Czichos H., Saito T., Smith L.E. (Eds.) Handbook of Metrology and Testing

Подождите немного. Документ загружается.

668 Part D Materials Performance Testing

12.7 Hydrogen-Induced Stress Corrosion

Cracking .............................................. 714

12.7.1 Electrochemical Processes ............ 715

12.7.2 Theories of H-Induced Stress

Corrosion Cracking....................... 716

12.7.3 Environment

and Material Parameters.............. 717

12.7.4 Fractographic and Mechanical

Effects of HISCC............................ 717

12.7.5 Test Methods .............................. 718

12.8 High-Temperature Corrosion ................. 718

12.8.1 Main Parameters

in High-Temperature Corrosion..... 718

12.8.2 Test Standards or Guidelines......... 719

12.8.3 Mass Change Measurements ......... 721

12.8.4 Special High-Temperature

Corrosion Tests............................ 729

12.8.5 Post-Test Evaluation of Test Pieces 731

12.8.6 Concluding Remarks .................... 732

12.9 Inhibitor Testing and Monitoring

of Efﬁciency ......................................... 732

12.9.1 Investigation and Testing

of Inhibitors ............................... 733

12.9.2 Monitoring of Inhibitor Efﬁciency.. 734

12.9.3 Monitoring Inhibition

from Corrosion Rates ................... 736

References .................................................. 738

12.1 Background

According to ISO 8044 [12.2] corrosion is deﬁned as an

interaction between a metal and its environment that re-

sults in changes in the properties of the metal, and which

may lead to signiﬁcant impairment of the function of

the metal, the environment, or the technical system,

of which these form a part. This deﬁnition has solved

a conﬂict because previously the term corrosion had

been used to mean the process, results of the process

and damage caused by the process. In most cases the

interaction between the metal and the environment is

an electrochemical reaction where thermodynamic and

kinetic considerations apply. From a thermodynamic

point of view the driving force as in any electrochem-

ical reaction is a potential difference between anodes

and cathodes in a short-circuited cell.

The result of corrosion is a corrosion effect which

is generally detrimental and may lead to loss of ma-

terial, contamination of the environment with corrosion

products or impairment of a technical system. It is im-

portant to notice that corrosion damage resulting from

an attack to the metal has to be distinguished from cor-

rosion failure which is characterized by the total loss of

function of the technical system. Corrosion protection

measures are available to inﬂuence corrosion process

with the objective of avoiding corrosion failures.

Corrosion and protective measures against corro-

sion result in costs. Many attempts have been made

to estimate the ﬁnancial expenditure to the commu-

nity caused by corrosion. These include the costs which

can arise in the form of corrosion protection mea-

sures, through replacement of corrosion-damaged parts

or through different effects deriving from corrosion,

such as shut-down of production or accidents which

lead to injuries or damage to property. Several estima-

tions have arrived at the conclusion that the total annual

corrosion costs in the industrialized counties amount to

about 4% of the gross national product. Part of these

costs is unavoidable since it would not be economi-

cally viable to carry out the necessary precautions to

eliminate completely corrosion damage. It is, however,

certain that one can reduce losses considerably solely

by better exploiting the knowledge we have today and,

according to one estimate, about 15% of corrosion costs

are of this type [12.3].

Due to the technical, scientiﬁc and economic impor-

tance of the problem numerous textbooks [12.1, 4–18]

and publications in journals or proceedings from confer-

ences are available. As mentioned before the corrosion

reaction is in most cases of electrochemical nature.

Therefore many of the published information deal with

the electrochemical reactions and especially the kinet-

ics of the corrosion process. For the engineer durability

aspects in a technical plant or equipment or the safety

of a process play a more important role. However, both

approaches follow the same route in yielding informa-

tion about the corrosion rate of a material in a speciﬁc

environment [12.4].

The schematic diagram in Fig. 12.1 [12.4]showsthe

various reactions that can occur sequentially and simul-

taneously in the corrosion process. Material transport

and chemical reactions can supply or remove impor-

tant reaction components. In addition to adsorption or

desorption, a phase-boundary reaction occurs which

mostly involves electrochemical reactions and which

can be affected by external currents. An example is the

formation of hydride on lead where the partner to the

Part D 12.1

Corrosion 12.1 Background 669

Medium Phase boundary Material

Transport

(diffusion,

convection,

migration)

Pre- and

post-reactions

General degradation

(demaging resultant reactions,

contaminants, etc.)

