CET IIA Corrosion and Materials

Lectures 1-8: Corrosion

Michaelmas term 2017
Dr Eric J. Rees

Outline of Corrosion Lectures

1. Corrosion and material selection
2. Thermodynamics of aqueous corrosion
3. Rates of aqueous corrosion
4. Localised corrosion mechanisms and corrosion testing
5. Corrosion protection
6. High-temperature oxidation of metals

Recommended Textbooks
The following textbooks are written at an appropriate level for this course
1. M.G. Fontana, “Corrosion Engineering” (3rd ed., 1986), McGraw-Hill
2. K.R. Trethewey and J. Chamberlain, “Corrosion for Science and Engineering”
(2nd ed., 1995), Longman
3. D.A. Jones, “Principles and Prevention of Corrosion” (1995), Prentice Hall
4. E. Mattsson, “Basic Corrosion Technology for Scientists and Engineers” (2nd
ed., 1996), Institute of Materials
5. P.R. Roberge, “Handbook of Corrosion Engineering” (2nd ed., 2012; or 1st ed.,
2008), McGraw-Hill
6. A.C. Fisher, “Electrode Dynamics” (1996) OUP
7. Z. Ahmad, “Principles of Corrosion Engineering” (2006), Elsevier
Fontana and Tretheway have been recommended by the department for many years,
and should be easy to find in college libraries.
Online course material at MSM: http://www.doitpoms.ac.uk/tlplib/aqueous_corrosion/

Further Reading and References

1. R.W. Revie “Uhlig’s Corrosion Handbook” (2nd ed., 2000), Wiley.
2. “Perry’s Chemical Engineers’ Handbook” (7th ed., 1997), McGraw-Hill.
3. L.L. Shreir, G.T. Burstein, R.A. Jarman, Corrosion (1998), Butterworth-
Heinemann

Examples Questions
Two examples papers will be distributed during this course.
Tripos questions for reference are in CET IIA Paper 1 Q 4 to 6 (2006 - 2013); Paper 3
Q 4 to 6 (2014); Paper 3 Q 4 to 5 (2015 - 2017);
 1. Introduction to Corrosion

1.1 Material selection

The design phase of a project can involve optimising a choice of material, and
corrosion-tolerance is an essential constraint:

1.2 What is corrosion?

A modern definition of corrosion (from Fontana) is

Corrosion is the deterioration of a material by

chemical reaction with its environment.

Some examples of corrosion are

o Dissolution of metal such as zinc coatings
o Embrittlement of reinforced concrete due to tensile stress from rust
o Many localised corrosion phenomena (see section 4).
Fontana’s definition of corrosion is used in this course.
 Some authors restrict the definition of corrosion to metals [Ahmad p.2 A,F].
However, to solve real problems involving metals we often must consider
other materials.
o Cement in reinforced concrete
o Polymer or glass coatings

Corrosion processes may be harnessed for an indirect benefit.

o Etching circuit boards
o Anodisation for hard coatings
o Aesthetic improvements

Beware: different textbooks on corrosion use various different conventions for

many key aspects of the subject, not just in minor details of its definition.
 X-Y axis convention for Evans diagrams (we use log-current on the x-axis)
 Positive and negative currents (we use positive values for anode currents)
 Standard pressure. Now 105 Pa, but used to be 101325 Pa (1 atm).
Reference http://goldbook.iupac.org/S05921.html
EJR 2
1.3 Why is corrosion important?
1.3.1 Corrosion leads to huge costs…
 Replacement (cost of new materials; downtime)
 Inspection and treatment
 Decreased process efficiency (e.g. of heat exchangers)
 Loss of product through leakage. (Water utilities often lose 20% of their
supply through leaks).
 Loss of services and the need for backup.
 The annual cost of corrosion in the USA has often been estimated at
between 3-6% of GDP.
o Between 10-40% of these costs are thought to be avoidable [e.g.
Roberge, pp 15-20].

1.3.2 Failure of structures or products…

 Finite lifetime. Causing vehicles/oilrigs/products to be scrapped
 Unexpected failures.
o E.g. electrical components.
o E.g. pipeline leaks and losses.
o … which may lead to further problems: fires; loss of services…

1.3.3 Contamination…
 Contamination of product streams from
o By leaching of dissolved ions
o By leaks through perforated metal (pitting corrosion)
 Fouling of cooling water cooling due to corrosion of heat exchangers
o Corrosion inhibitors / systematic design needed

1.3.4 And poisoning

 Leaching of Ni2+ from alloys
o Ni2+ is a sensitising allergen, and nickel compounds can be toxic
o Affected Euro coins
o Justifies testing in food processing
As stainless steel is a preferred material
 Dissolution of Cu2+ into acidic solutions
o Cu2+ is toxic
EJR 3
1.3.5 Disasters…
 Over 25% of accidents in the oil and gas industry are caused by corrosion
o Skikda explosion of natural gas compression train

Unexpected corrosion mechanism

o Prudhoe Bay oil spill(s)

Insufficient monitoring

 Guadalajara sewer explosion (1992)

Incompatible corrosion protection methods

 Bridge collapses (E.g. Silver Bridge collapse in West Virginia 1967)

Combination of factors. Unexpected mechanism and a design that limited

monitoring.

1.3.6 Corrosion Engineering

Corrosion engineering is “the use of science and art to prevent or mitigate these
problems in an economic way” [Fontana]. We can:
 Predict whether corrosion will occur…
o Severely, by active corrosion,
o Slowly, due to corrosion resistance or passivity
o Not at all, because a state of immunity to corrosion exists
 Predict the rate of corrosion
 Design systems that have sufficient corrosion resistance, so that the
lifetime capital / operating costs of a process are optimised.
 Identify, measure, and treat the mechanisms of corrosion

EJR 4
1.4 Corrosion of metals in chemical engineering
Metals are often used in chemical engineering because their properties
(strength, ductility, conductivity, etc.) are essential for many applications
(structural elements, heat exchangers, etc.)
 There are essentially three ways in which a metal can respond to a
chemical environment.

1.4.1 Active corrosion

 In a particular environment, some metals may undergo continuous and
rapid deterioration called active corrosion.
 This happens if the dissolved ion is more stable that the metal phase.
 A simple example is the dissolution of zinc in aerated (oxygenated) solution
Zn → Zn2+ + 2e−
O2 + 2H2 O + 4e− → 4OH −

 Some active corrosion processes are much more complex. E.g. rust.
Fe → Fe2+ + 2e−
O2 + 2H2 O + 4e− → 4OH −

Fe2+ +2OH− → Fe(OH)2 ↓

1
2Fe(OH)2 + H2 O + 2 O2 → 2Fe(OH)3
or 2Fe(OH)2 + 2OH − → 2Fe(OH)3 + 2e−

o The iron dissolution step is similar to the simple case of zinc dissolution,
and this rate is used to quantify the speed of corrosion in this course.
 Actively corroding materials are often used, but methods to predict the
speed of corrosion (hence lifetime, or replacement costs) are important.

1.4.2 Immunity to corrosion

 Thermodynamically stable metals and alloys are immune from corrosion.
 This is rare in engineering unless thermodynamic corrosion protection
methods are used.

1.4.3 Passivity
 It is very common to select metals which are passive against corrosion.
 This means a thermodynamically stable corrosion product is solid and that
the metal reacts to produce an impermeable (hydr–)oxide surface layer
which prevents further corrosion.

EJR 5
1.5 Important metals in corrosion science
1.5.1 Zinc
Zinc is a common coating on other metals. It often dissolves actively, but slowly.
 Questions. How fast does zinc dissolve in a given environment?
 Issues. Surprisingly slow kinetics for dissolution, given the large
thermodynamic driving force. Sacrificial anodes.

1.5.2 Iron alloys and carbon steel

Widely used. Active corrosion is common. Rust, which consists of hydrated iron
(hydr-)oxides, is exclusive to iron alloys.
 Questions. How fast is active corrosion? How much does corrosion
protection cost (often by using sacrificial anodes or cathodic protection to
make the iron immune from corrosion)? How does that compare with over-
specifying the metal thickness?
 Issues. Carbon steels are classified by carbon content, but often treated as
pure iron in corrosion models. (‘Iron’ is < 0.015% carbon. Then ‘steel’ < 2%
carbon. Then ‘cast irons’ < 5% carbon.)

1.5.3 Aluminium
Aluminium is passive in neutral aqueous environments because its oxide is
thermodynamically stable and impermeable to water.
 Questions. Does passivation occur (depending on pH)? How does
passivation occur? How does corrosion affect the material properties (e.g.
does stress corrosion cracking decrease fracture toughness)?
 Passive oxide layer formation:

 Issues. Abrasion and repassivation. Acid-base chemistry. Pitting corrosion.

Liquid metal embrittlement.

