Aluminium Electrochemistry in

Electrocoagulation Reactors

Martin Mechelhoff
Imperial College London
Department of Chemical Engineering
London SW7 2AZ (UK)

213th ECS Meeting, Phoenix, 18th – 23th May 2008

 Conventional Coagulation
raw water coagulant (AlIII or FeIII) + pH adjustment

treated water outlet

sludge outlet sludge outlet

• Advantage: well known process

• Disadvantages: large footprint
large amounts of liquid chemicals
performance exhausted
2
Imperial College London 2008
 Electrocoagulation Reactors
• Water for purification passed through
electrochemical reactor

• Anode usually Fe or Al; cathode often steel

• Inter-electrode gap: < 5 mm

• Dissolution, hydrolysis and precipitation of Power Supply

metal ions from electrode

• Simultaneous evolution of H2 gas

H2O
• Destabilisation of suspension → coagulation

Main advantage: H2 + OH -
Me+
No addition of chemicals
Anode Cathode
3
Imperial College London 2008
 Performance and Challenges
Comparison Chemical dosing - Electrocoagulation
at inlet pH 7.8
100
90
80
70
removal [%]

60
50
40
30
20
10
0
DOC colour (UV254) Jiang, J.Q.; Graham, N.
et al. (2002); Water
chemical dosing electrocoagulation
Research, 36, 4064-4078
But:
• High cell voltages, thus high specific energy consumption
• Electrode passivation → slow reaction rates
4
Imperial College London 2008
 Potential-pH diagram for Al-H2O system
for 10-6 M Al(III) activity at 298 K
1.4
1.2
1.0
0.8
electrode potential E

Al(OH) 3 (am)
O2
0.6
0.4
Electode Potential (SHE) / V

+
0.2 H H2O
0.0
-0.2 H2
-0.4
-0.6
-0.8 overpotential η
-1.0
-1.2 Al3+
-1.4
Al2O3 (c) Al(OH)4-
-1.6
-1.8
-2.0
-2.2 Al
-2.4
-2.6
0 1 2 3 4 5 6 7 8 9 10 11 12 13 14
pH

5
Imperial College London 2008
 Electrode Reactions
Power
supply

Al → Al 3+ + 3e − Al Anode + _ Inert
2 H 2O + 2e − → H 2 + 2OH −
Cathode
(O → 4OH − )
+
2 Al + 3 H 2 O ⎯⎯
→ Al2 O3 + 6 H + 6e −

2 + 2 H 2O + 4e − ⎯⎯

( Al + 3H O ⎯⎯→ Al ( OH )
2 3
+ 3H + + 3e − )
Precipitation:

H2 bubbles Al 3+ + 3H 2O ⎯⎯
→ Al (OH )3 + 3H +

Water flow

6
Imperial College London 2008
 Voltammetric Behaviour of Al
• Passivation impedes dissolution

1
0.5 2 H 2O + 2e − → H 2 + 2OH −
0
-0.5
log (j/A m -2)

-1
+
2 Al + 3 H 2 O ⎯⎯
→ Al2 O3 + 6 H + 6e −
-1.5
-2 High purity Al RDE
-2.5 ( Al + 3H O ⎯⎯→ Al ( OH )
2 3
+
+ 3H + 3e −
) mirror finish
-3
2000 rpm
-3.5
5x10-4 kmol m-3
-4
Na2SO4
-2.3 -2 -1.7 -1.4 -1.1 -0.8 -0.5 -0.2 0.1 0.4
1 mV s-1
electrode potential E(SCE) / V

7
Imperial College London 2008
 Impedance Spectrum
• Three overlapping, slightly depressed semi-circles
• -Z’’(max) of second semi-circle increases with growing potential

2.5

2.0
0.5 Hz
-Z'' / ohm m2

1.5

1.0

0.5
104 – ca. 10-2 Hz
1 Hz
Stationary
1 Hz
0.0 Al1050 (99.5%)
0 1 2 3 4
Z' [ohm m2] 5x10-4 M Na2SO4 +
10 mgl-1 humic acid +
-1.0 V -0.6 V -0.2 V 1.5x10-4 mM NaCl

8
Imperial College London 2008
 Effect of Potential on Maximum –Z’’
• -Z’’(max) linearly dependent on applied potential
• Lower values when adding humic acid and NaCl
Æ interference with surface properties, ε=?

1.6
2
R = 0.994
Na2SO4 1.4

1.2
2
-Z'' / ohm m

2 1.0
R = 0.9559
0.8
Na2SO4 + humic acid
0.6
2
R = 0.9593
0.4
Na2SO4 + humic
acid + NaCl 0.2
0.0
-1.2 -1 -0.8 -0.6 -0.4 -0.2 0
electrode potential (SCE) / V

9
Imperial College London 2008
 Passive Layer Thickness
• Equivalent electrical circuit used for data fitting:
R1 R2 R3

Q2

Q1

3.0E-09
2
R = 0.9991
layer thickness d / m

2.5E-09

2.0E-09

1.5E-09

1.0E-09
d = 1.9 nm V −1 ⋅η
5.0E-10

0.0E+00 5x10-4 kmol m-3

-1.2 -1 -0.8 -0.6 -0.4 -0.2 0 Na2SO4

Electrode potential (SCE) / V With ε = 9

10
Imperial College London 2008
 Cell Voltage at Constant Current Density
At Yorkshire Water’s electrocoagulation pilot plant:
• Constant current density applied: ca. 67 A m-2
Æ Rapid cell voltage increase due to passive layer growth / fouling
400