Weight loss

uniform

localized

Microstructural

changes,

internal cracks

Adsorption

and

desorption

Anodic reaction

Cathodic reaction

Chemical reaction

Absorption

Liquid metal reactions

Metal-physical

and

chemical

processes

in the

material

External and galvanic currents

Mechanical stresses

Cracking

Corrosion effects

Corrosion damage

Requirements for

the system

Fig. 12.1 Flow diagram of corrosion processes and types of damage (after [12.4])

hydrogen reaction can arise from the cathodic process.

Hydrogen can also penetrate the metal microstructure,

where it can have a physical or chemical effect, causing

a degradation of the mechanical properties of the ma-

terial. In this type of corrosion cracks can form during

mechanical testing, leading to fracture without the loss

of any metal.

12.1.1 Classiﬁcation of Corrosion

Numerous concepts have been applied to classify cor-

rosion into categories. Due to the complex interaction

between a material and its environment different clas-

siﬁcation schemes have developed to classify corrosion

by the type of attack, the rate of attack, the morphol-

ogy of the attack, or the properties of the environment,

which may change in the ongoing corrosion process.

Type of damage have been predominantly used.

This can be classiﬁed as uniform or localized metal re-

moval, corrosion cracking or detrimental effects to the

environment. Local attack can take the form of shallow

pits, pitting, selective dissolution of small microstruc-

ture regions or cracking. It is usual, where different

results of reactions lead to deﬁnite forms of corrosion

effects, to classify them as particular types of corrosion.

The most important are uniform corrosion, pitting cor-

rosion, crevice corrosion, intergranular corrosion, and

those with accompanying mechanical-stress corrosion

and corrosion fatigue. The two latter are most feared.

Most failures arise from pitting corrosion. Uniform cor-

rosion almost always occurs in practice but seldom

leads to failure.

From [12.19] it can be taken that there are generally

eight forms of corrosion.

1. General (uniform) corrosion;

2. Localized corrosion;

3. Galvanic corrosion;

4. Cracking phenomena;

5. Intergranular corrosion;

6. Dealloying;

7. High-temperature corrosion.

The eight forms of corrosion can be divided into three

categories.

Group I Those readily identiﬁed by visual examina-

tion (forms 1, 2 and 3);

Part D 12.1

670 Part D Materials Performance Testing

Group II Those which may require supplementary

means of examination (forms 5, 6 and 7);

Group III Those which usually should be veriﬁed by

microscopy, optical or scanning electron

microscope, although they are sometimes ap-

parent to the naked eye.

The individual phenomena are illustrated in Fig. 12.2.

12.1.2 Corrosion Testing

The safe use of materials in technical plants, structures

or equipment requires sufﬁcient information for assess-

ing corrosion risks – especially if mechanical stresses

are superimposed on the electrolytic load and failure

of components due to stress corrosion cracking (SCC)

Group I

General attack

Original surface

Velocity phenomena

Intergranular attack

Erosion

Cracking phenomena

Static

stress

Dynamic

stress

Scale

Fissure

High-temperature attack

Stress corrosion cracking

Corrosion fatigue

Scaling

Internal attack

Dealloying attack

Layer Plug

Cavitation

Fretting

Load

Vibration

Weld

Weld decay

Exfoliation

HAZ

Bubbles

Flow

Localized attack

Galvanic attack

Base

metal

Noble

metal

Localized corrosion

Pitting

Crevice corrosion

Group II Group III

Fig. 12.2 Types of corrosion

cannot be excluded. If sufﬁcient knowledge about the

corrosion behavior is not available then tests to deter-

mine this under practical conditions will be required.

In this context a distinction must be made between

the terms corrosion investigation and corrosion test.

According to the International Standards Organization

(ISO) standard ISO 8044 [12.2] a corrosion investiga-

tion includes corrosion tests and their evaluation and is

directed towards the following objectives.

•

The explanation of corrosion reactions,

•

Obtaining knowledge on corrosion behavior of ma-

terials under corrosion load,

•

Selecting measures for corrosion protection.

This enables statements about the properties of a corro-

sion system to be made. A corrosion test is a special

Part D 12.1

Corrosion 12.2 Conventional Electrochemical Test Methods 671

case of a corrosion investigation; the corrosion load

and the assessment of the results being prescribed by

certain regulations (standards, test sheet) and/or agree-

ments such as, for example, delivery conditions.

Corrosion tests mainly help in quality control dur-

ing production and further treatment of materials as well

as for the assessment of corrosion protection measures.