1.5.4 Stainless Steels

These iron alloys contain chromium which can form a passive layer. Many
alloys exist, classified by composition and crystal structure. A common alloy is
“Type 304”, which is an 18/8 stainless steel (18% Cr, 8% Ni, max 0.08% C).
Very important in subsea oil pipelines and the process industries.
 Questions. Is an alloy passive in practice? How are problems measured?
 Issues. ‘Breakdown’ of stainless steel passivity is a huge topic. Many of the
localised corrosion mechanisms in section 4 relate to this. (Pitting and
crevice corrosion; stress corrosion cracking; welding and sensitisation due
to local chromium loss.)

EJR 6
1.5.5 Copper alloys
Copper is immune to corrosion in some environments (not including sea water,
where it dissolves slowly). Used in pipes / electronics.
 Questions. Calculate the boundary between active corrosion and immunity.
How can copper corrosion cause problems (toxicity / bimetallic corrosion)?
 Issues. Copper resists biomacrofouling due to toxicity combined with slow
dissolution.

1.5.6 Gold
Gold is usually immune from corrosion.
 Questions. Special cases when gold dissolves actively.
 Issues. Electroplating. Leaching of nickel from gold alloys can be mistaken
for a gold allergy.

1.5.7 Titanium
Passivates by forming titanium oxide, which is generally more passive than
aluminium or stainless steels.
 Questions. No mainstream questions (!) because only non-aqueous
reducing environments really cause loss of passivity, and these are
complicated and rare.
 Issues. Biocompatibility excellent. Chloride resistance. Some reducing
environments (e.g. natural gas) cause depassivation. Titanium also reacts
with chemically dry methanol and disintegrates.

EJR 7
 2. Thermodynamics of aqueous corrosion

The purpose of thermodynamics in corrosion is to determine, for a given

situation, whether a metal is thermodynamically stable and therefore immune
from corrosion, or if is there an oxidation product with a lower free energy. It is
also important to know whether the stable corrosion product is a dissolved ion
which indicates active corrosion will occur, or if is it a solid oxide (or hydroxide)
which may protect the metal and produce passivity.
In aqueous environments, corrosion science treats corrosion as two electrode
reactions. The oxidation of a metal is an anode electrode. A mixture of several
oxidising reactions may provide the cathode. A corroding electrode has an
electrical equilibrium potential at which no net reaction occurs. The metal
electrode will corrode if the mixture of oxidising species in the environment have
a more oxidising equilibrium potential than the metal.
As well as oxidation of metal, which is an electron transfer process, the stability
of solid, protective oxides with respect to dissolved ions are often determined
by pH.

2.1 Types of corrosive environment

Example question: what are some factors that speed up corrosion?

Most corrosion involves oxidation (M → M n+ + ne− ), but it is common to

distinguish between two situations
2.1.1 Chemical corrosion
Takes place in dry gases or non-ionic liquids
Chemical thermodynamics describes this process
 Oxidation of a copper in air (e.g. at 1 bara and 700 °C)
4Cu + O2 (g) → 2 Cu2 O (2.1)
o ΔGreaction indicates if this will occur.

2.1.2 Electrochemical corrosion

Occurs via electrode reactions
 Requires an ion-conducting solution (usually water).
 Electrochemical thermodynamics is used to study this process, because it
supports rate-models that chemical thermodynamics does not.
 Electrochemical corrosion is important in the natural environment and
many industrial processes.
 It happens under aqueous solution, and also in air above ~70% relative
humidity due to the formation of surface moisture.
EJR 8
2.2 Electrode notation: describing corrosion as electrochemistry
The dissolution of zinc in acid is a simple corrosion reaction
 There are only two phases: solid Zn and dissolved Zn2+. No zinc oxides or
hydroxides are stable in acid, so they are neglected.
 There are several possible chemical reactions for zinc dissolution
Zinc-protons in any acid
Zn + 2H + (aq) → Zn2+ (aq) + H2 (g) (2.2)
Zinc-oxygen in aerated acid
2Zn + O2 (aq) + 4H + (aq) → 2Zn2+ (aq) + 2H2 O (2.3)
 But the zinc experiences the same electrode process (or ‘half reaction’)
Zn(s) → Zn2+ (aq) + 2e− (2.4)
This is an anode half reaction because electrons are lost, i.e. the zinc is
oxidised.
 Equation (2.2) is referred to using the following electrode notation
Zn|Zn2+ (2.4a)
The reduced species (metal) is on the left, the oxidised species (metal ions)
is on the right, and the vertical line represents a phase boundary.
 Electrode notation is used in algebra. For example, the speed of the
reaction (per unit area of metal) might be described as
rate(Zn|Zn2+ ) = 1 mol cm−2 year (2.5)
or 𝑖(Zn|Zn2+ ) = 6 mA cm−2 (2.6)
where 𝑖 is the current density describing this rate (see section 3).

The proton or oxygen reactions can be written separately. They are cathodes,
because the species here are reduced and gain electrons.
 Hydrogen evolution (uses up protons in aqueous solution)
2H + + 2e− → H2 (2.7)
Or (H2 |H + ) which is the electrode notation (2.7a)
 Oxygen reduction (uses up dissolved oxygen molecules)
O2 + 4H + + 4e− → 2H2 O (2.8)
Or (H2 O|O2 , H + ) (2.8a)

Models of corrosion use electrochemical notation because:

 On metals, the electrode processes often takes place separately – they can
be several metres apart.
 Fewer equations are needed (only one metal anode equation). The effect
of several oxidising agents can be treated as a single term ‘the cathode’
and accounted for by a measurement using a voltmeter and a reference
electrode. Electrical currents and voltages are simple to measure.
 Electrochemical equations are compatible with kinetic models which
include ion transport.
EJR 9
2.3 Electrochemical cell notation
Aqueous corrosion can be described as an electrochemical cell, as shown
below. An electrochemical cell consists of
 One anode
 One cathode
 Electronic contact between the electrodes. An electrical potential difference
of Δ𝐸 volts might exist between the electrode surfaces. If both electrode
processes taken place on the same piece of metal, then Δ𝐸 is likely to be
zero volts.
 Ionic contact. An ion resistance may exist, 𝑅ion , but it is often neglected.

2.3.1 Diagram of an electrochemical cell

V
e-
Load

Anode Cathode

 Temperature must be given. E.g. 298 K.

 Ion activities (or concentration) must be given. E.g. 𝑎(Zn2+ ) = 1, pH = 0

This cell can be described without a diagram by the following cell notation
Zn|Zn2+ ||H+ |H2 (2.9)

 A single electrode always has the reduced species on the left, and the
oxidised species on the right. Neglect spectator ions and electrons.
Zn|Zn2+ (2.10)
H2 |H + (2.11)
 For the complete cell, the left hand side is the anode, and the right hand
side is the cathode. At each electrode, the reagent is on the left of the single
vertical line, and the product on the right. So in a complete cell the anode
is written (reduced | oxidised) cathode is written (oxidised | reduced).
 The double vertical line in equation (2.5) is a ‘salt bridge’. It represents a
distance too large for rapid mixing of the electrolyte, so that different ions
are involved at the anode and cathode.

EJR 10
2.3.2 Examples of electrochemical cell notation
 An electroplating cell is an example of a cell with no salt bridge. This cell
would need to be driven by electrical energy supplied by the ‘load’.
Cu|Cu2+ |Cu (2.12)

 A concentration cell could be used to plate copper at the cathode by using

a salt bridge to separate a high and a low concentration of dissolved ions.
Cu|Cu2+ (0.1 mol dm−3 )|| Cu2+ (1 mol dm−3 )|Cu (2.13)

 A battery with zinc dissolving at the anode and silver chloride being
reduced to silver at the cathode could be written as follows. A comma
separates species in the same phase.
Zn|Zn2+ , Cl− |AgCl|Ag (2.14)
or Zn|ZnCl2 (aq)|AgCl|Ag (2.14a)

 Sometimes it is useful to specify that the hydrogen reaction takes place on

a metal surface, such as platinum. Therefore the reaction (H2 |H+ ) is
sometimes written
Pt|H2 (g)|H + (2.15)
or Pt|H2 (g)|HCl(aq) (2.15a)

2.3.3 What is cell notation used for?

Electrochemical cell notation is used to refer to a cell in algebra.
 Electromotive force is the voltage supplied by a cell when it is working as a
battery. Consider the following descriptions in text, and in algebra.
A cell contains hydrogen gas dissolving at the anode and silver ions (Ag + )
electroplating onto silver metal at the cathode. All species have unit activity,
meaning pH=0 and 𝑎(Ag + ) = 1. The temperature is 25°C. It is found that
the potential difference between the electrodes, Δ𝐸, is 0.799 volts and that
this is an electromotive force that supplies energy to the load.
Δ𝐸(H2 |H + ||Ag + |Ag) = +0.799 V (2.16)
At pH 0, 1 bara, with a(Ag+)=1, and temperature 25°C

 Equation (2.16) describes a special situation used to measure the ‘standard

electrode potential’, E°, of the silver electrode (Ag|Ag+). [See section 2.7.]
𝐸°(Ag|Ag + ) = +0.799 VSHE (2.16a)

