Towards Elimination of the Anode Effect and Perfluorocarbon Emissions

from Hall-Héroult Cells

Hongmin Zhu & Donald R. Sadoway

Department of Materials Science & EngineeringMassachusetts Institute of Technology

Cambridge, Massachusetts U.S.A.

Sadoway, MIT ECS Meeting, Philadelphia, May 2002

background

p on anode effect Hall-Héroult cell produces CF4 and C2F6 (PFCs)

p PFCs have high GWP

p in US, Al smelting is #1 point source of PFCs

Sadoway, MIT ECS Meeting, Philadelphia, May 2002

premise of the current work:

understanding the anode effect

+ understanding PFC generation

+ plan for reducing emissions

Sadoway, MIT ECS Meeting, Philadelphia, May 2002

outline of today’s talk:

p electroanalytical studies p gas analysis during electrolysis

in laboratory cellsp method to avert AE and PFC

generation in lab cellsp extension to industrial cells

Sadoway, MIT ECS Meeting, Philadelphia, May 2002

results of studies in lab cells

p anodic reaction inhibited when potential exceeds a critical value,EC ≈ 3.0 - 3.5 V

p anode becomes active again when potential is reduced below a so-called recovery value, ER ≈ 2.8 V

p PFCs generated only at extreme potentials, i.e., E > 4.0 V

Sadoway, MIT ECS Meeting, Philadelphia, May 2002

results of studies in lab cells

+ AE linked to formation of highly resistive surface film

Sadoway, MIT ECS Meeting, Philadelphia, May 2002

our experiments:

3-electrode cell featuringO rotating rod anode (WE)O reference electrodeO large counter electrode

electrolysis cell featuringO tubular anode for

improved gas sampling

Sadoway, MIT ECS Meeting, Philadelphia, May 2002

cell for electrokinetic studies

cell cover

mullite reaction tubecounter electrodereference electrode

working electrode

furnace

crucible

electrode

BN sheath

thermocouplegas or vacuum

Sadoway, MIT ECS Meeting, Philadelphia, May 2002

various types of electrodes

disk rodtubular

BN sheath

electrode surface

BN sheath

electrode surface

Sadoway, MIT ECS Meeting, Philadelphia, May 2002

evidence of a passivating film84% Na3AlF6 - 11% AlF3 - 5% CaF2

E / V vs Al / AlF3

I / m

Acm

-2

-500

0

500

1000

1500

2000

2500

= 100 mV s-1v

0 RPM200 RPM400 RPM600 RPM

1% Al2O3

1.0 2.0 3.0 4.0 5.0 6.0

T=970ºC

CV on graphite RDE

Sadoway, MIT ECS Meeting, Philadelphia, May 2002

evidence of a passivating filmI /

mA

cm-2

-500

0

500

1000

1500

2000

2500

1.0 2.0 3.0 4.0 5.0 6.0

0 RPM200 RPM400 RPM600 RPM

= 100 mV s-1v

84% Na3AlF6 - 11% AlF3 - 5% CaF2

E / V vs Al / AlF3

T=970ºC

CV on graphite RDE

1% Al2O3

Sadoway, MIT ECS Meeting, Philadelphia, May 2002

evidence of a passivating film

0

400

800

1200

1600

0 2 4 6 8 10 12 14

average current duringa potential step,tubular electrodelinear sweep, rod electrode