The signiﬁcance of the corrosion tests selected for an

investigation differs. While corrosion tests under loads

that are characteristic of service conditions give the best

estimation of the real corrosion behavior, tests with in-

creased corrosion load or accelerated corrosion tests are

less suitable to give a reliable prognosis of behavior in

practice as the corrosion mechanism could differ from

that under practical conditions. Nevertheless, acceler-

ated corrosion tests are more suitable for determining

certain material properties (e.g. resistance to intergranu-

lar corrosion or stress corrosion cracking) although they

can be reliably used only if sufﬁcient experience exists

for the transfer of the test results to practical situations.

12.2 Conventional Electrochemical Test Methods

Corrosion is the degradation of materials by gaseous

or liquid environments. Atmospheric corrosion involves

the inﬂuence of both media on the surface of a solid, of-

ten with alternating wet and dry periods. Any kind of

materials may be attacked, including metals, semicon-

ductors and insulators as well as organic polymers. This

chapter describes corrosion of metals in electrolytes and

therefore some fundamentals of electrochemistry and

related test methods are discussed.

Large losses are caused by corrosion in modern

industrial societies. One estimate is that 4.0% of the

gross national product is lost by corrosion, a third of

which may be avoided if all knowledge is taken into

account. Traditionally corrosion is subdivided into at-

tack by dry and hot gases and by electrolytes. In this

section corrosion in aqueous electrolytes only is de-

scribed, although solutions with organic solvents may

play an important role in praxis. Corrosion is a reaction

at the metal/electrolyte phase boundary. Although the

bulk properties of the solid and the solution may play an

important role for the degradation of the materials, the

properties of the metal surface and the reactions at the

solid/electrolyte interface are most important. Therefore

corrosion science and corrosion engineering apply the

concepts of electrochemistry and electrode kinetics. In

Sect. 12.2.1 a basic introduction to the equipment and

experimental procedures is given. Corrosion involves

electrode processes which are ruled by the electrode

potential and thermodynamic and kinetic factors. For

this reason the equilibria and the related thermody-

namic properties of the metal/electrolyte interface and

their electrode kinetics are discussed. In Sect. 12.2.3

a short introduction to electrochemical thermodynam-

ics and equilibrium is given followed by an introduction

to electrode kinetics in Sect. 12.2.5. Finally this chap-

ter is completed by a description of the application of

electrochemical transient measurements and some elec-

troanalytical methods.

12.2.1 Principles of Electrochemical

Measurements and Deﬁnitions

Electrochemical measurements are performed in an

electrochemical cell, which contains a three-electrode

arrangement in its most simple form Fig. 12.3. The

out

Salt

CE WE

Fig. 12.3 Standard electrochemical cell with working elec-

trode (WE), counter-electrode (CE), reference electrode

(RE), Haber–Luggin capillary (HL), magnetic stirring

(MS), water jacket with thermostat water (TW) and nitro-

gen inlet

Part D 12.2

672 Part D Materials Performance Testing

Table 12.1 Some commonly used reference electrodes and their related standard potential E

from [12.20]

Electrode E

/V Nernst equation

Calomelelectrode:Hg/Hg

/Cl

−

0.268 E = E

– 0.059 log [Cl

−

]

electrode: Hg/Hg

/SO

2−

0.615 E = E

– 0.059 log [SO

2−

]

HgO electrode: Hg/HgO/OH

−

0.926 E = E

– 0.059 log [OH

−

]

AgCl electrode: Ag/AgCl/Cl

−

0.222 E = E

– 0.059 log [Cl

−

]

PbSO

electrode: Pb/PbSO

/SO

2−

−0.276 E = E

– 0.029 log [SO

2−

]

working electrode (WE) of the material under study,

the counter electrode (CE) of a metal which is not at-

tacked by the electrolyte, in most cases a platinum or

gold electrode, and ﬁnally the reference electrode (RE)

with an electrolytic contact to the bulk electrolyte via

the Haber–Luggin (HL) capillary. Usually such an elec-

trochemical cell contains an inlet for protective gases

such as nitrogen or argon, a stirrer for the electrolyte

and a water jacket to keep the electrolyte at the de-

sired temperature using a thermostat. Numerous special

designs for electrochemical cells have been applied de-

pending on the special problem under study. In some

cases small electrolyte vessels are used as e.g. for mi-

croscopic in situ investigations or in situ measurements

with a scanning tunneling microscope (STM). Another

approach is cells that restrict the investigated surface

area to the size of droplets in the μm range. In all these

cases a three-electrode set up is used with the excep-

tion of in situ STM investigations. Here two working

electrodes are required, the material under study and the

STM tip. The potentials of both electrodes have to be set

independently and therefore a bipotentiostat is required.

A similar situation with two or even more working

electrodes exists for a rotating ring-disc electrode. This

analytical method will be described in Sect. 12.2.11.