2.4 Name convention for anode and cathode

The anode is the site of an oxidation reaction, which loses electrons.
 A common mistake is to label one terminal of a battery as the “anode”
based on its location or material.
EJR 11
2.5 Corrosion mainly occurs at the anode of a cell
Corrosion mainly involves loss of metal by dissolution or oxide formation, e.g.
Zn → Zn2+ + 2e− (2.17)
Cu + H2 O → CuO + 2H + + 2e− (2.18)
Both of these are anode processes.
Only a few corrosion problems occur at a cathode
 Hydrogen embrittlement
 Deposition corrosion

2.6 Relating electrochemical cell voltage to the Gibbs energy of reaction

Consider the following electrochemical cell.
Zn(s)|Zn2+ (aq)|| H + (aq)|H2 (g) (2.19)
 The chemical reaction inside the cell is
Zn(s) + 2H + (aq) → Zn2+ (aq) + H2 (g) (2.19a)
 The standard Gibbs energy change per mole of zinc, assuming unit ion
activity, i.e. 1 molar effective concentration of all ions, is:
Δ𝐺°reaction (Zn|Zn2+ | |H+ |H2 ) = −147 kJ mol−1 (2.20)
 And at non-standard concentrations, the Gibbs energy change is
Δ𝐺reaction (Zn|Zn2+ | |H+ |H2 ) = Δ𝐺°reaction + 𝑅𝑇 ln(𝑄r ) (2.21)
 Where
𝑎(Zn2+ ) 𝑎H2
𝑄r = ( 2 ) (2.22)
𝑎Zn 𝑎(H +)

Here, 𝑎(Zn2+) is the Zn2+ activity, 𝑎(H+ ) is given by the pH, 𝑎H2 is the partial
pressure of H2 in bar, and the activity of any pure solid or liquid phase is unity,
so 𝑎Zn = 1. Unit activity is also used for impure water.

A purely chemical process will go forwards spontaneously if Δ𝐺reaction < 0 and

it would be reversible if Δ𝐺reaction = 0. The equation Δ𝐺reaction = 0 can be used
to find conditions where a chemical reaction just begins to go forwards.

In electrochemistry, a cell involves a chemical reaction, plus charge flow

between the electrodes. The total Gibbs energy change of a cell therefore
includes some electrical work done on the load, Δ𝑊electrical
Δ𝐺total = Δ𝐺reaction + Δ𝑊electrical (2.23)

An electrochemical process will go forwards spontaneously if Δ𝐺total < 0, and

an electrochemical cell is reversible if Δ𝐺total = 0. A cell can be made to operate
reversibly by controlling the electromotive force that is drawn by the load.
In a reversible electrochemical cell
Δ𝐺total = Δ𝐺reaction + Δ𝑊electrical = 0 (2.24)

EJR 12
Electrical work can be written as the charge transferred across the cell
multiplied by the electromotive force experienced by the load. Assuming the cell
has negligible ion resistance and is reversible, the electrical work on the load is
Δ𝑊electrical = 𝑄𝐸reversible cell (2.25)
Where
 𝑄 = 𝑧𝐹 is the charge transferred through the cell
 𝑧 = 2 is the number of electrons per zinc atom in (Zn|Zn2+||H+|H2).
 𝐹 = 96485 Coulombs per mole of electrons.
 𝐸reversible cell is the reversible cell potential (Δ𝐸 volts) which is supplied to
the electrical load by the cell acting reversibly (i.e. when Δ𝐺total = 0).

Chemical thermodynamic data can therefore be converted to the voltage of a

reversible electrochemical cell by the important relation
Δ𝐺reaction = −𝑧𝐹𝐸reversible cell (2.26)

 A cell is reversible if all of the free energy change of the reaction is

converted to useful electromotive force.
 This value of 𝐸reversible cell can be measured by a voltmeter in a cell
operating reversibly.
 A high impedance voltmeter between two electrodes can find 𝐸reversible cell .
(This measurement may become difficult if a piece of metal actually hosts
multiple electrode reactions on its surface. Neglect this for now.)
Purely chemical reactions proceed spontaneously if Δ𝐺reaction < 0.
In electrochemistry, a corrosion reaction proceeds spontaneously if
𝐸reversible cell > 0 which is equivalent to Δ𝐺reaction < 0 and therefore Δ𝐺total < 0
in the spontaneous case when zero electrical work is involved.

2.6.1 Example: converting thermodynamic data from kJ/mol to volts

What is the electromotive force (potential difference) exerted by the following
zinc-acid cell when it is operated reversibly, at pH 0 with an effective dissolved
Zn2+ concentration of 1 M?
Zn(s) | Zn2+(aq) || H+(aq) | H2(g)
Data:
Δ𝐺reaction (Zn|Zn2+ | |H + |H2 ) = −147 kJ mol−1 under these conditions.

EJR 13
2.7 Standard electrode potential – a definition by example
The standard electrode potential of a half-reaction is the value of 𝐸reversible cell
in a special case. For zinc, it can be defined as follows.
The standard electrode potential of the zinc electrode, 𝐸°(Zn|Zn2+ ), is defined
as the value of 𝐸reversible cell for a cell with a zinc cathode (Zn|Zn2+ ), and a
hydrogen anode (H2|H+), with all reagents at unit activity and the cell operating
reversibly. It is usually defined at 25°C. The units are volts, and should be
written VSHE to make clear that the anode is a standard hydrogen electrode.

𝐸°(Zn|Zn2+ ) = 𝐸reversible cell (H2 |H + ||Zn2+ |Zn) [unit activities] (2.27)

= −Δ𝐺°reaction (H2 + Zn2+ → 2H + + Zn)/𝑧𝐹
= −(+147 kJ/mol)/(2 ∗ 96485 C /mol )
= −0.76 volts in the given conditions
= −0.76 VSHE
The reaction in this cell is zinc reduction by hydrogen.
… This reaction has negative E° (and positive Δ𝐺°reaction ). This is consistent
with zinc existing in oxidised forms in the natural world.

2.8 Using standard electrode potentials

The electrochemical cell
(Zn|Zn2+ | |Cu2+ |Cu) (2.28)
Involves materially the same change in state as
(H2 |H + | |Cu2+ | Cu) + (Zn|Zn2+ | | H + |H2 ) (2.29)
Which is also the same as
(H2 |H + | |Cu2+ | Cu) − (H2 |H+ | |Zn2+ | Zn) (2.30)
So the Gibbs energy changes for (2.28) and (2.30) are the same
Δr 𝐺(Zn|Zn2+ ||Cu2+ |Cu) = Δr 𝐺(H2 |H + | |Cu2+ | Cu) − Δr 𝐺(H2 |H + | |Zn2+ | Zn) (2.31)
Dividing (2.31) by -zF and using ° to indicate unit activities
𝐸°reversible cell (Zn|Zn2+ ||Cu2+ |Cu) = 𝐸°(Cu|Cu2+ ) − 𝐸°(Zn|Zn2+ ) (2.32)
More generally, this could be done for any electrodes
𝐸°reversible cell (anode||cathode) = 𝐸°cathode − 𝐸°anode (2.33)

This makes it simple to predict if a metal will corrode! If 𝐸°reversible cell > 0 for a
chosen anode and cathode, the anode will corrode (into unit activity solutions).
 At non-unit activities, the Gibbs energy changes Δr 𝐺 are different
from Δr 𝐺°, but the equivalent expression to equation (2.33) is
𝐸reversible cell = 𝐸0,cathode − 𝐸0,anode (2.34)

𝐸 ∘ values can be looked up. A correction can be added to find 𝐸0,anode for the
metal at non-unit activity. Crucially, the effective value of 𝐸°cathode affecting the
metal can be measured with a voltmeter, even if several reactions are involved.
𝐸unknown cathode = Ereversible cell (H2 |H + || unknown cathode) (2.35)
EJR 14
2.9 The IUPAC definition of standard electrode potential

The standard electrode potential, abbreviated E° or E (with a

superscript plimsoll character, pronounced "E standard" or "E nought"),
is the measure of the potential of a reversible electrode at standard state,
which is with solutes at an effective concentration of 1 mol dm −3, and
gases at a pressure of 1 bar.
[http://dx.doi.org/10.1351/goldbook.S05912 ]
The standard electrode potential is the value of the standard emf of a cell
in which molecular hydrogen under standard pressure is oxidized to
solvated protons at the left-hand electrode.

The standard electrode potential of a test electrode such as (M|Mz+) is found

when:
 Reagents are at unit activity, and
 The cell is reversible
 Standard electrode potentials are often tabulated at 25°C.
The standard electrode potential can be measured with a high-impedance
voltmeter between a test electrode and a standard hydrogen electrode.
 The standard hydrogen electrode is written (H2|H+) at pH 0 and 1 bar H2.
 Standard electrode potentials are therefore stated in volts (SHE) or VSHE.

