Diapositive 1 INPG. Oct. 2003 Corrosion in brief

Corrosion in brief by Bernard Baroux Institut National Polytechnique de Grenoble + Arcelor R&D • Introduction to Corrosion processes • Corrosion Electrochemistry • Passivity and Passive films • Stability and Breakdown of passivity • Pitting corrosion • Crevice corrosion • Intergranular corrosion of stainless steels • Concluding remarks 1

Corrosion is an irreversible phenomenon which results from the basic thermodynamic characteristics of the materials and the nature of their environment. In this lecture we will look at the water corrosion phenomena applicable to metals and alloys, but excluding both high temperature oxidation phenomena and corrosion of non-metallic materials.

Diapositive 2 INPG. Oct. 2003 Corrosion in brief

Experimental Evidences • Formation of rust and other corrosion products

Metallurgical effects • Electrochemical dissolution

M →M+++2e-

⇒ loss of weight and pollution… 2

Corrosion phenomena may concern the most of the devices of our day to day life, when they are immersed insufficiently severe environments, or made from an insufficiently resistant material (which may happen for cost consideration, even when the appropriate solution is known, but expansive). Corrosion risks may also arise from inappropriate secondary processing conditions, such as welding or drawing. The conjunction The corrosion of metals takes two different forms: 1) Dissolution of the metal and therefore loss of matter. This consists of passage of the metal cations into an aqueous solution (anodic dissolution), the electrons p^roduced by this dissolution being consumed by a cathodic reaction. As such, the dissolution of metals differs greatly from ordinary dissolution such as when sugar is added to water. Aqueous corrosion is above all an electrochemical phenomenon. 2) The formation of rust, which is the common name given to certain iron oxides. In this case, the product of corrosion is not soluble but solid and often considered unsightly.

Diapositive 3 INPG. Oct. 2003 Corrosion in brief

The oxidation of metals (example of a divalent metal)

M → M + + + 2e − Dry oxidation

Passivation Anodic dissolution Precipitation of an oxide or a hydroxide (rust)

1 M + O2 → MO 2

1 O2 + 2e − → O − − 2 M + + + O − − → MO

M + H 2O → MO + 2 H + + 2e − M + nH 2O → M + + ( H 2O ) n + 2e −

M + + + H 2O → MO + 2 H + ou

M + + + 2 H 2O → M (OH ) 2 + 2 H + 3

The corrosion of metals may take different forms: Oxidation in oxygen containing atmospheres at high temperatures (higher than the water dew point in this atmosphere). This produces a gain of matter (the oxygen trapped in the oxide) and will be investigated later. Morever, It is well known that most metal ores are found in their oxide form. However, the overall reaction involving formation of the oxide splits into several simple reactions including the dissociation of its cations and electrons. An oxide can also form by interaction with water (passivation reaction). In fact, this consists of both an oxidation reaction and an acid-base reaction which may include several steps. In certain conditions, metal cations are transferred to solution in hydrated form (anodic dissolution), resulting in a loss of matter. In aqueous corrosion, the electrons produced by the anodic dissolution reaction having to be consumed by a cathodic reaction. As such, the dissolution of metals differs greatly from ordinary dissolution such as when sugar is added to waterfor instance. Aqueous corrosion is above all an electrochemical phenomenon. Lastly, the cations anodically dissolved may cause hydrolysis, leading to the formation of a hydroxide or an oxide which can precipitate in the form of rust, which is the common name given to certain iron oxides and often considered unsightly. This 3 forms have in common to initiate from a chemical oxidation process , namely deelectronation , involving an oxidising element (for example oxygen, or ferric iron) which is able to trap the electrons from a reducing species (for instance oxhydrile ion, or ferrous ion). The tendency of a metal to be oxidized is related to its capacity to combine with the oxidizing agent.

Diapositive 4 INPG. Oct. 2003 Corrosion in brief

The role of water The oxygen in the iron oxides forming the rustis drawn essentially from the molecules of water and not at all or only very slightly from the oxygen in the air, as is most often thought (the oxygen only encourages the cathodic reaction as will be explained later). In temperate climates, atmospheric corrosion may be considered a water corrosion phenomenon due to the presence of a film of water on the surface of the metal which generally constitutes a cold surface encouraging the formation of condensation.

