Electrochemical Pathways Towards Sustainability Donald R. Sadoway Department of Materials Science & Engineering Massachusetts Institute of Technology Cambridge, MA 02139-4307 U.S.A.

1

outline of this morning’s talk  the energy storage landscape  innovation in energy storage  electrometallurgical approach for stationary storage applications  innovation in metals extraction  electrochemical approach to zero-emissions smelting

outline of this morning’s talk  the energy storage landscape  innovation in energy storage  electrometallurgical approach for stationary storage applications  innovation in metals extraction  electrochemical approach to zero-emissions smelting

misconceptions about batteries ๏ not much has changed: not true!

electrochemistry and energy storage: noble origins

5

electrical energy storage (Wh/kg)

(MJ/kg)

lead acid

35

0.13

NiCd

45

0.16

NaS

80

0.28

NiMH

90

0.32

Li ion

150

gasoline

12000



0.54 43

misconceptions about batteries ๏ not much has changed: not true! ๏ no Moore’s Law (transistor count doubles every 2 years) ๏ all microelectronics are silicon-based ๏ all new batteries are based on entirely new chemistries  radical innovation

different approaches for different applications ๏ don’t pay for attributes you don’t need ๏ cell phone needs to be idiot-proof ๏ car needs to be crashworthy ๏ how about service temperature? human contact? ๏ stationary batteries: more freedom in choice of chemistry but very low price point

market price points application

price point

laptop computer

$2,000 - $3,000 / kWh

communications

$1,000 / kWh

automobile traction

$250 / kWh

stationary storage

$100 / kWh

severity of service conditions

price

storage is the key enabler ๏ for deployment of renewables: intermittency obstructs contribution to baseload ๏ for load leveling, load following, frequency regulation, off-peak capture: colossal battery ๏ for grid-level storage, battery vs combustion  need to think differently ๏ today’s Li-ion batteries fail badly  the whole is less than the sum of its parts:

 plinergy ๏ confine chemistry to earth-abundant elements

 to make it dirt-cheap, make it out of dirt

outline of this morning’s talk  the energy storage landscape  innovation in energy storage  electrometallurgical approach for stationary storage applications  innovation in metals extraction  electrochemical approach to zero-emissions smelting

how to think about inventing a colossal yet cheap battery ๏ look at the economy of scale of modern electrometallurgy:  aluminium smelter  bauxite, carbon, 13 kWh electricity, $5000/tonne capital cost

 metal cost < 50¢/lb

a modern aluminium smelter 1886

Charles Martin Hall, USA Paul L.T. Héroult, France

15 m × 3 m × 1 km × 0.8

−2 A⋅cm

13

how to think about inventing a colossal yet cheap battery: pose the right question  start with a giant current sink convert this…

aluminium potline 350,000 A, 4 V

…into this

why is an aluminium cell not a battery?

frozen bath

960°C

 produce liquid15 metals at both electrodes

work started 5 years ago with internal funding from the Deshpande Center and the Chesonis Family Foundation

liquid metal battery

refractory lining

on discharge

liquid metal battery

refracto

on discharge Mg(liquid) !

liquid metal battery

2 Mg +

+ 2 e-

䎪

refracto

on discharge Mg(liquid) !

2 Mg +

+ 2 e-

Mg2+ + 2 e- ! Mg(liquid alloy)

liquid metal battery

䎪

refracto

䎩

our sponsors

$4 million $7 million

laboratory-scale test cell

1 Ah “shotglass” 21

cell section after cycling 48 h at 700°C

electropositive anode

1 Ah “shotglass”

molten salt electrolyte

electronegative cathode

23

“hockey puck”

“personal pizza” 24

cycle testing of cell 11 (20 Ah) Cell

Current density

Cycles

Cycles analyzed

mA / cm2 11

250

100

10

Columbic efficiency

Energy efficiency

Fade rate

Capacity density

Utilization

Electrode cost

%

%

% / cycle

Ah/cm2

%

$ / kWh

99

67

0

0.6

77

90

Reason for decommission

Test complete

25

attributes of all-liquid battery  all-liquid construction eliminates any reliance on solid-state diffusion

 long service life  liquid-liquid interfaces are kinetically the fastest in all of electrochemistry

 capable of handling high currents  all-liquid configuration is self-assembling  expected to be scalable at low cost 26

???