Then the standard electrode potential is, with unit activities:

𝐸°(M|M 𝑧+ ) = 𝐸°reversible cell (H2 |H+ | | M 𝑧+ |M) (2.36)

2.9.1 Practical measurement using a silver chloride electrode

In practice, E° is often measured using silver chloride anodes, which are easy
to fabricate and use.
𝐸°(M|M 𝑧+ ) = 𝐸°reversible cell (Ag|AgCl|HCl(aq)| | M 𝑧+ |M) + 0.223 V (2.37)

Since
(H2 |H + | | M 𝑧+ |M) = (Ag|AgCl|H + , Cl− | | M 𝑧+ |M) + (H2 |H+ , Cl− |AgCl|Ag) (2.38)

And at unit activities at at 25°C

𝐸°(Ag|AgCl|HCl) = 0.223 VSHE (2.39)

Which produces equation (2.37) using the same reasoning as (2.28 – 2.33).

EJR 15
2.10 Non-standard concentrations
Most corrosion processes involve ion activities smaller than unity.
2.10.1 Example.
Seawater has a pH ≈8. Predict whether copper dissolves in this environment,
where there is 10−8 mol/dm3 of H+(aq). The copper metal is a solid phase, and
has unit activity, and the dissolved copper may be at a low concentration.
The standard electrode potential of copper is
𝐸°(Cu|Cu2+ ) = +0.337 VSHE (2.40)
So in a corrosion cell with copper as the anode
𝐸°reversible cell (Cu|Cu2+ | | H + |H2 ) = −0.337 V (2.41)
Therefore copper would not spontaneously dissolve by reaction with protons in
a solution which contained 1 mol dm-3 of dissolved copper ions at pH 0.
 But copper actually does dissolve into seawater, to a low concentration of
dissolved ions. This means, for example, that copper barrels containing
nuclear waste could not safely be disposed of by dropping them to the
bottom of the ocean!

2.10.2 The Nernst equation

Consider copper dissolution at 298 K in an aqueous solution of pH 0 and 1 pM
dissolved copper ion concentration.
Cu + 2H + → Cu2+ + H2 (2.42)
The Gibbs energy change, using equation (2.21), is
Δ𝐺reaction (Cu|Cu2+ | |H+ |H2 ) = Δ𝐺°reaction + 𝑅𝑇 ln(𝑄r ) (2.43)
= +65.03 kJ mol−1 + 𝑅𝑇 ln(𝑄r )
Where Qr is the reaction quotient
𝑎(Cu2+ ) 𝑎(H2 )
𝑄r = ( 2 ) (2.44)
𝑎(Cu) 𝑎(H+)

Hence, evaluating the Gibbs energy change

10−12 ×1
Δ𝐺reaction = +65.03 kJ mol−1 + 𝑅𝑇 ln (1×(10−0 )2 ) (2.45)
= −3.43 kJ mol−1
Or equivalently
Δ𝐺reaction 𝑅𝑇
𝐸reversible cell = − − 𝑧𝐹 ln(𝑄𝑟 ) (2.46)
𝑧𝐹

𝑅𝑇 𝑎(Cu2+ ) 𝑎(H2 )
𝐸reversible cell = 𝐸°reversible cell − 𝑧𝐹 ln ( 2 ) (2.47)
𝑎(Cu) 𝑎(H +)

𝐸reversible cell = +0.0177 V (2.48)

So copper would dissolve into a very dilute solution of Cu2+ ions at pH 0.

 This is part of the reason why copper dissolves in seawater.

EJR 16
Equation (2.46) is the Nernst equation for a complete cell reaction. If we know
the properties of both electrodes in a corrosion process, then we can evaluate
𝐸reversible cell . If 𝐸reversible cell > 0 then the metal anode will corrode
spontaneously.
However, the main reason for using electrochemical notation to study corrosion
is because we only know the exact properties of the metal (the possible anode).
Fortunately, the multi-reaction environment (the possible cathode) is something
we can measure.
 Equation (2.47) can be expanded by separating the logarithm term
2
𝑅𝑇 𝑎(Cu2+ ) 𝑅𝑇 𝑎(𝐻+)
𝐸reversible cell = 𝐸°reversible cell − 𝑧𝐹 ln ( ) + 𝑧𝐹 ln ( 𝑎 ) (2.49)
𝑎Cu (𝐻2 )

2
𝑅𝑇 𝑎(Cu2+ ) 𝑅𝑇 𝑎(𝐻+)
𝐸reversible cell = 𝐸°cathode − 𝐸°anode − 𝑧𝐹 ln ( ) + 𝑧𝐹 ln ( 𝑎 ) (2.50)
𝑎(Cu) (𝐻2 )

 We can spot that there are separate terms for the anode and cathode, and
the difference between them gives the reversible cell potential at non-
standard conditions.
2
𝑅𝑇 𝑎(𝐻 +)
𝐸0,cathode = 𝐸°cathode + 𝑧𝐹 ln ( 𝑎 ) (2.51)
(𝐻2 )

𝑅𝑇 𝑎(Cu2+ )
𝐸0,anode = 𝐸°anode + 𝑧𝐹 ln ( ) (2.52)
𝑎(Cu)

𝐸reversible cell = 𝐸0,cathode − 𝐸0,anode (2.53)

In equations (2.51-2.53), 𝐸0,cathode is the equilibrium potential of the cathode at

a non-standard concentration. 𝐸0,anode is the equilibrium potential of the anode.
These equilibrium potentials obey the Nernst equation for a single electrode
𝑅𝑇 [𝑜𝑥𝑖𝑑𝑖𝑠𝑒𝑑]
𝐸0,cathode = 𝐸°cathode + 𝑧𝐹 ln ( [𝑟𝑒𝑑𝑢𝑐𝑒𝑑] ) (2.54)
𝑅𝑇 [𝑜𝑥𝑖𝑑𝑖𝑠𝑒𝑑]
𝐸0,anode = 𝐸°anode + 𝑧𝐹 ln ( [𝑟𝑒𝑑𝑢𝑐𝑒𝑑] ) (2.55)

The notation [reduced] represents the product of the concentrations (or

pressures where gases are involved) of all of the species that appear on the
reduced side of the electrode reaction, raised to the power of their
stoichiometric coefficients. The notation [oxidised] represents the same for the
oxidised side of the electrode reaction.

The Nernst equation is sometimes written in log10. This makes it easier to deal
with concentration differences in terms of the number of powers of ten which
are involved.
𝑅𝑇 [𝑜𝑥𝑖𝑑𝑖𝑠𝑒𝑑]
𝐸0,electrode = 𝐸°electrode + 2.303 𝑧𝐹 log10 ( [𝑟𝑒𝑑𝑢𝑐𝑒𝑑] ) (2.56)

We can compare the equilibrium potential of the metal anode, 𝐸0,anode [which
we can calculate exactly using (2.56)] with the measured potential of the
environment, 𝐸0,cathode . Then if 𝐸reversible cell > 0, we predict the metal corrodes.

EJR 17
2.11 Concentration gradient cells
Example
An electrochemical cell consists of a silver anode dissolving into an aqueous
solution containing silver ions at 10 mM concentration, and a cathode
connected by a salt bridge in which silver ions at 100 mM concentration are
electroplated onto a silver block.
The temperature is 298 K.
Calculate the potential difference across the cell using the Nernst Equation, if
the cell is operated reversibly.

Solution
Cell notation: Ag(s) | Ag+(aq, 0.01 M) || Ag+(aq, 0.1 M) | Ag(s)

𝐸reversible cell = 𝐸0,cathode − 𝐸0,anode

𝑅𝑇 [𝑜𝑥𝑖𝑑𝑖𝑠𝑒𝑑] 𝑅𝑇 0.1
𝐸0,cathode = 𝐸°cathode + 𝑧𝐹 ln ( [𝑟𝑒𝑑𝑢𝑐𝑒𝑑] ) = 𝐸°(Ag|Ag + ) + 𝑧𝐹 ln ( 1 )
𝑅𝑇 [𝑜𝑥𝑖𝑑𝑖𝑠𝑒𝑑] 𝑅𝑇 0.01
𝐸0,anode = 𝐸°anode + 𝑧𝐹 ln ( [𝑟𝑒𝑑𝑢𝑐𝑒𝑑] ) = 𝐸°(Ag|Ag + ) + 𝑧𝐹 ln ( )
1

𝑅𝑇
𝐸reversible cell = 𝐸0,cathode − 𝐸0,anode = (ln(0.1) − ln(0.01)) = +59 mV
𝑧𝐹

… Reality check: positive Ecell means this process is expected to go forwards

spontaneously, which (since the process is the flow of silver ions from high to
low concentration) is correct.
... For a reaction with 1 electron per atom, the equilibrium potential of an
electrode at 25°C shifts by 59 mV per decade change in concentration of the
atom (or other species). This often means 59 mV per unit of pH.