4

The predominant role of water in corrosion phenomena must be underlined. The oxygen in the iron oxides is drawn essentially from the molecules of water and not at all or only very slightly from the oxygen in the air, as is most often thought (the oxygen only encourages the cathodic reaction as will be explained later). In our temperate climates, atmospheric corrosion may be considered a water corrosion phenomenon due to the presence of a film of water on the surface of the metal which generally constitutes a cold surface encouraging the formation of condensation. Moreover, the example of very old iron parts found in hot and dry climates which have not rusted because the ambient moisture content is too low to cause condensation of water on the surface of a metal illustrates this point.

Diapositive 5 INPG. Oct. 2003 Corrosion in brief

Water assisted Corrosion and passivation processes Passivation M

H

M

O

M

+ M +H

O

H M

O

e-

O

O

H

O

H M

M

+ H+

+ H+

H

+ H+

H

In common

H O

e-

M O + H

MOH+

first step

M++ + OHCorrosion

5

The dipolar water molecule adsorbs to metal surfaces due to either physical or chemical interactions. Adsorbed water can combine with the metal by loosing a proton and an electron, giving birth to a MOH type adsorbed complex. This second step (formation of the absorbed complex) is common between water corrosion and passivation mechanisms described below. In turn, the complex MOH may (or not) oxidise by loosing an electron. The oxidised form MOH+ is unstable and dissolves by giving up the cation to the water environment. This a typical water corrosion process (Bockris mechanism for divalent iron dissolution). 4) If, instead of oxidizing and dissolving, the MOH complex is again deprotonated, we observe the formation of a stable species closely linked to the metal and inhibitor of any subsequent aqueous corrosion (the water is no longer absorbed on the metal but on the MO oxide). This the typical water passivation process

Diapositive 6 INPG. Oct. 2003 Corrosion in brief

Formation of a passive film: ion hopping and water deprotonation

M

H O M O O O O

M M M

O O

H

(a) Formation of the first oxide monolayer

H

(b) Cation hopping and formation of a second layer.

O O

M

H

H H

H

(a) First (hydrated) oxide layer

(c) Vacancy „ migrates toward the metal (d) The process is repeated (Passive Film growth)

H O O O O O

M

O

„

O

M M

H M M

O O

M H

H

(b) ion hopping and film growth 6

In ferrous alloys, the passive film growth proceeds from cation hopping and electromigration of the resulting vacancies toward the metal

Diapositive 7 INPG. Oct. 2003 Corrosion in brief

The developped passive film M

M O O O O O metal

M

O

M

O

M

O

M M

O

M

O O

M M

O O O

M M

O

M film few nm

M

M O

M

O

M

O

M

O

M

O

M

O

H H

M M M

O

H M

M Electrolyte

IHP 7

Diapositive 8 INPG. Oct. 2003 Corrosion in brief

Chloride assisted corrosion process H O

MCl

H e-

M

Cl-

MCl+ + Cl-

M++ + 2Cle-

MCl

• Adsorption of chloride ions, complexation, dissolution 8

In chloride containing environments, the adsorbed water can be replaced by a chloride ion. A similar mechanism to the one described for water corrosion is then possible but the MCl complex can only dissolve (by restituting the chloride ions in solution after dissolution of the cation) and cannot form a protective passivating compound. Finally, the chloride ion functions like a cation pump and the mechanism can be repeated. This mechanism is known as chloride dissolution.

Diapositive 9 INPG. Oct. 2003 Corrosion in brief

Oxidation reactions

2 H 2O + O2 + 4e − → 4OH − metal

2 H 3O + + 2 e − ↔ 2 H 2 O + H 2 Examples of other oxidation processes + ++

−

++

NO

Fe +e →Fe

−

3

→NO2+ 1O2+e 2

−

Etc… 9

Water is not only an acid/base amphoter The hydrogen and oxygen evolution reactions are in fact two aspects of the oxydoreduction of water The deprotonated form of water (hydroxyl ion) behaves like a reducing agent, able to give up electrons with release of oxygen. Its protonated form (hydronium ion) behaves like an oxidizing agent, capable of trapping electrons with release Hydrogen These reactions are controlled by the solution pH and, in the presence of an electron reservoir (for example metal), by the difference of potential between this reservoir and the solution.

