Rechargeable Batteries, Old and NewJohn B. Goodenough

The University of Texas at Austin

Modern Society is enabled by the chemical energy stored in fossil fuels.Unsustainable Cost• National vulnerabilities from uneven, finite

distribution• Spoiled environment by extraction• Air pollution and global warming by use

Sustainable Energy Supply

• Generation of electrical energy by wind, solar, nuclear, and other sources

• Storage of electrical energy at an acceptable cost

• What is the status of rechargeable batteries? Electrochemical capacitors?

Battery Principle

Rechargeable Electrochemical Cell: Pcharge > Pdischarge Battery=Stack of cells connected in series to increase V, in 

parallel (or electrode area) to increase I=dq/dt and/or Δt. Challenges: Cost, Safety, Energy Density, Life, Rate.

Chemical Energy(in electrodes)

Electrical Energy(P=IV for Δt)

I, V

Δt

Traditional Rechargeable Cells

AnodeCd0 + 2H2O Cd(OH)2 + 2 H+ + 2e‐

Pb0 + H2SO4 PbSO4 + 2H+ + 2e‐

CathodeNiOOH + H+ + e‐ Ni(OH)2PbO2 + H2SO4 + 2H+ + 2e‐ PbSO4 + 2H2O

Reductant Oxidant

Anode CathodeElectrolyte

‐ +Separator

e‐e‐

M2+

H+

NiO6/3 layer

NiOOH

H+ H+

KOH

H2SO4

H2SO4

KOH

Vdis

Vocdis

I

(i) (ii) (iii)

Vdis = Voc − ηdis(I,q) Relectrolyte = (dV/dI)ii

Vch = Voc + ηch(I,q)

TYPICAL BATTERY DISCHARGE

V

t or q

VV

t or q

Two Phase

η

V

t or q

VOC

Diffusion‐limited (end of life)

VV

t or q

η

(i) (ii) (iii)

VOC

Cell Energy at Constant I = dq/dt

V(q) = VOC – η(q, I); VOC = (μA – μC)/e ≤ electrolyte Eg

η(q, I) increases with resistance to working-ion current

Problem: Maximize VOC, Q(I) and # cycles to Q/Q0 = 0.8; minimize η

Q(I)/wt or vol = specific or volumetric capacity

Energy = ; ∆ ∆

Electrolyte

Aqueous Electrolyte: Eg = 1.23 eV• LUMO = H2O/H2; HOMO = O2/H2O

Kinetic Stability• NiOOH/KOH/Cdo Voc = 1.5 V

Limited Life• PbO2/H2SO4/Pbo Voc = 2.0 V

LUMO

HOMO

1.1 eV

0.3 eV

4 eVoc

EF(Li)

3.2 eV

Electrolyte, Separator LixCLi1‐xCoO2

E

O2‐:2p6

Co4+/Co3+

S2‐:3p6

2.6 eV

First Li-ion Battery

LiCoO2//C Cell

SEI layer

Separator

LixC6Li1‐xCoO2

Lix[M2]O4 Spinel Electrodes

a0

8a16c

16dO2‐

2 quadrants of structureEdge‐shared MO6/5 octahedra3D Li+ insertion into close‐packed oxygen array 

Note: Li1+x[Li1/3Ti5/3]O4: V=1.5 VLi1‐x[Ni0.5Mn1.5]O4: V=4.7 VNi(II)  Ni(IV) with little step

Lix[Mn2]O4

NASICON Framework

NASICON

Na1+3xZr2(P1-xSixO4)3

LixFe2(SO4)3: V =3.6 V

LixFe2(MO4)3: M = Mo, W: V =3.0 V

Na3-xV2(PO4)3: V = 3.4 V

c

a

Capacity Challenge(Limited specific capacity further reduced by anode

SEI layer)

0 50 100 150 2001.5

2.0

2.5

3.0

3.5

4.0

4.5

5.0

Pot

entia

l (V

) vs.