The reference electrode is usually an electrode of the

second kind. Its potential is ﬁxed by a weakly soluble

compound of the metal and an anion of the solution.

Table 12.1 gives some examples of mercury and silver

ΔU

Fig. 12.4 Block diagram of potentiostat with operational ampliﬁer

(OA), potential setting ΔU, resistance R for current measurement

as RI

and connections to working electrode (WE), counter-

electrode (CE) and reference electrode (RE)

electrodes [12.20]. These reference electrodes are easy

to prepare and to handle and their electrode potential is

very stable.

For most experiments the electrode potential is

ﬁxed by an electronic potentiostat and the current is

measured. Potentiodynamic polarization curves involve

a linear change of the electrode potential with time. Po-

tentiostatic measurements keep the potential constant.

Potentiostatic transients require its rapid change. With

a fast potentiostat and some precautions a potential

change can be achieved in a few microseconds so that

fast electrode processes can be followed with a high

time resolution starting in the μs range. In most cases

of corrosion studies a ms resolution of the current mea-

surement is sufﬁciently fast.

Figure 12.4 explains with a simple block diagram

the characteristics of a potentiostat built with an op-

erational ampliﬁer (OA). The reference electrode, RE,

is given directly to the inverting input of OA and the

desired electrode potential is determined by the set-

ting of the voltage ΔU from a power source fed into

the noninverting input. The working electrode WE is

set to common. The high gain of the operational am-

pliﬁer (factor 10

) ensures a negligibly small deviation

of both inputs. Thus RE has a potential difference ΔE

to WE equal to ΔU. The input resistance is very high

(10

Ω) so that almost no current ﬂow occurs into the

OA. Its output resistance is very low so that it may sup-

ply a sufﬁciently high current I

to WE via CE and the

electrolyte which may be required for the rate of the

electrode processes related to the chosen potential. I

measured via its ohmic drop RI

at the resistor R. This

simple electronic circuit stabilizes the chosen electrode

potential automatically, independent of the required cur-

rent density I

. However, a good potentiostat contains

additional booster units to provide high currents and to

improve the setting of potential changes with a fast re-

sponse. Usually potentiostats contain additional circuits

built with operational ampliﬁers for necessary options

such as, e.g., an adder which permits the addition of var-

ious potentials forming a potential program with ramps

and pulses for potentiostatic transients. A special circuit

Part D 12.2

Corrosion 12.2 Conventional Electrochemical Test Methods 673

compensates for ohmic drops within the electrolyte be-

tween WE and HL which may occur in the case of high-

current densities and electrolytes with low conductivity.

Galvanostatic measurements require the setting of

a current I

while the change of the electrode potential

ΔE is followed with time. This is achieved by a gal-

vanostat (constant-current source) depicted as a simple

block diagram in Fig. 12.5. I

is set via a proportional

voltage ΔU from a built-in power source. ΔU is given

to the noninverting input of the operational ampliﬁer

OA, which is compared to the ohmic drop RI

across

the resistor R at the output. This voltage drop passes

a differential ampliﬁer (DA), which avoids the setting of

a second point of the circuit to common. As both inputs

of OA have to be at the same potential as a consequence

of its high magniﬁcation, I

is adjusted to the desired

value RI

= ΔU. Usually galvanostats are built in to

potentiostats. Appropriate voltages ΔU may be given

to the input with speciﬁc time characteristics to permit

also galvanodynamic measurements including galvano-

static transients controlled by appropriate voltage pulses

given to OA.

12.2.2 Some Deﬁnitions

An electrochemical cell usually consists of two elec-

trodes. Figure 12.6 givesasanexampleaniron

electrode in a Fe

solution in contact with a hydro-

gen electrode via a diaphragm. The H

electrode

consists of a platinum electrode in a solution of a given

pH with the introduction of pure hydrogen. If the solu-

tion is 1 M for the H

concentration, i. e. if pH is 0, and

the gas pressure of H

is 1 atm =1.013 bar, this is the

standard hydrogen electrode (SHE). Potential drops at

the electrode–electrolyte interface cannot be measured

separately in principle. The contact of a voltmeter to

the electrolyte introduces automatically a second inter-

face so that the measured voltage necessarily includes

the potential drop of two electrodes. The potentials of

any electrode is therefore given relative to the SHE.

The electrode potential of the standard hydrogen elec-

trode is thus set to 0 V by deﬁnition. Its value relative

to the zero point of the potential scale in physics, i. e.

the charge-free vacuum level, has been determined to

−4.6 V [12.21,22], so that both scales may be converted

to each other.