2.11.1 Analogy between temperature and oxidation potential

EJR 18
2.12 Predicting corrosion from thermodynamic data
A metal is expected to corrode if it is the anode in a cell where:
𝐸reversible cell = (𝐸0,cathode − 𝐸0,anode ) > 0 (2.57)
Where the equilibrium potentials usually need to be calculated using the Nernst
equation (2.56), because usually the ion activities are not unity.
A relevant value for 𝐸°anode can be looked up for a metallic element in a table
of standard electrode potentials. This can be used to find 𝐸0,anode using the
Nernst equation (2.32) for the given concentration of dissolved ions.
A relevant value for the cathode usually consists of contributions from several
reactions due to chemicals in the environment reacting on the metal surface.
This can be measured. Tables listing measured values in common situations
are called practical galvanic series.

Table 2.1. Some standard electrode potentials in aqueous solution (for unit activities)
Electrode Standard equilibrium
potential / VSHE
(Au|Au3+) +1.40
(H2O|O2, H+) at pH 0 +1.22
(Pt|Pt2+) +1.19
(Ag|Ag+) +0.80
(H2O|O2, H+) at pH 14 +0.40
(Cu|Cu2+) +0.34
(H2|H+) at pH 0 0.00
(Pb|Pb2+) -0.13
(Fe|Fe2+) –0.44
(Zn|Zn2+) –0.76
(H2|H+) at pH 14 –0.83
(Al|Al3+) –1.66
(Mg|Mg2+) –2.36

Table 2.2. Practical galvanic series for metals in neutral soils or water [Roberge p.109]
Metal (and condition) Potential on surface / VSHE
Platinum +0.3
Mild steel in concrete +0.1
Copper, brass, bronze +0.1
Mild steel (rusted) –0.2 to +0.1
Lead –0.2
Mild steel (shiny) –0.5 to –0.2
Pure aluminium –0.5
Aluminium-zinc alloy (5% Zn) –0.75
Zinc –0.8
Magnesium –1.4

EJR 19
2.13 Practical Galvanic Series
The electrochemical series lists thermodynamic data describing individual
electrode equilibrium potentials in volts (SHE).
A galvanic series is a table of the electrode potentials measured on the surface
of real metals in a particular environment.
 A practical galvanic series is measured for particular conditions (e.g. “metal
in neutral soils”).
o It tabulates the potential of a metal surface which may host several
reactions, involving environmental chemicals as well as the metal.
o Temperature, pH, and salinity may affect the potential on the metal
surface, so there are different galvanic series for different environments.
 When there are many half-cell reactions on a metal surface, the galvanic
potential measured on its surface is called a mixed potential:

The galvanic series tabulates the potential observed when all these
electrochemical reactions occur at once. It is a weighted average of the various
equilibrium potentials.
 The galvanic series is used for:
o Determining if the mixed electrode potential measured on a metal
surface is sufficiently oxidising to cause it to corrode.

o Determining which of two metals may suffer increased corrosion

if they are bolted together.

Galvanic potential on a metal can be very different from its equilibrium potential.
 Some metal electrodes are kinetically slow. The rate of reaction of
(Pb|Pb2+) is slower than reactions like (H2|H+), and the galvanic potential will
be close to the equilibrium potential of the faster reaction.
 Some materials are strong electrocatalysts. Rusty iron is a stronger
electrocatalyst than pure iron for (H2|H+) and (H2O|O2).
o This is why rust is sometimes called ‘autocatalytic’. The galvanic potential
of iron becomes closer to the oxidising potentials of (H2|H+) and (H2O|O2)
due to rust formation, and this produces a stronger cathode for corrosion.
 Some metals have passive oxide surfaces. Aluminium’s surface chemistry
is mainly affected by various reactions taking place on its Al2O3 surface,
and not by the reaction (Al|Al3+).

EJR 20
2.14 Corrosion and pH
pH affect aqueous corrosion for several reasons.
 pH affects the ion conductivity of solutions. Ion mobility is needed for
electrochemical corrosion, and can be rate-limiting.
 pH determines the oxidising potential exerted by protons dissolved in
water. From equations 2.3 and 2.10, at 298 K.
2
𝑅𝑇 𝑎(𝐻 +)
𝐸0,cathode = 𝐸°cathode + 𝑧𝐹 ln ( 𝑎 ) (2.58)
(𝐻2 )

Example
Effect of pH on reactions in copper corrosion at 298 K, and unit activity of
dissolved copper.
Corrosion reaction Notation Equilibrium potential / VSHE

(A) Hydrogen (H2|H+) 𝐸0 = 0.00 − 0.059 pH

evolution

(B) Oxygen (H2O|O2, H+) 𝐸0 = 1.228 − 0.059 pH

reduction

(C) Copper (Cu|Cu2+) 𝐸0 = +0.337

dissolution

(D) Copper oxide Cu + H2O = CuO + 2H+ + 2 e- 𝐸0 = 0.570 − 0.059 pH

formation

(E) Copper oxide Cu2+ + H2O = CuO + 2H+ 𝐸0 is not determined for a
dissolution pure acid-base reaction

log10 (𝑎(Cu2+ )) = 7.9 − 2 pH

EJR 21
2.15 Mapping out corrosion products on pH-potential diagrams
The pH-potential diagram, also called a Pourbaix diagram, maps out the
thermodynamically stable phase of a metal in aqueous solution as a function of
the oxidation potential and pH of the environment.

Example: Construct the pH-potential diagram for copper in aqueous solution,

from pH 0 to 10 and oxidation potential -1.5 VSHE to + 1.5 VSHE.

1.5

1
E, VSHE

0.5

-0.5
0 2 4 6 8 10
pH

Instructions:
 pH on x-axis, oxidation potential on y-axis.
 Dashed lines showing oxidising agents (half-reactions A, B, above)
 Label common environments (e.g. de-aerated neutral water).
 Draw the horizontal line of the Cu | Cu2+ equilibrium.
 Draw the vertical line showing the Cu2+ | CuO equilibrium.
 Draw the diagonal line showing Cu | CuO equilibrium.
 Label the stable phases.

 Note this diagram was simplified by neglecting:

o CuO22- which is stable at pH > 12.
o A small region of Cu2O stability between the hydroxide and metal.
o All of the chloride complexes which can appear in salt water.
EJR 22
2.16 Simplified Pourbaix (pH-potential) diagrams for zinc and iron
pH-potential diagram for 1  M Zn ions in aqueous solution
1.5

0.5
Zn2+ Zn(OH)2
E, VSHE

-0.5

-1 ZnO2-
2
Zn

-1.5
0 2 4 6 8 10 12 14
pH

pH-potential diagram for 1  M Fe ions in aqueous solution

1.5

3+
1 Fe

Fe2O3 passivity
0.5
Fe3O4
E, VSHE

Fe2+ corrosion
0
Fe(OH)2

-0.5
HFeO2-

-1 Fe immunity

-1.5
0 2 4 6 8 10 12 14
pH

Reference: Marcel Pourbaix, “Atlas of Electrochemical Equilibria in Aqueous

Solution.” Pergamon (1974).

EJR 23
2.17 The purpose and limitations of pH-potential diagrams
Purpose
 Data on a small number of equilibria can predict stable phases at any
possible value of { pH , E }.
 Can measure pH and oxidation potential, then refer to Pourbaix diagrams
to find a suitable, passive metal.
 Represents redox and acid-base interactions visually
o Redox: electron transfer, y-axis, so pure redox equilibria are
separated by horizontal lines of constant oxidation potential
o Acid-base: proton transfer is only affected by pH, so pure acid-
base reactions (between ions and oxides at the same metal
oxidation state) are separated by vertical lines of constant pH
o Mixed ( metal | metal oxide ) reactions combine redox and acid
base, and are separated by diagonal lines.

Limitations
 Does not predict kinetics. Driving force in volts can be measured on the
diagrams, but this is just one input to kinetic models
 Simple versions neglect chloride and other complexing ions – would have
to add a third axis, or multiple diagrams, to show the effect of different
chloride concentration.
 Lines refer to equilibria at particular solution concentrations. Need to adjust
line position for different concentrations. Lines are often drawn for
equilibrium with 1 M or 1 micromolar activities, with the latter used to
determine stability against corrosion in most environments.
 Identifies when solid oxidation products are stable, but does not give
information on whether they are impermeable (protective) and thus lead to
passivity.

EJR 24
 3. Kinetics of aqueous corrosion

Thermodynamics can determine whether aqueous corrosion will occur, and its
driving force. To predict the rate, a kinetic model is needed.
The rate of an electrochemical reaction is expressed as an electronic current
density which is related to the molar rate by Faraday’s law.
In models of electrochemical kinetics, a zero rate of corrosion for an electrode
such as (Cu|Cu2+) means there is dynamic equilibrium with an equal rate of
oxidation and reduction. When the environment of a metal exerts an oxidising
action, the electrode potential is polarised away from the equilibrium potential,
and the oxidation of the metal proceeds faster than reduction. This leads to a
net positive rate of corrosion. The rate of corrosion is predicted by two factors:
the baseline rate of the dynamic equilibrium reactions, called the exchange
current density, and by the way this baseline rate is increased by polarisation.