350

300
cell voltage / V

250

200

150

100

50

0
0 50 100 150 200 250 300
time / min

11
Imperial College London 2008
 Characterisation of Electrodes
• Objective:

– Investigate factors controlling passivation behaviour

Æ Define means of diminishing passivation

• Procedure:

– Use of ca. 30 mm2 plate electrodes with various degrees of

surface roughness

– Measurement of electrode potential as function of time in

response to applied current density
12
Imperial College London 2008
 Effect of Surface Roughness
12 Limit of
potentiostat
electrode potential / V(SCE)

10

0
0 50 100 150 200 Rough Al electrode
time / s
1.4

1.2

electrode potential E / V(SCE)

Smooth Al electrode
1

0.8

0.6

0.4

0.2
Solution without Cl- 0
0 100 200 300 400 500 600
Fresh surface
time / s
13
Imperial College London 2008
 Effect of Storage Time in Solution
1.4

1.2
electrode potential E / V(SCE)

0.8

0.6

0.4

0.2

0
After 72 hours in solution
0 100 200 300 400 500 5
600
time / s
4.5

Fresh surface electrode potential / V(SCE) 3.5

2.5

1.5
1

Rough electrode 0.5

0
Solution without Cl- 0 100 200 300 400 500 600
time / s
14
Imperial College London 2008
 Effect of Chloride Ions
12
electrode potential / V(SCE)

10

0 With Cl-
0 50 100 150 200
1.4
time / s
1.2

electrode potential / V(SCE)

1
Without Cl-
0.8

0.6

0.4

0.2

-0.2
Smooth electrode
-0.4

Fresh surface 0 100 200 300 400 500 600

time / s
15
Imperial College London 2008
 Impedance Spectrum after Corrosion
• Initially, high -Z’’(max)-value Æ thick passive layer
• Significantly reduced after constant current Æ layer thinned

20

16 before
2
-Z'' / ohm m

12 0.05 Hz

104 – 0.05 Hz
8 0.2 Hz
Stationary
4 High purity Al
after
Fresh electrode
0
0 4 8 12 16 20 5x10-4 kmol l-1
2 Na2SO4 + 10 g m-3
Z' / ohm m
humic acid + 1.5x10-4
kmol l-1 NaCl
16
Imperial College London 2008
 Effect of Real Surface Area of Al Electrode
• AFM image of ‘mirror finish’ surface:

• “Micro-roughness” may create cracks in passive layer

17
Imperial College London 2008
 Finite Element Model of Electrochemical Reactor
Reactor design and reactions included:
Out-flow

0.01 m
Dissolution of Al: Production of H2:
Anode: Cathode:
Al → Al 3+
+ 3e −
Al inert 2 H 2 O + 2e − → H 2 + 2OH −
0.14 m

Homogeneous precipitation
of Al3+ in two steps:
0.19 m 3+ 2+ +
0.002 1 Al + H 2O U Al (OH ) + H
m

2 Al (OH ) 2+ + 2 H 2O U Al (OH )3 + 2 H +

In-flow: c(Al3+)=0, pH 7

18
Imperial College London 2008
 Model Prediction of pH Profile
• Anode Potential +0.05 V(SCE)
• Resultant current density ca. 6.8 A m-2

Al Anode
Cathode
(inert)

fluid flow

inter-electrode gap
19
Imperial College London 2008
 Effect of pH Profile on Dissolution
• Low local pH at Al anode dissolves oxide layer
1.4
1.2
1.0
0.8

Al(OH) 3 (am)
O2
0.6
0.4
Electode Potential (SHE) / V

+
0.2 H H2O
0.0
-0.2 H2
-0.4
-0.6
-0.8
-1.0
-1.2 Al3+
-1.4
Al2O3 (c) Al(OH)4-
-1.6
-1.8
-2.0
-2.2 Al
-2.4
-2.6
0 1 2 3 4 5 6 7 8 9 10 11 12 13 14
pH

pH at anode
20
Imperial College London 2008
 Conclusions
• Dissolution of Al anode enabled by low local pH due to Al3+
hydrolysis generating H+ and precipitating Al(OH)3

• This leads to local dissolution and thinning of passive layer

Æ spontaneous depassivation

• Rough and freshly prepared surfaces more likely to exhibit

spontaneous depassivation

• Corrosion promoters, e.g. Cl-, enhance depassivation

21
Imperial College London 2008
 Thanks
• Yorkshire Water, Bradford (UK), for financial support

• Prof. Geoff Kelsall (Imperial College, Dep. of Chemical

Engineering)

• Prof. Nigel Graham (Imperial College, Dep. of Civil and

Environmental Engineering)

22
Imperial College London 2008
 Questions?

23
Imperial College London 2008
 24
Imperial College London 2008
 High Frequencies
• Frequency range 104 – 50 Hz

0.08

0.06
-Z'' / ohm m2

0.04

0.02

IR drop
0.00
0.00 0.03 0.06 0.09 0.12 0.15 0.18
Z' / ohm/m2

-1.0 V -0.6 V -0.2 V

25
Imperial College London 2008
 Full Spectrum
• Frequency range 104 – 10-3 Hz

16

14

12
-Z'' / ohm m2

10

0
0 5 10 15 20 25 30
Z' / ohm/m2

-1.0 V -0.6 V -0.2 V -0.2 V fitted

26
Imperial College London 2008