There are two main kinds of electrodes, metal/metal-

ion electrodes and redox electrodes. For the ﬁrst case

the electrode reaction requires the transfer of metal ions

across the electrode–electrolyte interface. The Fe/Fe

electrode is an example. Redox electrodes involve the

ΔU

ΔU =RI

ΔE

Fig. 12.5 Block diagram for galvanostat with operational ampliﬁer

(OA), differential ampliﬁer (DA), voltage supply ΔU for current

setting I

and connections to working electrode (WE), counter-

electrode (CE) and reference electrode (RE)

transfer of electrons, which is the case for the hydro-

gen electrode. A Pt electrode within a solution of Fe

and Fe

ions or Fe(CN)

4−

and Fe(CN)

3−

are other

examples. If no reaction occurs, the electrode is in

electrochemical equilibrium, i. e. both reactions – Fe

dissolution or Fe deposition – compensate each other.

The same is the case for the redox reaction.

One has to distinguish anodic and cathodic pro-

cesses. An anodic process involves the transfer of

positive charge from the electrode to the electrolyte,

such as Fe dissolution to Fe

, or the transfer of

negative charge in the opposite direction, e.g. the ox-

idation of Fe

to Fe

. A cathodic process involves

the transfer of positive charge from the electrolyte to

the electrode, such as the deposition of Fe

from the

solution as Fe metal or the transfer of negative charge

in the opposite direction, e.g. the reduction of Fe

. The currents of anodic processes get a positive

–

= 1.013 bar

]=1M

Fe Pt

Fig. 12.6 Electrochemical cell with an Fe/Fe

and

electrode

Part D 12.2

674 Part D Materials Performance Testing

sign, those for cathodic reactions a negative sign. In

conclusion, it is the kind of process which determines

whether an electrode acts as an anode or a cathode. De-

pending on the applied electrode potential and further

electrochemical conditions an electrode may be either

an anode or a cathode.

If anodic and cathodic currents of a redox or

a metal/metal-ion electrode compensate each other the

electrochemical equilibrium is established at the equi-

librium potential E

. A positive deviation η = E −E

leads to an anodic current. η is called the overvoltage.

A deviation in the negative direction with a negative

overvoltage η = E −E

< 0 causes a cathodic reaction.

If two different electrode processes compensate each

other, such as an anodic Fe dissolution and cathodic

hydrogen evolution with vanishing current in the ex-

ternal circuit, the electrode is at its rest potential E

A positive deviation π = E −E

> 0 is called a positive

polarization and π = E −E

< 0 a negative polariza-

tion.

12.2.3 Electrochemical Thermodynamics

The equilibrium potential E

is related to the change of

the Gibbs free energy for standard conditions ΔG

the related electrode process, i. e. for activities of so-

lutes of a = 1 M and gas pressures of p = 1.013 bar.

As all electrode potentials are referred to the stan-

dard hydrogen electrode its reaction has been chosen

as a compensating process for thermodynamic calcula-

tions. Equations (12.1, 12.2) describe the process of an

electrochemical cell with the Fe/Fe

and the H

electrode; (12.3) describes their combination. ΔG

for

the electrode process of interest has to be calculated for

the cathodic direction and the compensating process of

the reference electrode in the anodic direction to avoid

mistakes in the sign (12.4). The relation between ΔG

and the electrode potential is given by (12.5). The con-

centration dependence of the equilibrium potential E

given by the Nernst equation (12.6, 12.7) for both elec-

trodes. Equation (12.7) describes the pH dependence of

the hydrogen electrode

−

+Fe

→Fe , (12.1)

→2H

+2e

−

, (12.2)

+2H

→Fe +2H

, (12.3)

ΔG

298



i,i

=−G

=78.9kJmol

−1

(12.4)

ΔG

298

=−nFE

, (12.5)

=−

ΔG

298

=−

78.94 × 10

=−0.409 V , (12.6)

= E

ln a(Fe

)

=−0.409+

0.059

lg a(Fe

)/V , (12.7)

=0+



a(H

)

p(H

)



=−0.059 pH/V , (12.8)

with pH =−lg



a(H

)



. (12.9)

Electrode reactions that involve solid compounds are

treated in a similar way. As an example the equilibrium

of Fe with Fe

isgivenin(12.10–12.13). This equi-

librium describes the electrochemical formation of an

oxide layer at the metal surface. These equilibriums are

important for the calculation of potential–pH diagrams,

so called Pourbaix diagrams, which will be discussed in

the next section.