3.1 Measuring rates of corrosion

Methods of measuring corrosion depend on its rate and mechanism.

Measurement Units Applications

Weight change As mm/year loss of Most metals

thickness (speed)

% weight change Rare. Need standard test objects.

Permeation Volume flux per Special gases (hydrogen blistering).

unit pressure

Penetration mm Depth of corrosion pits

mm Depletion of alloy component

Tensile properties Pa m0.5 Change in fracture toughness due to

(sometimes) embrittlement

Appearance ‘Surface texture’ Counting number of pits

‘Colour’ E.g. copper alloys

Effect on Toxicity / contamination

environment

Current density A/m2 Electrochemical corrosion

EJR 25
3.1.1 Current density
Electronic current density, i, (measured in mA/cm2 or A/m2), is used to quantify
reaction rate in aqueous corrosion.
 Current is easy to measure.
 Current density is related to molar corrosion rate by Faraday’s law.
 It is easily converted to mm/year.

3.2 Relation of current density to speed of corrosion

In a cell, the corrosion rate at the anode (M → M 𝑧+ + 𝑧e− ) is proportional to the
electronic current, I.
 One mole of metal M reacting gives out z moles of electrons.
 The total charge that passes through the cell when N moles of metal M
react is Q.
𝑄 = 𝑧𝐹𝑁 (3.1)
 The current flowing through an electrochemical cell in which the metal M is
oxidised at a rate of dN/dt moles per unit time is given by Faraday’s law:
𝐼 = 𝑧𝐹(d𝑁/d𝑡) (3.2)
𝑧𝐹
or 𝑖 = ( 𝐴 ) (d𝑁/d𝑡) (3.3)
 Where the current density, i, given the electrode area A, is
𝑖 = 𝐼/𝐴 (3.4)

An anodic current density is positive and can be written 𝑖a , and a cathodic

current density is negative, and can be written as 𝑖c . Sometimes, the absolute
values |𝑖a | and |𝑖c | may be written for clarity. The following equation uses |𝑖a |
and |𝑖c | to describe copper behaviour near a dynamic equilibrium:
𝑖net (Cu|Cu2+ ) = 𝑖a + 𝑖c = |𝑖a | − |𝑖c | (3.5)
= |𝑖(Cu → Cu2+ + 2e− )| − |𝑖(Cu2+ + 2e− → Cu)|

3.2.1 Example
Calculate the net current density equivalent to the steady-rate dissolution of
1 mm/year of steel. Assume the reaction is Fe = Fe2+ + 2e-.
Data:
Density of iron = 7874 kg/m3
Atomic mass of iron = 55.85

EJR 26
3.3 Polarisation (sometimes called ‘overpotential’)
In a real environment, there are many electrochemical reactions taking place
on the surface of a metal object.
 The metal surface has an electrochemical potential, E.
 E may be called Emixed if the potential is caused by a mixture of chemical
species, or Eapplied if the potential is applied by external electronics.
 𝐸 may be different from the equilibrium potential, E0, of an individual
electrode reaction. Suppose a piece of copper in water has an electrode
potential 𝐸 ≠ 𝐸0 (Cu|Cu2+ ). Then the electrode is said to be polarised with
respect to (Cu|Cu2+ ). The polarisation, η, is measured in volts.
𝜂 = 𝐸 − 𝐸0 (3.6)
Or 𝜂(Cu|Cu2+ ) = 𝐸 − 𝐸0 (Cu|Cu2+ ) which specifies the reaction

The polarisation of an electrode is the driving force for corrosion. Compare

equation (3.6) with thermodynamics (sections 2.2 – 2.12).
Recall that a two-electrode process proceeds if
𝐸reversible cell = (𝐸0,cathode − 𝐸0,anode ) > 0 (2.57)
Corrosion of copper, (Cu|Cu2+ ), in a complex environment proceeds if
𝐸reversible cell (Cu|Cu2+ || ′environment′) > 0 (3.7a)
Or 𝐸 − 𝐸0 (Cu|Cu2+ ) > 0 (3.7b)
Or just 𝜂(Cu|Cu2+ ) > 0 (3.7c)

 𝜂(M|M + ) > 0 means the local environment will make (M|M+) an anode and
so the metal will be oxidised.
 𝜂(M|M + ) < 0 means the local environment will not corrode (M|M+). In this
case:
𝐸reversible cell (Cu|Cu2+ || ′environment′) = 𝜂(Cu|Cu2+ ) < 0
There is a driving force for the reverse process in the cell – electroplating.
 η = 0 indicates there is zero driving force for corrosion or electroplating.

Electrochemical kinetics assumes that a zero net rate of corrosion, when η =

0, involves a dynamic equilibrium, with equal rates of dissolution and plating.

Oxidised species + 𝑧e− ↔ Reduced species

For example, a (Cu|Cu 2+) electrode with 𝜂(Cu|Cu2+ ) = 0 V might experience:

𝑖net (Cu|Cu2+ ) = |𝑖a (Cu|Cu2+ )| − |𝑖c (Cu|Cu2+ )| = 𝑖0 − 𝑖0 = 0 (3.8)

Where i0 is property called the exchange current density of the electrode.

EJR 27
3.4 Exchange current density
Exchange current densities, 𝑖0 , can be found from experimental data.
Exchange current densities depend on the surface chemistry.
 E.g. in electrolysis of water, the hydrogen electrode (H2|H+) involves several
steps: (1) electron transfer, (2) H atom diffusion, and (3) desorption of H2
molecules. The rates of these steps depend on electrocatalysis, so different
values of 𝑖0 exist for the same reaction on different metals.

Table 3.2. Exchange current densities for 2H+ + 2e- = H2 [Bockris, Textbook of
electrochemistry, 1951].
Metal 𝑖0 (H2 |H+ )
Platinum 1 x 10-3 A/cm2
Nickel 4 x 10-6 A/cm2
Iron 2 x 10-6 A/cm2
Zinc 1 x 10-8 A/cm2
Lead 2 x 10-13 A/cm2

Polarisation and exchange current density are the two key concepts needed to
relate corrosion rate to thermodynamic driving force.

3.4.1 Diagram of factors which may limit electrochemical reaction rate

e-
V
Anode Cathode

In this model, consider:

−Δr 𝐺
( 𝑧𝐹
) = 𝐸cell + |𝜂anode | + |𝜂cathode | + |𝜂ion resistance | + |𝜂diffusion | (3.9)

EJR 28
3.5 The Tafel equation: a model for the current density of an
electrode when the only significant limiting factor is activation
The Tafel equation is used to estimate the current density of an electrode
reaction such as (Cu|Cu2+), assuming:
 The reaction is slow, so there are negligible energy losses due to mass
transport, which means 𝜂diffusion = 𝜂ion resistance = 0
 The surface of the metal is at a potential, E
𝜂anode (Cu|Cu2+ ) = 𝐸 − 𝐸0 (Cu|Cu2+ ) (3.10)

The following ‘distance model’ is used in electrochemistry. The Gibbs energy of

an electrode is plotted as a metal atom is moved a distance Δx into the solution,
where the atom eventually becomes a fully dissolved ion.

The net rate of the (Cu|Cu2+) reaction is obtained by adding up simultaneous

forward and reverse reactions. The forward and reverse reactions are both
described as Arrhenius-limited with attempt rates A and activation energies Δ𝐺 ∗ .
Δ𝐺∗for
−( )
𝑖for = 𝑖(Cu → Cu2+ + 2e− ) = 𝐴for e 𝑅𝑇
(3.11)
Δ𝐺∗
−( rev )
𝑖rev = 𝑖(Cu2+ + 2e− → Cu) = 𝐴rev e 𝑅𝑇 (3.12)

(A) When the polarisation is zero, the forward and reverse activation energies
are the same (Δ𝐺 † ), and the net reaction rate is zero (from equation 3.7).
Therefore we can identify:
Δ𝐺† Δ𝐺†
−( ) −( )
𝑅𝑇 𝑅𝑇
𝑖for = 𝑖rev = 𝐴for e = 𝐴rev e = 𝑖0 (3.13)

(B) When there is some polarisation, 𝜂 > 0, the forward activation energy is
decreased, and the reverse activation energy is increased, to the values given
in the figure. (Often, α=0.5.) The reaction rates becomes:
Δ𝐺† −𝛼𝑧𝐹𝜂 Δ𝐺† +(1−𝛼)𝑧𝐹𝜂
−( ) −( )
𝑅𝑇 𝑅𝑇
𝑖net = 𝑖for − 𝑖rev = 𝐴for e − 𝐴rev 𝑒 (3.14)
Which simplifies to the Butler-Volmer equation (Databook p.33):
𝛼𝑧𝐹𝜂 (1−𝛼)𝑧𝐹𝜂
𝑖net (Cu|Cu2+ ) = 𝑖0 [ exp (+ ) − exp (− )] (3.15)
𝑅𝑇 𝑅𝑇

EJR 29
3.5.1 Figure showing the exact kinetic behaviour of an electrode, according to
the Butler-Volmer model (with α=0.5, i0=1 µA/cm2, 298 K, z=1).