Liquid Metal Battery

27

LMB status report  liquid metal battery works:  almost 400 cells tested  many chemistries: alloys and salts  capacity fade as low as 0.05% / cycle  accelerating scale-up to self-heating cell  startup company  Liquid Metal Battery Corp.

towards commercialization LIQUID METAL BATTERY CORPORATION ! 

founded 2010

! 

series A: Bill Gates & TOTAL

䘛  patient investors 䘛  significant ability to support subsequent

capital intensive investment ! 

focus on commercialization & scale-up

outline of this morning’s talk  the energy storage landscape  innovation in energy storage  electrometallurgical approach for stationary storage applications  innovation in metals extraction  electrochemical approach to zero-emissions smelting

problems with metals extraction  unfavorable by-products L 

2

steelmaking makes CO2  2 FeO + C = 2 Fe + CO2

(½ kg C / kg Fe) x 1.8 billion tonnes



sundry HAPs including Mn & Pb, polycyclic organics, benzene, & CS2

31

why is metal production so dirty?

 many processes are over 100 years old r attitude then of indifference towards the environment

2

32

where do metals come from?  occur naturally as compounds  beneficiated  high-purity feed  reducing agents: H, C, M, e options for sustainability? 2

33

where do metals come from?  occur naturally as compounds  beneficiated  high-purity feed  reducing agents: H, C, M, e options for sustainability? 2

34

beyond the blast furnace  most metals are found in nature as oxides  “like dissolves like”  e- is the best reducing agent

 molten oxide electrolysis:

extreme form of molten salt electrolysis where pure oxygen gas is the by-product

MMMMM 2

35

replace C with

e:

reductant and fuel

๏ electrolytic route from ore to liquid metal viable at industrial scale: aluminium worldwide capacity exceeds 45 million tpy 1886 Charles Martin Hall, USA Paul L.T. Héroult, France

๏ decompose Al2O3 dissolved in Na3AlF6 (T = 960°C)  liquid Al (-) and CO2 (+) 3  find an inert anode & molten oxide electrolyte

36

molten oxide electrolysis (MOE) (FeOx ) = Fe(l) + x 2 O2 (g) NOT TO SCALE

๏ temperature above 1538°C ๏ current flow generates heat by Joule effect

iron

๏ carbon-free iron product in the liquid state ๏ oxygen by-product:  environmentally beneficial  commercial value liquid iron

๏continuous process:  periodic feeding of iron oxide  periodic removal of liquid iron

3

37

attributes of MOE (1) ๏ extraction is carbon-free  no emission of CO2, SO2, NOx ๏ cell operates at 1600°C  production of molten steel in a single reactor ๏ iron oxide fed directly into the cell  fewer unit operations  lower cost ๏ tonnage oxygen also produced  marketable by-product coke oven sintering

blast furnace

molten oxide electrolysis

2

basic oxygen furnace

refining, casting, rolling, shaping

refining, casting, rolling, shaping

38

T = 1600°C

\ 39

+

anode lead: making oxygen

cathode collector: making liquid iron

electrolysis of Fe2O3 at 1570°C as seen through port in cell cap 40

constant-current electrolysis at 1575°C current density: ~1 A cm-2

Mo crucible

electrolyte

iron

41

more electrolytic production of molten iron:

iron

42

producing oxygen on an inert anode

anode after 2.5 h electrolysis at 1.5 A.cm-2, T = 1565°C

5 mm

frozen slag

oxide layer metallic alloy core

 point defect model (D.D. MacDonald) 3

43

next step: internally heated cell ๏ MOE industrial cell will be self-heated by the Joule effect

Radiation! (-)!

Natural convection! (-)!

Joule effect! (+)!

Reaction heat! (-)!

๏ energy efficiency and metal purity can be assessed only in an internally heated cell 3

44

notional design of self-heating cell anode ø 40 cm

NOT TO SCALE

slag ø 4 cm cathode ø 50 cm

3

45

other attributes of MOE ironmaking

† supply chain for iron oxide feed uses existing ๏

๏ produces metal of superior quality in liquid state (no carbon, sulfur, nitrogen, or hydrogen) ๏ lower threshold tonnage at lower capital cost ๏ zero carbon emissions from smelter ๏ potential to produce high-quality steels, e.g., stainless

2

46

production of nickel by MOE

metal ball at bottom of cathode 2

metal ball on floor of cell 47

production of ferrochromium by MOE

2

48

towards electrolytic stainless steel Fe-Ni-Cr alloy

production of liquid titanium by MOE Mo crucible frozen electrolyte

T = 1725°C

titanium puddle

cathode: Mo anode: C current density ∼1 A/cm2 50

production of rare-earth metals by MOE?

stay tuned!

51

electrochemistry and energy storage: noble origins bright future

52

electrochemistry and energy storage: noble origins bright future

Ernest Rutherford 53