+8H

+8e

−

→3Fe+4H

O (12.10)

ΔG

298

=1013.23−4(237.13) =64.71 kJ/mol

(12.11)

=−

64.71 × 10

=−0.084 V (12.12)

=−0.084+

0.059



a(H

)



=−0.084−0.059 pH/V (12.13)

Thermodynamic data and especially standard Gibbs

free energies ΔG

for the formation of compounds

from the elements are listed in special publications for

T = 298 K, but also for other temperatures [12.23, 24].

They permit the calculate of data for electrochemical

equilibriums as shown for (12.1–12.13). For most equi-

libriums the standard electrode potentials E

are listed.

Their dependence on the concentrations of the reacting

species, i. e. their Nernst equations, are used to calculate

potential–pH diagrams [12.25]. Figure 12.7 shows the

diagram for iron, which is a very important example in

corrosion. The horizontal line (1) at E =−0.589 V de-

scribes the pH-independent Fe/Fe

equilibrium with a

−6

M, i. e. a negligibly small concentration of Fe

At E = 0.771 V the Fe

/Fe

equilibrium is given

as a further horizontal line (2). Line (3) depicts the

pH-dependent formation of Fe

according to (12.10–

12.13). Line (4) corresponds to the Fe

/Fe

equilibrium. Lines (5) and (6) refer to the oxidation

Part D 12.2

Corrosion 12.2 Conventional Electrochemical Test Methods 675

E (V)(SHE)

Immunity

Passivation

Flade

potential

Corrosion

1.8

1.6

1.4

1.2

0.8

0.6

0.4

0.2

–0.2

–0.4

–0.6

–0.8

–1

–1 1211109876543210–2 13

–1.2

–1.4

–1.6

2.2

–1.8

Fig. 12.7 Potential–pH diagram of Fe with equilibria (1–

7) and (a) and (b) and domains of corrosion, immunity

and passivation, as described in the text, Flade potential

describing passivation in acidic electrolytes

of soluble Fe

to Fe

and Fe

respectively. The

vertical line (7) describes the potential-independent

dissolution equilibrium of Fe

/Fe

that is related

to a speciﬁc pH. According to this diagram one has

to distinguish three main ﬁelds. Corrosion, i. e. metal

dissolution, should occur where soluble products are

stable, i. e. at the upper left end of the diagram Fig. 12.7.

At the lower end of the diagram one has the ﬁeld of

immunity where Fe is the stable species. In the upper

right part the surface should be protected by anodic ox-

ide layers. Here thin anodic ﬁlms should form at the

metal surface, which block further oxidation and lead to

passivity. The lines (a) and (b) describe the equilibrium

potential of the hydrogen and oxygen electrode, which

are important redox systems for aqueous corrosion.

These diagrams have been calculated at 298 K for all

elements, metals, semiconductors and elements with in-

sulating properties [12.26,27]. More recently they have

also been calculated for elevated temperatures. They

provide an excellent possibility to get a ﬁrst idea of the

corrosion properties of a metal. However, one should

always keep in mind that they are calculated on the ba-

sis of thermodynamic data only. Any kinetic properties

are not included. This means that conclusions on the

basis of these diagrams may be misleading. Pourbaix

diagrams tell us whether a reaction may occur or not.

Whether the predictions really happen or not depends

on the kinetic parameters of the related reactions. A typ-

ical exception is the passive behavior of iron in strongly

acidic electrolytes such as 0.5MH

or 1 M HClO

Iron should dissolve at potentials above E =−0.409 V.

However, passivation occurs when the potential gets

above E = 0.58–0.059 pH[V]. This condition is de-

picted as a dashed line in Fig. 12.7. Surface analytical

investigations have shown that Fe forms a poreless few-

nm-thick layer of Fe

a spinell-type oxide which is

very slowly dissolving in these acidic electrolytes. De-

tailed electrochemical studies come to the conclusion

that the transfer of Fe

cations from the oxide surface

to the electrolyte is an extremely slow process, although

the oxide layer is far from its dissolution equilibrium.

As a consequence this surface ﬁlm passivates the metal

underneath. It has been shown that the passivation po-

tential may be explained by the thermodynamic data for

the anodic oxidation of Fe

to Fe

[12.28]. Appar-

ently the structure and the electrochemical properties

of Fe

provide sufﬁcient stability for dissolution,

i. e. the activation energy to transfer Fe

ions from

the potential well of the O

2−

matrix of an oxide is

very high. Similar arguments hold for passive layers

onCr.HereaCr

ﬁlm of only 1–2 nm thickness

dissolves in acidic solutions so extremely slowly that

Cr is protected very effectively against corrosion in

strongly acidic electrolytes [12.29–31]. This is also the

case for Cr-containing alloys, such as Ni/Cr or Fe/Cr.