Diapositive 10 INPG. Oct. 2003 Corrosion in brief

The electrode potential :

V

A consequence of the charge transfers

Electrolyte VH

x IHP Metal

OHP

Double layer (Helmoltz)

The Charge transfers (electrons and ions) between the metal and the aqueous solution produces a potential difference 10

The charge transfers between metal and solution lead to a difference of potential between the metal and the core of the solution. The figure has been drawn in the case of a net positive excess of charge on the metal surface and of a net negative excess of charge in the electrolyte. One has assumed that the solution excess of charge was concentrated at the immediate vicinity of the metal (outer Helmoltz plan), which is true for sufficiently conductive solutions. In this case , the interface behaves as a capacitor , the two plates of this capacitor beeing the metal surface (Inner Helmoltz plan) and the Outer Helmotz plan Both IHP and IHP form the so called Electrochemical double layer Remark: In the more general case, the charge transfered to the lectrolyte is not concentrated at the interface and form the so called « diffuse layer »

Diapositive 11 INPG. Oct. 2003 Corrosion in brief

M Anodic (electron producing)

M

(+H2O)

MO + 2H++2eM++ + 2e-

and cathodic (electron consuming)

reactions: an outlook

Ox e

-

Red 2 H 3O + + 2 e − ↔ 2 H 2 O + H 2

Ox +e- → Red

4OH − ↔ 2 H 2O + O2 + 4e −

Fe+++ + e- → Fe++

11

The anodic dissolution results in a transfer of cations (positive charges) from the metal to the solution From another hand, the metal forms a reservoir of electrons which it can exchange with other oxidizing species present in solution. This transfer is called "cathodic reaction". In an acid environment, the usual cathodic reaction is the one of so-called hydrogen evolution which reduces proton to hydrogen. In a neutral oxygenated environment, the oxygen is consumed (evolution of oxygen) thereby increasing the local pH. In ferric salt environments, the ferric ions can be transformed into ferrous salts. And so on … In each case, the cathodic reaction operates below the redox standard potential (above it is the opposite reaction which can take place, for instance release of oxygen by decomposition of the water).

Diapositive 12 INPG. Oct. 2003 Corrosion in brief

Corrosion equilibria At a metal electrode, several potential and pH dependent reactions occur

Assuming local equilibria, the mass action law writes

(1) Anodic dissolution: M ↔M+++2e(2) Dissolution or Precipitation of the oxide M+++H2O ↔MO+2H+

V = cst. + 2.3

1 pH + log (c) = cst. 2

(3) Passivation M+H2O ↔ MO+2H++2eWhere

kT log(c) 2q

V + 2.3

kT pH = cst. q

c is the concentration in dissolved cations near the metal interface.

C results from the balance between the dissolution rate and the diffusion flow J from the metal surface to the solution bulk (where c=0)

µc

The dissolution rate is then proportionnal to c 12

Diapositive 13 INPG. Oct. 2003 Corrosion in brief

Pourbaix representation

2

M++

MO

[

Potentiel

M+++H2O ↔MO+2H+

c= M

++

]

(1) V = cst . + 2 . 3 1

M+H2O ↔ MO+2H++2eM

↔M+++2e-

kT log( c ) 2q

1 log (c) = cst. 2 kT ( 3 ) V + 2 .3 pH = cst. q ( 2 ) pH +

3

M pH

13

The typical reactions of metal-water interactions involve both electrons (oxydoreduction) , protons (acidobasicity), and solute cations (solution chemistry) Anodic disssolution: M M++ + ePrecipitation of rust M+++H20 MO + 2H+ Passivation: M+H20 MO + 2H+ + 2e-

Diapositive 14 INPG. Oct. 2003 Corrosion in brief

Pourbaix diagram of Aluminium

(3)

Al+++

(4)

(Al2O3, 3H20)

AlO2-

(5) (2)

Al

(1) Al+++ + 2 H2O ↔ AlO2- +4 H+ (2) Al + 3 H2O ↔ Al2O3 + 6 H+ + 6 e(3) 2 Al+++ + 3 H2O ↔ Al2O3 + 6 H+ (4) Al2O3 + H2O ↔ 2 AlO2- +2 H+ (5) Al ↔ Al+++ + 3 e(6) Al + 2 H2O ↔ AlO2- +4 H+ + 3 e-

(6)

NB: this diagram ise drawn for 1µM/l concentrations in solute species

14

The actual Pourbaix Diagrams are more complex, due to the co-existence of several electrochemical reactions The Al diagramme shows 2 soluble species (3) log (Al+++) = 5,7 - 3 pH (5)

V = -1,663 + 0.02 log (Al+++)

(6)

V = -1,2662 - 0.08 pH + 0,02 log (AlO2-)

Diapositive 15 INPG. Oct. 2003 Corrosion in brief

Pourbaix diagram of Iron

Fe+++

Fe++ Fe2O3

Fe

HFeO2-

15

For Iron, several forms of solute iron (Fe++, Fe+++ , HFeO2--, …) have to be considred , defining several corrosion regions in the diagram…. (see figure).

