]

0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 1.80

0.05

0.1

122124126128

Electrochemical reaction rate

Electric potential

X [m]

Y [m

]

0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 1.8 20

0.05

0.1

-0.02

-0.01

5 tabs in each current collector 5 min after 5C discharge

Inner electrode pair


Positive current collector

Negative current collector

0 2 4 6 83

3.2

3.4

3.6

3.8

Time [min]

V out [V

]

Topview

Bottomview

Current mainly flows in the winding direction

More energetic reactions near tabs


0 2 4 6 825

30

35

40

45

50

Time [min]

Tem

pera

ture

[o C]

X [m]

Y [m

]

0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 1.8 20

0.05

0.1

0.3

0.3

26

State of charge

TemperatureX [m]

Y [m

]

0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 1.80

0.05

0.1

0.32

0.34

SWC Model ResultsInner electrode pair


High rate of discharge with a moderate heat transfer condition

Heat generation dominates temperature distribution in the system

Temperature difference in the system is relatively small

- More usage of electrode near tabs

Innovation for Our Energy Future27

Discharge kinetics rate comparison

- Discharge KineticsImpact of # of tabs

X [m]

Y [m

]

0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 1.8 20

0.05

0.1

120 121 122 123 124 125 126 127 128 129 130

X [m]

Y [m

]

0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 1.8 20

0.05

0.1

X [m]

Y [m

]

0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 1.8 20

0.05

0.1

X [m]

Y [m

]

0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 1.8 20

0.05

0.1

i” [A/m2]

0.2%

2.2%

6.6%

32.2%

Δi”/ i”avg

2 tabs

5 tabs

Continuous tab

10 tabs

in the inner electrode pair at 5 min

Innovation for Our Energy Future28

- TemperatureImpact of # of tabs

X [m]

Y [m

]

0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 1.8 20

0.05

0.1

X [m]

Y [m

]

0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 1.8 20

0.05

0.1

-0.4 -0.3 -0.2 -0.1 0 0.1 0.2 0.3 0.4T-Tavg[°C]

Temperature comparison

0.19°C

0.37°C

0.78°C

3.25°C

2 tabs

5 tabs

Continuous tab

10 tabs

ΔTat 5 min discharge



29





Atomic Scale

Particle Scale






Model Prediction for Cylindrical Cell Degradation

30

ξ1

ξ2ξ3

x1

x2x3

X1

X2X3



r

1D porous electrode model+ empirical life model 2D SPPC model

Spirally wound cell design D40, H100 mm form factor 10 Ah PHEV10 application

SVM

Sub-model Choice


SVM FVM


Life Modeling ApproachNCA datasets fit with empirical, yet physically justifiable formulas

31

ResistanceGrowth

RelativeCapacity

•Data shown above: J.C. Hall, IECEC, 2006.•Model also fit to DOE/TLVT, Southern CA Edison & NASA data

Qactive = e0 + e1 N

R = a1 t1/2 + a2 N

Calendar fade• SEI growth (partially

suppressed by cycling)• Loss of cyclable lithium • a1, d1 = f(∆DOD,T,V)

Q = min ( QLi , Qactive )

QLi = d0 + d1 t1/2

Cycling fade• active material structure

degradation and mechanical fracture

• a2, e1 = f(∆DOD,T,V)

• Portability + Physical interpretation• Applicable to complex real-world storage and cycling scenarios

Rela

tive

Cap

acit

y (%

)

Time (years)

r2 = 0.942

Li-ion NCA chemistry

Res

ista

nce

Gro

wth

(mΩ

)

Tafel-Wohler model describing a2(∆DOD,V)


US06 – Nonuniform Capacity Loss• Regions near terminals

suffer most significant capacity lossLarge overpotential Excessive cycling

• Inner core loses capacity faster than outer cylinder wallHigh temperature Material degradation0 months:

8 months:16 months:

+

-

+

-

+

-

US06

32


US06 – Ah Imbalance (Nonuniform Cycling)

33

0 months:0.7% Ah Imbalance 8 months:

1.7% Ah Imbalance 16 months:4.8% Ah Imbalance

• Later in life, those same areas are most degraded and are cycled least

+

-

+

-

+

-

Preferentially cycled regions shift early in life

Imbalance continually grows throughout life

• Early in life, inner core and terminal areas are cycled the most



34





Atomic Scale

Particle Scale






Modeling Thermal Runaway

35

Constructed empirical reaction models using calorimetry data for component decompositions: approach practiced by J. Dahn’s group

Enhanced understanding of the interaction between heat transfer and exothermic abuse reaction propagation for a particular cell/module design

Provided insight on how thermal characteristics and conditions can impact safety events of lithium-ion batteries

20 40 sec8 28

Total Volumetric Heat Release from Component Reactions

4 2416 3612 32

Internal T External T

(°C)

0 20 40 (sec)

SEI decomposition positive/electrolyte negative/electrolyte


Multi-Physics ISC Model

36

• Developed an integrated model for multi-physics internal short circuit (ISC) of lithium-ion cells by linking and integrating NREL’s unique electrochemical, electrothermal, and abuse reaction kinetics models

• Performed 3D multi-physics internal short simulation study to characterize an internal short and its evolution over time

Current Density

Temperature

Electrothermal Model

Abuse Kinetics Model Electrochemical Model

050

100150

200

0

50

100

15059.5

60

60.5

61

61.5

62

X(mm)

soc [%]

Y(mm)

Internal Short Model Study

socT


40 mm

35 mm

3 mm

Shorted Spot

Shutdown Separator for Large Cells ?

37

Short Between Al & Cu Metal FoilsAl

Cu

• Cell Capacity: 20 Ah 0.4 Ah

Rshort ~ 10 mΩIshort ~ 300 A (15 C-rate)

Rshort ~ 7 mΩIshort ~ 34 A (85 C-rate)

Joule Heat for Short Temperature @10 sec after short

800oC

25oC

surface temperature

internal temperature

T

Q

30 sec

T

Q

40 sec

T

Q

50 sec

T

Q

60 sec

T

Q

20 sec

100 110 120 130 1400

0.1

0.2

100 110 120 130 1400

0.1

0.2

110

140

120

130

100

surface temperature

temperature [oC]

volu

me

fract

ion

volu

me

fract

ion

0 2 4 6 8 10 12 14 16 180

200

400

600

800

without shutdownshutdown functioned


Summary

38

1. Introduction to the NREL’s MSMD model

• The MSMD model is a modularized multiphysics multiscale lithium battery model framework

2. Model application to large Li-ion battery performance

• The model enhances understanding of interactions among varied scale physics beyond what’s possible with experimentally measurable quantities only

• Thermal/electrical design variation of a cell impacts internal battery kinetics

3. Model application to large Li-ion battery degradation

• Internal imbalance of cell use grows continually throughout life

4. Model application to large Li-ion battery safety

• Cell heating pattern is affected by cell characteristics (e.g. Ah, rate)


Acknowledgments

39

Vehicle Technology Program at DOE• Dave HowellCAEBAT operation at DOE• Stephen Goguen, Brian Cunningham

Thank you for your attention!

Date post:	17-Apr-2018
Category:	Documents
Upload:	dohanh
View:	216 times
Download:	3 times

Integrated Lithium-Ion Battery Model Encompassing Multi ... · Major Activities and...

Documents