The strong bonds between Cr

ions and its ligands

in its complexes are well known from inorganic chem-

istry. The exchange of ligands is so slow that many of

these Cr

compounds are almost insoluble, although

this description is misleading; they dissolve extremely

slowly, which has the same effect. As an example the

extremely slow dissolution of CrCl

should be men-

tioned, which leads to its description as an insoluble

compound. However, this behavior is only caused by an

extremely slow dissolution rate [12.31]. This becomes

faster at elevated temperatures when the related high

activation energy for the exchange of ligands has been

overcome.

In conclusion the Pourbaix diagrams provide ex-

cellent initial information on the corrosion behavior of

metals on the basis of thermodynamic data. However,

a reliable discussion has to include kinetic parameters

Part D 12.2

676 Part D Materials Performance Testing

–1.0 0 1.0 2.0

Au/Au(CN)

–

Au/Au

E (V)

I (A/cm

)

Cu/Cu

Fe/Fe

Fe Cu

Fig. 12.8 Schematic diagram of E

, E

, and anodic dissolution of

Fe, Cu and Au and the shift of E

of Au by complexation with

cyanide with oxygen reduction as counter-reaction

Preceding reaction

Adsorption

Charge transfer reaction

Desorption

Following reaction

Diffusion

Reaction overvoltage η

Charge transfer overvoltage η

Reaction overvoltage η

Diffusion overvoltage η

η =η

+ η

Fig. 12.9 Elementary reaction steps of an electrode process with

related overvoltages

in order to avoid misleading conclusions which might

contradict experimental results.

12.2.4 Complex Formation

The formation of strong complexes of dissolved cations

with anions has an important inﬂuence on the dissolu-

tion of metals; hey shift the concentration of free cations

to very small values and thereby shift the equilibrium

potential of the related metal/metal-ion electrode to

very negative values [12.20, 32, 33](Fig.12.8). A well-

known effect is the strong complexation of Au

−

or of Au

by CN

−

as depicted in (12.14, 12.15).

The standard potential of Au dissolution is shifted by

−

from 1.42V to 0.994 V or by CN

−

from 1.68 V

to −0.60 V, which is a consequence of the small value

of the dissociation constant K

= 2.2×10

−22

of the

AuCl

−

and K

= 2.3×10

−39

for the Au(CN)

−

com-

plex. Therefore Au dissolution occurs in a 1 :3 mixture

of HCl and HNO

(aqua regia) whereas pure HNO

does not attack. Similarly Au may be dissolved in CN

−

containing solution by air oxidation (Fig. 12.8). The

latter example is shown by (12.16, 12.17) in detail, and

has been applied for extraction of Au from minerals.

AuCl

3−

→Au

+4Cl

−

, K

=2.2×10

−22

(12.14)

Au(CN)

−

→Au

+2CN

−

, K

=2.3×10

−39

(12.15)

a(Au

) = K

a(AuCN)

−

(CN

−

)

(12.16)

E = E

ln a(Au

)

=1.68+0.059 log K

+ 0.059 log

a[Au(CN)

−

]

(CN

−

)

(12.17)

AuCN

=1.68−0.059 log 2.3×10

−39

=1.68−2.28 =−0.60 V (12.17a)

Complexation of cations also plays a role in the en-

hanced dissolution of passivating layers. The formation

of complexes of cations with halides has been proposed

as a mechanism for the breakdown of passivity and lo-

calized corrosion [12.29–31].

12.2.5 Electrochemical Kinetics

The kinetics of electrode processes inﬂuence decisively

corrosion phenomena. As usual in kinetics the rate of

a chemical reaction at the electrode surface is ruled by

the activation energy of the slowest of a sequence of

elementary reaction steps. This could be the charge-

transfer process or any preceding or following reaction

step. Any electrode process requires the transport of

the reacting species to the electrode surface, a possi-

ble chemical reaction within the solution in front of the

electrode, the charge-transfer process, a possible fol-

lowing chemical reaction and ﬁnally the transport of

the product from the electrode to the bulk electrolyte.

In addition adsorption and desorption processes may

be involved. Any of those steps may be the slowest

and thus may be rate-determining for the overall reac-

tion. Figure 12.9 describes the sequence of elementary

reaction steps, including the charge-transfer reaction,

that together form the overall reaction. Each of these

steps may contribute with its special overvoltage η

the total overvoltage η =



of the whole reaction

(Fig. 12.9). Usually one reaction step is the slowest and

thus rate-determining for the overall reaction. Its over-

voltage η

dominates η.