Li+ /L

i

Capacity (mAh/g)0 50 100 150 200 250 300

2.0

2.5

3.0

3.5

4.0

4.5

5.0

Pot

entia

l (V

) vs.

Li+ /L

i

Capacity (mAh/g)

Li1‐xCoO2

Li1‐xFePO4

LixTiS2& Li2[Ti2]S4

Lix[Mn2]O4

Li1‐x[Ni0.5Mn1.5]O4

MO2

LiMO2

Li MO2

Li MO2

ELECTRODE MATERIALS

LUMO

Na2MnFe(CN)6 and Na3V2(PO4)3 : V 3.4 V, Q(10C) > 100 mAh/g

Framework of Prussian Blue Analogues

Anode Problem with Organic Liquid Electrolyte

• An EF (Anode) > LUMO requires an SEI• SEI on Li0 or Na0 creates dendrites, so C

anode• Need VCh < Vplate, so C-buffered alloys• Need SEI permeable to Li+ or Na+: if Li+,

Na+ come from cathode, get capacity loss on initial charge.

• Reforming SEI gives capacity fade limiting cycle life

• SEI slows Li+ or Na+ transfer

Separators

• Celgard membrane: penetrated by dendrites; insertion-compound cathode

• Polymer-gel membranes: blocks dendrites; adds choice of liquid flow-through cathode

• Anode/Solid-electrolyte interface: prevents dendrites, allows choice of sulfur or liquid flow-through and air cathode if solid electrolyte stable in alkaline water

Al2O3/PEO Separator(K.S. Park, J.-H. Cho, C.J. Ellison

Oxide: Dried (400C) Al2O3 Powder (300-400nm)Polymer: DEGDVE = Di(ethyleneglycol) divnylether with ethylne-

oxide (EO) units PETT = tetrathiol crosslinker

Preparation: AIBN = thermal initiator (80C, 4h)Yield: Quantitative, homogeneous network

AIBN = Azobisisobutyrolnitrile

Photographs of a PEO/Al2O3 (2/1 in weight) composite membrane (25  20 cm2)

PEO/Al2O3 Composite-Membrane SeparatorK. Park et al.

Mechanical Stability

• Tensile test shows a good mechanical stability, especially after Al2O3incorporation.

• Stability against Li dendrite

Minor surface dent by Li dendriteAfter full charging

It blocks Li dendrite.

Presence of Osmosis• Presence of concentration gradient of a redox-molecule across the PEO/Al2O3

membrane

• Balancing the concentrations between catholyte and anolyte is necessary, for example, with high-molecular-weight Ionic liquids and PEGDME.

Charge/Discharge Cycle Property

Anolyte: 1M LiTFSI in EC.DEC w/0.1M PEGDME 500Catholyte: 1M LiTFSI in EC/DEC w/0.1M 6-Bromohexyl ferrocene

0 10 20 30 40 500

40

80

120 0.5 C2 C1 C0.5 C

GF/PVDF-HFP/PDA GF/PVDF-HFP Glass fiber

Spe

cific

Cap

acity

(mA

h g-1

)

Cycle number

0.2 C

PVDF‐HFP: Widely used as host for gel‐polymerelectrolytesBut with unsatisfactory mechanical strengthGlass‐fiber paper: Used to enhance the mechanical properties

Polydopamine: Biomimetic polymer of mussel adhesive protein Polymerized at room temperature in aqueous solutionModify the surface properties of PVDF‐HFP

Gel‐polymer electrolytes: Improve rate and cycle performance of air‐dried cathode of Na2MnFe(CN)6The composite membranes have good thermal 

stability and mechanical strength. Hongcai Gao, et al. Adv Energy Mater 2015

Current collector

Separator membrane

Solid electrolyte

cathode

Current collector

Separator membrane

cathode

Li+/Na+

Strategy for an Alkali-Metal Anode

Cost Targets• Cycle Life N > 10,000; calendar life

10 years

• Simple processing of inexpensive materials.

• Increased VQ(I) to reduce number of cells.

• Simplify battery management

Lithium-Sulfur Batteries