3.5.2 The Tafel equation (Databook p.33)

The Tafel equation is a simplified case of (3.15).
The Tafel equation is an approximation which considers only the forward or
reverse reaction. From Figure (3.5.1), it is a good approximation if the
polarisation is at least 50 mV, which is often a valid approximation in corrosion.

Corrosion (anodic polarisation)

Only the first term is significant if η > +50 mV, since |𝑖a | ≫ |𝑖c |

𝛼𝑧𝐹𝜂
𝑖 ≈ 𝑖for = 𝑖0 exp ( ) = 𝑖0 × 10𝜂/𝛽a (3.16)
𝑅𝑇

So 𝑖 = 𝑖0 × 10𝜂/𝛽a (3.17a)
Or 𝜂 = 𝛽a log10 (𝑖/𝑖0 ) (3.17b)
𝑅𝑇
Where 𝛽a = 2.303 𝛼𝑧𝐹 (3.18)
Equation (3.17) is the Tafel equation for an anode, and 𝛽a is the Tafel slope.

Reduction reaction (cathodic polarisation)

Only the second term is significant if η < –50 mV, since |𝑖a | ≪ |𝑖c |

𝑖 ≈ −|𝑖𝑐 | = −𝑖0 × 10𝜂/𝛽c with η and 𝛽c negative (3.19)

So 𝑖 = −𝑖0 × 10𝜂/𝛽c (3.20a)
𝜂 = 𝛽c log10 ( | 𝑖 | / 𝑖0 ) (3.20b)
𝑅𝑇
Where 𝛽c = −2.303 (1−𝛼)𝑧𝐹 (3.21)

Equation (3.20) is the Tafel equation for a cathode.

EJR 30
3.5.3 Notes on using the Tafel equation
Sometimes, the cathodic Tafel slope βc and the cathodic polarisation ηc
might both be treated as absolute values instead of being considered
negative.

Near equilibrium
For small polarisations, |η| < 50 mV, both terms in the Butler-Volmer
equation are significant. In simple reactions, 𝛼 = 0.5, in which case
𝑧𝐹𝜂
𝑖 = 2 𝑖0 sinh (2𝑅𝑇) (3.22)
𝑖0 𝑧𝐹𝜂
or 𝑖≈ for small values of η. (3.23)
𝑅𝑇

Hence linear behaviour is expected for low polarisations, meaning the Tafel
equation is not obeyed at low polarisations.
 We are normally interested only in the behaviour when the polarisation is
moderately large (i.e. the Tafel equation region).

EJR 31
3.6 Corrosion of a two-electrode system with no mass transport
limitation (and unit activities).
Example problem
Determine the corrosion current density of an iron object dissolving in
deaerated acid at pH 0 containing 1 M Fe2+. Assume the anode reaction is
Fe→Fe2++2e-, and the only cathode reaction is 2H++2e-→H2, and that there is no
mass transport limitation.
Note there is just one metal surface, so |𝑖a | = |𝑖c | = 𝑖corr at the solution
Data:
E0(H2|H+) = 0.00 VSHE (same as E° since activity = 1)
i0(H2|H+) = 10-3 mA/cm2 on iron
βc(H2|H+) = –0.118 V/decade
E0(Fe|Fe2+) = –0.44 VSHE (same as E° since activity = 1)
i0(Fe|Fe2+) = 10-5 mA/cm2
βa(Fe|Fe2+) = +0.059 V/decade

0.1

0
E / Volts SHE

-0.1

-0.2
ia(Fe|Fe2+) Tafel
-0.3
ic(H+|H2) Tafel

-0.4 ia(Fe|Fe2+) exact

ic(H+|H2) exact
-0.5
-6 -5 -4 -3 -2 -1 0 1 2 3 4
log10(i / (mA/cm2))

Figure. An Evans diagram is a plot of potential against log of current density. Here, the
corrosion rate of a two-electrode system is found by the Tafel equation and the Butler-
Volmer model, neglecting mass transport limitation.

EJR 32
3.7 Diffusion-limited corrosion rate
In the iron-acid example above, the chemical reaction was the rate-determining
step. However, diffusion of species to or from the electrode surface may be
rate-limiting.
 This is particularly important for corrosion caused by a dilute oxidising
agent, such as dissolved oxygen.
O2 + 4H + + 4e− → 2H2 O (H2O|O2, H+) cathode (3.24)
mA
𝐸0 = +1.22 VSHE , 𝑖0 = 10−5 cm2 , 𝛽c = −0.1 V/decade (3.25)
To model diffusion-limited corrosion, a linear diffusion boundary layer is
assumed to exist.

1 d𝑁
 The molar flux of oxygen is ( d𝑡 ) = 𝑘g (𝑐𝑏 − 𝑐𝑖 ), where kg is a mass transfer
𝐴
coefficient.
1 d𝑁 d𝐶
 Alternatively, by Fick’s law, ( ) = −𝐷 d𝑥 (3.26)
𝐴 d𝑡

 The current density is related to this rate of diffusion by Faraday’s law

|𝑖| = 𝑧𝐹𝑘g (𝑐b − 𝑐𝑖 ) or |𝑖| = 𝑧𝐹𝐷 (𝑐b − 𝑐𝑖 )/𝑥 (3.27)
 So the maximum current density of corrosion on this surface is
𝑖L = 𝑧𝐹𝑘g 𝑐b or 𝑖L = 𝑧𝐹𝐷 (𝑐b )/𝑥 (3.28)

 Crucially, the surface concentration of oxygen can be related to current

density. Using the versions of (3.27) and (3.28) involving 𝑘g
|𝑖| |𝑖|𝑐 |𝑖|
𝑐𝑖 = 𝑐b − 𝑧𝐹𝑘 = 𝑐b − 𝑧𝐹𝑘 b𝑐 = 𝑐b (1 − 𝑖 ) (3.29)
g g b L

 The depleted surface concentration of oxygen affects the equilibrium

potential of the (H2O|O2, H+) cathode according to the Nernst equation.
𝑅𝑇 𝑎
𝐸0 (H2 O|O2 , H + ) = 𝐸 0 + 2.3 𝑧𝐹 log10 (𝑎Oxidised ) (3.30)
Reduced
𝑅𝑇
= 𝐸 0 + 2.3 𝑧𝐹 log10 (𝑐𝑖 )
𝑅𝑇 𝑅𝑇 |𝑖|
= 𝐸 0 + 2.3 𝑧𝐹 log10 (𝑐b ) + 2.3 𝑧𝐹 log10 (1 − 𝑖 )
L

𝑅𝑇 |𝑖|
= lim( 𝐸0 ) + 2.3 𝑧𝐹 log10 (1 − 𝑖 )
𝑖→0 L

 So diffusion limitation can be modelled by a voltage loss

𝑅𝑇 |𝑖|
|𝜂diffusion | = 2.3 𝑧𝐹 log10 (1 − 𝑖 ) (3.31)
L

EJR 33
Diffusion-limitation can be added to the existing corrosion model.
Example problem
Determine the corrosion current density of a copper object dissolving in aerated
acid at pH 0 containing dissolved Cu2+ and O2 both at unit activity (this is
unrealistically high for oxygen). Assume the anode reaction is Cu→Cu2++2e-,
and the only cathode reaction is O2+4H++4e–→2H2O, and that there is diffusion
limitation of oxygen.
Note there is just one metal surface, so |𝑖a | = |𝑖c | = 𝑖corr at steady state
Data:
E0(Cu|Cu2+) = +0.34 VSHE E0(H2O|O2, H+) = +1.22 VSHE
i0(Cu|Cu2+) = 10-5 mA/cm2 i0(H2O|O2, H+) = 10-5 mA/cm2
βa(Cu|Cu2+) = +0.059 V/decade βc(H2O|O2, H+) = –0.059 V/decade
𝑖L = 2 mA/cm2 for oxygen

The solution to this problem is the intersection of the current-potential curves of

the anode and cathode
𝜂anode = 𝐸 − 𝐸0 (Cu|Cu2+ ) (3.32)
𝑖
𝛽a log10 (𝑖a ) = 𝐸 − 𝐸0 (Cu|Cu2+ ) (3.32a)
0

𝜂cathode + 𝜂diffusion = 𝐸 − 𝐸0 (H2 O|O2 , H + ) (3.33)

|𝑖 | 𝑅𝑇 |𝑖c |
𝛽c log10 ( 𝑖c ) + 2.3 𝑧𝐹 log10 (1 − ) = 𝐸 − 𝐸0 (H2 O|O2 , H + ) (3.33a)
0 𝑖L