12.2.6 The Charge-Transfer Overvoltage

If the charge transfer is rate-determining the related

current density follows an exponential relationship, the

Butler–Volmer equation. As mentioned earlier an ion or

Part D 12.2

Corrosion 12.2 Conventional Electrochemical Test Methods 677

an electron may be the transferred species. This process

occurs within the Helmholtz layer, i. e. within a dis-

tance of the radius of solvated ions of typically less than

1 nm. The charge transfer of (12.18) occurs via a tunnel-

ing process that is seriously inﬂuenced by the potential

drop Δφ = φ

−φ

within this layer. Combining the

usual rate equation with Faraday’s law one obtains a re-

dox process (12.19) which includes the two opposite

parts due to the anodic and the cathodic reactions. At

the equilibrium potential these are equal and compen-

sate each other. F is the Faraday constant and n is the

number of charges transferred for this reaction, which

is usually 1 and may be larger if several one-electron

processes occur in a sequence, k

and k

−

are the rate

constants for the anodic and cathodic reaction steps and

c(Red) and c(Ox) are the concentrations of the oxidized

and reduced species of the redox system involved. For

metal dissolution equivalent (12.20, 12.21) hold where

is the surface coverage for metal atoms which is

proportional to their surface concentration, and c(Me

)

is the concentration of the cations within the electrolyte.

One has to distinguish in general between the charges

n which are involved in the total electrode reaction of

(12.18, 12.20), and the charge z of the species which

is transferred across the electrode–electrolyte interface

during the charge-transfer process. In the case of a sim-

ple metal dissolution, according to (12.20) n equals z.If

the dissolving cation forms a complex with anions from

the solution in a reaction before the charge-transfer step

these number may be quite different. For a redox reac-

tion z equals 1. If however, more than one electron is

transferred in a fast sequence of several steps, i. e. in the

case of a charge-transfer reaction with more electrons n

will be larger

Red →Ox+ne

−

, (12.18)

i = nF



−dc(Red)

−

−dc(Ox)



=nF



c(Red)−k

−

c(Ox)



, (12.19)

Me →Me

+ne

−

, (12.20)

i = nFk

−nFk

−

c(Me

) . (12.21)

The rate equations of both types of charge-transfer

processes may contain the concentration of additional

species that are involved in their mechanisms. An ex-

ample is the catalysis of active Fe dissolution by OH

−

ions. The exponent of the concentrations of OH

−

,i.e.

its electrochemical reaction order, is 1.5 [12.34]. Re-

action orders may be any number for a given reaction

depending on the details of its mechanism.

Energy

Helmholtz layer

Distance

ΔE

0,A–

ΔE

0,A+

ΔE

A+

ΔE

A–

FΔφ

αFΔφ

(1 – α)FΔφ

αFΔφ

Distance

Metal Electrolyte

Fig. 12.10 Activation energy ΔE

0,A

and its change to ΔE

for the anodic (+) and cathodic reaction (−) due to the po-

tential drop Δφ across the electrode–electrolyte interface

As usual, both rate constants k

and k

−

depend

exponentially on their activation energy ΔE

0,A+

and

ΔE

0,A−

(12.22, 12.23). According to Fig. 12.10 the ac-

tivation energies ΔE

A,+

and ΔE

A,−

for the transfer of

charged species across the electrode–electrolyte inter-

face are inﬂuenced by the potential drop Δφ.However,

only a fraction of Δφ will modify ΔE

0,A+

. The part

α takes into account the fraction of Δφ which is kineti-

cally effective, i. e. which part will act on the species up

to the maximum of the energy–distance curve; α usu-

ally has a value of 0.2–0.8. If α refers to the anodic

process 1 −α will hold for the cathodic reaction. Thus

the activation energy ΔE

A,+

will change by −αzFΔφ

for the anodic and ΔE

A,−

by +(1 −α)zFΔφ for the

cathodic process, respectively; z is the charge of the

species which is transferred across the double layer,

where z = 1 for electrons, i. e. for charge-transfer pro-

cesses, and z may be larger for metal ions, e.g. z = 2

for Fe

. Equations (12.22, 12.23) describe the inﬂu-

ence of Δφ on the rate constants of both processes

and (12.24) gives the full kinetic equation for a re-

dox reaction. An equivalent equation is obtained for

a metal/metal-ion electrode. The unknown potential

drop Δφ within the double layer is replaced by the elec-

trode potential E and the drop Δφ

within the double

layeroftheSHE (Δφ = E −Δφ

). As Δφ

is con-

stant, any changes of Δφ will equal that of the electrode

Part D 12.2