I / m

Acm

-2

E / V vs Al / AlF3

T=970ºC

Sadoway, MIT ECS Meeting, Philadelphia, May 2002

evidence of film formation

t / s

-100

0

100

200

300

400

500

600

0 5 10 15 20 25 30 35 40

I / m

Acm

-2

0.85 V

2.0 V

3.0 V

3.5 V 0.85 V

potentiostatic electrolysis

Sadoway, MIT ECS Meeting, Philadelphia, May 2002

evidence of film formation

-100

0

100

200

300

400

500

600

0.0

2.0

4.0

6.0

0 20 40 60 80 100 120

I EI /

mA

cm-2

E / V vsAl / AlF

3

t / s

Sadoway, MIT ECS Meeting, Philadelphia, May 2002

cell for analysis of anode gas

cell cover

mullite reaction tubecounter electrodereference electrode

working electrode

furnace

crucible

BN sheath

thermocouplegas or vacuum

electrode

Sadoway, MIT ECS Meeting, Philadelphia, May 2002

controlled-potential electrolysis

04080

120160200240280

1.5 2.0 2.5 3.0 3.5 4.0

CC

F4/ p

pm

E / V vs Al / AlF3

Sadoway, MIT ECS Meeting, Philadelphia, May 2002

controlled-potential electrolysis

0.00.51.01.52.02.53.03.54.0

2.0 4.0 6.0 8.0 10.0 12.0 14.0 16.0

log

(CC

F4/ p

pm)

E / V vs Al / AlF3

Sadoway, MIT ECS Meeting, Philadelphia, May 2002

controlled-potential electrolysis

in the high-voltage regimeln c = α + β E,

where β = 0.325 V–1

c.f. Nissen & Sadoway (1997)r CF4 = a exp (b E ),

where b = 0.331 V–1

Sadoway, MIT ECS Meeting, Philadelphia, May 2002

data from an industrial cell

0

5

10

15

20

25

30

0 10000 20000 30000 40000 50000 60000

cell

volta

ge /

V

t / s

anode effect

Sadoway, MIT ECS Meeting, Philadelphia, May 2002

a closer look at an anode effect

4.5

5.0

5.5

6.0

800 810 820 830 840 850

cell

volta

ge /

V

t / s

5.0 V, 10 s

4.9 V, 23 s

Sadoway, MIT ECS Meeting, Philadelphia, May 2002

data from a laboratory cell

200

400

600

800

10000 500 1000 1500 2000 2500 3000 3500 4000

I / m

At / s

800 mA cm-2

galvanostatic electrolysis0.6 wt% Al2O3

Sadoway, MIT ECS Meeting, Philadelphia, May 2002

data from a laboratory cell

2.0

4.0

6.0

8.0

10.0

12.0

14.0

16.0

0 500 1000 1500 2000 2500 3000 3500 4000

E/ V

vs

Al /

AlF

3

t / s

galvanostatic electrolysis0.6 wt% Al2O3

Sadoway, MIT ECS Meeting, Philadelphia, May 2002

data from a laboratory cell

2.0

2.5

3.0

3.5

0 500 1000 1500 2000 2500 3000 3500

9oc26jj, 9nv04ca, ha

700 mA600 mA466 mA

E / V

vs

Al /

AlF 3

t / s

s = 0.58 cm 2

Sadoway, MIT ECS Meeting, Philadelphia, May 2002

AE averted by current stepping

2.4

2.6

2.8

3.0

3.2

3.4

600

700

800

900

0 500 1000 1500 2000

I / mA

t / s

current

potentialE / V

vs

Al /

AlF 3

Sadoway, MIT ECS Meeting, Philadelphia, May 2002

major findings:

onset of anode effect is potential dependent

AE can be averted by reducing current so as to prevent voltage from exceeding a critical value

Sadoway, MIT ECS Meeting, Philadelphia, May 2002

in an industrial cell

accurate voltage sensing

robust current shunting

control circuit

enabling technologies:

Sadoway, MIT ECS Meeting, Philadelphia, May 2002

vacuum current breaker

Sadoway, MIT ECS Meeting, Philadelphia, May 2002

our proposal

shunting ≈1% cell current drops cell voltage for ≈1 min∴ 300 kA cell 3 kA/min

no AE no PFCs

150di Asdt

−=

+ no ANODE EFFECT!!!

Sadoway, MIT ECS Meeting, Philadelphia, May 2002

acknowledgments

The Aluminum Association

U.S. Environmental Protection Agency,Climate Protection Division