This can be plotted

1.4
ia(Cu|Cu2+)

ic(O2|H2O), iL=2mA/cm2
1.2

1
E / Volts SHE

0.8

0.6

0.4

0.2
-6 -5 -4 -3 -2 -1 0 1 2 3
2
log10(i / (mA/cm ))

 Concentration polarisation at the anode (caused by slow desorption of

metal ions), is rarely important.
EJR 34
 The diffusion-limited current density 𝑖L , depends on several factors.
o Diffusivity of oxygen in water (affected by temperature)
o Dissolved oxygen concentration in the bulk
o Reynolds number (affects boundary layer thickness)

1.4
ia(Cu|Cu2+)
ic(O2|H2O), iL=2mA/cm2
1.2
ic(O2|H2O), iL=20mA/cm2
ic(O2|H2O), iL=500mA/cm2
1
E / Volts SHE

0.8

0.6

0.4

0.2
-6 -5 -4 -3 -2 -1 0 1 2 3
2
log10(i / (mA/cm ))

Figure 3.6. Effect of increasing iL on the diffusion-limited corrosion rate

EJR 35
3.8 Ion conductivity limited corrosion rate (non examinable)
The final limiting factor in aqueous corrosion is resistance to ion flow.
Consider a model of spontaneous corrosion of copper (Cu|Cu2+||O2,H+|H2O)
where
 The anode and cathode are activation limited
 Diffusion limitation is not significant (𝑖L = 500 mA/cm2 )
 There is significant resistance to ion flow between the electrodes

−Δr 𝐺
− 𝐸cell = |𝜂anode | + |𝜂cathode | + |𝜂oxygen diffusion | + |𝜂ion resistance |
𝑧𝐹

(3.34)

For illustrative purposes, it is possible to write, for an electrolyte of resistivity ρ

and a characteristic length scale l.
𝜂ion resistance = 𝑖𝜌𝑙 (3.35)
It is also possible to construct a model so that
𝜂anode + 0.5 𝜂ion resistance = 𝐸 − 𝐸0 (Cu|Cu2+ ) (3.36)
𝜂cathode + 𝜂diffusion − 0.5 𝜂ion resistance = 𝐸 − 𝐸0 (H2 O|O2 , H + ) (3.37)

Given some estimates of ρ (0.3 Ω m for seawater, 50 Ω m for tap water, and
5000 Ω m for pure water) and l (1 cm).

1.2
ia(Cu|Cu2+), sea

1.1 ia(Cu|Cu2+), tap

ia(Cu|Cu2+), pure
1 ic(H2O|O2), sea
ic(H2O|O2), tap
E / Volts SHE

0.9 ic(H2O|O2), pure

0.8

0.7

0.6

0.5
-1 -0.5 0 0.5 1 1.5 2 2.5 3
2
log10(i / (mA/cm ))

This model shows

 Seawater (~550 mM NaCl) can support higher corrosion rates than fresh
water due to higher ion conductivity. Temperature is also important.
EJR 36
3.9 The Area Effect in corrosion models
When two different electrodes are connected, one will be cathodic while the
other will be anodic.
 This situation can occur on a single metal surface. In the iron-acid and
copper-oxygen examples above, the anode and cathode were treated
as having the same area, since they took place on the same metal
object. In reality, the surface would have hosted a mixture of ‘anodic
sites’ and ‘cathodic sites’ due to concentration gradients.
 Often, corrosion takes place where two or more metals are joined
together. In this situation, there is usually one dominant anode reaction,
and one dominant cathode reaction, but they take place over different
total areas.

Example
A zinc anode (Zn|Zn2+) of area 1 m2 is attached to a sheet of iron of area
10 m2. This metal is entirely submerged in acid at pH 0, containing
dissolved Zn2+ at unit activity. Given that the only significant reactions are
a zinc anode (Zn|Zn2+) and hydrogen evolution on the iron surface (H+|H2),
calculate the anode current density. Assume there is no mass transport
limitation.
Data:
E0(Zn|Zn2+) = –0.76 VSHE E0(H2|H+) = +0.00 VSHE
i0(Zn|Zn2+) = 10-8 mA/cm2 i0(H2|H+) = 10-3 mA/cm2 on iron
βa(Zn|Zn2+) = +0.059 V/decade βc(H2|H+) = –0.059 V/decade

Solution
 The entire surface is at a potential E
 No charge accumulation, and Tafel-limited kinetics on both electrodes
𝐴a |𝑖a | = 𝐼cell = 𝐴c |𝑖c |

(Zn|Zn2+ ))/𝛽a + |H
𝐴a 𝑖0 (Zn|Zn2+ ). 10(𝐸−𝐸0 = 𝐴c 𝑖0 (H + |H2 ). 10(𝐸−𝐸0 (H 2 ))/𝛽c

2+ + +
𝐸−𝐸0 (Zn|Zn ) 𝐸−𝐸0 (H |H2 ) 𝐴c 𝑖0 (H |H2 )
− = log10 ( 2+ )
𝛽a 𝛽c 𝐴a 𝑖0 (Zn|Zn )

Since βc = – βa this simplifies to

+
𝐴c 𝑖0 (H |H2 )
2𝐸 = 𝛽a log10 ( 2+ ) + 𝐸0 (Zn|Zn2+ ) + 𝐸0 (H + |H2 )
𝐴a 𝑖0 (Zn|Zn )

𝐸 = −0.203 VSHE
(Zn|Zn2+ ))/𝛽a
𝑖a (Zn|Zn2+ ) = 𝑖0 (Zn|Zn2+ ). 10(𝐸−𝐸0 = 27.5 mA cm−2
… check that |ic| is 10% of this…

EJR 37
3.10 Summary of kinetic models
 We can predict corrosion rates for a single pair of corrosion half-reactions,
using the Tafel approximation and either (a) algebra, or (b) plotting a
suitable graph, called an Evans diagram, showing log of current density
and potential.
 The rate of corrosion depends on:
o The electrode equilibrium potentials of the reactions involved
o The exchange current densities (which depend on surface chemistry)
o The Tafel parameters
o Any externally applied electronic currents
 The charge conservation condition, 𝐴a |𝑖a | = 𝐴c |𝑖c |, implies corrosion will
be particularly rapid if 𝐴c ≫ 𝐴a . A general principle in corrosion prevention
is to avoid the situation of a large cathode and small anode area.
 More sophisticated corrosion models can explain: (a) mass transport or
diffusional limitations and (b) electrolyte ion resistance limitations.

EJR 38
3.11 Kinetics of Passivation
Passivation is the formation of a protective, solid oxidised layer.
pH-potential diagram for 1  M Fe ions in aqueous solution
1.5

3+
1 Fe

Fe2O3 passivity
0.5
Fe3O4
E, VSHE

Fe2+ corrosion
0
Fe(OH)2

-0.5
HFeO2-

-1 Fe immunity

-1.5
0 2 4 6 8 10 12 14
pH

Figure 3.8. A Pourbaix (pH-potential) diagram shows thermodynamic passivation.

1.5 ia(Fe|Fe2+)
ic(weak oxidation)
ic(strong oxidation)
1
E / V SHE

0.5

-0.5
-8 -6 -4 -2 0 2 4
2
log10(i / (mA/cm ))

Figure 3.9. An Evans diagram shows the kinetics of passivation.

 At low potentials “normal” corrosion occurs: active dissolution region
 At higher potentials, the corrosion products cause passivation, and so the
rate of corrosion is reduced (or becomes negligible).
 At very high potentials, the corrosion products themselves undergo
chemical reaction (the “transpassive” region).

EJR 39
3.11.1 Factors affecting passivation
 Oxidation potential and pH of environment.
 Composition. Transition metals such as iron tend to have solid higher-
oxides and so they can show active-passive behaviour. (I.e. they go from
immune active  passive  transpassive with increasing oxidation
potential and so can be ‘anodically protected’). Other metals may just have
immune active behaviour.
 Alloying. In passivated stainless steel, the surface is effectively chromium
oxide. A (very simplified) description of chromium’s effect is that it
significantly expands the regions of passivity that are already present on
the pH-potential diagram of iron, because the alloy will be passive if either
the iron or the chromium are passive. In order for stainless steel to benefit
from this property, the chromium content needs to be at least 11%, and is
typically 18% to allow for local depletion.
 Aggressive chemicals. Chloride ions can cause breakdown of many
passive oxides, shrinking the passive region on the pH-potential diagram,
and increasing the passive current density.
 Abrasion and cracking. These processes may be the rate-limiting step for
determining the passive current density.

Glossary of symbols for electrochemistry

Symbol Description Units Section
i Current density A m-2 3.11
i0 Exchange current density A m-2 3.4
z Electrons per reaction - 2.6
A Area m2 -
E Potential VSHE 3.3
E° Standard electrode potential VSHE 2.7
E0 Equilibrium potential VSHE 2.10
F Faraday constant C mol-1 -
I Current A 3.2
R Universal gas constant J mol-1 K-1 -
T Temperature K -
β Tafel slope V 3.5
η Polarisation V 3.3

EJR 40