Diapositive 16 INPG. Oct. 2003 Corrosion in brief

Corrosion kinetics and polarisation curves i

- iK icor

iA V Vcor

iK

iG=iA+iK

The anodic and cathodic reaction rates depend on the electrode potential

Polarisation resistance (V=Vcor) : Rp=dV/di

(Ω.cm2)

16

The intensity of the cathodic and anodic reactions is related to the difference of potential between the metal and the electrolyte. This is set using a potentiostat and the current density is measured; we obtain then a polarisation curve (see above). The overall measured current iG is in fact the sum of the anodic current iA (positive) and the cathodic current iK (negative). This overall current is nil at the free corrosion potential Vcor (no potential applied). The cathodic and anodic currents are then equal in absolute values to the corrosion current icor The corrosion potential and current correspond to the intersection of the individual polarisation curves for the anodic and cathodic reactions. The slope of the curve i(V) at the corrosion potential has a dimension which is the inverse of a resistance (by unit surface) called polarisation resistance. There are different techniques for measuring the polarisation resistance and deducing the corrosion current.

Diapositive 17 INPG. Oct. 2003 Corrosion in brief

The Tafel law CU in H2SO4(0.5M) + CuSO4(0.075M)

The Tafel law:

V = cst + b log i

I = B.exp (2.3 V/ b)

In the cathodic domain: Proton reduction In the anodic domain: Anodic dissolution

17

Diapositive 18 INPG. Oct. 2003 Corrosion in brief

The Corrosion Current

Evans diagram

(Tafel Laws)

i = iA - iK = BA exp 2.3V/bA- BK exp –2.3 V/ bK V=Vcor ⇒ η=O et iA=-iK=icor

i = icor(exp2.3η/ bA− exp-2.3η/ bK) Where

η= V-Vcor is the overvoltage

Stern-Geary law

η→0 ⇒ i ∼ i cor2.3 η(1/ b A+1/ b K) = η / Rp 1/Rp

= 2.3 icor(1/bA+1/bK) 18

Lorsque les deux réactions cathodique et anodique sont controlées par une loi de Tafel, le courant global résultant est une différence d ’exponentielles. On peut reformuler la loi composée ainsi obtenue en rapportant les potentiels au potentiel de corrosion (courant global nul), c ’est à dire en utilisant la surtension définie plus haut. Dès que l ’on s ’éloigne du potentiel de Corrosion, l ’une des deux réactions anodique ou cathodique devient dominante et l ’une des exponentielles peut etre négligée devant l ’autre. La surtension dépend alors logarithmiquement du courant global (Loi de Tafel). La Constuction d’ Evans permet alors de déterminer graphiquement le potentiel et le courant de corrosion Near the corrosion potential, the current varies linearly with the electrode potential. A polarisation resistance is defined. The Stern-Geary law gives then the corrosion current as a fonction of this resistance and Tafel coefficcients

Diapositive 19 INPG. Oct. 2003 Corrosion in brief

Galvanic coupling LE SS N O B LE

i

N O B LE R ic or

V

Vm A

V m = m ixt poten tial B

Icor = i A .S A = -i K .S K

• Coupling of 2 material with 2 different polarisation curves • Or of 2 zones of the same material exposed to different electrolyte compositions 19

2 materials electrochemically different exhibit different polarization curves when placed in tyhe same electrolyte. When these 2 materials are in contact each of the other, it results in an electrochemical coupling and the so called « galvanic corrosion ». Folowing the galvanic corrosion rules, the less noble material specialises in anodic reaction, then corrodes, while the nobler specialises in cathodic reaction, then is protected This is only a schematic view of galvanic corrosion problems and things may be more complex in real situations :We can give the example of corrosion due to contact between two different metals. In certain cases the most noble metal is protected but accelerates the corrosion of the less noble metal (galvanic corrosion). In other cases nothing happens, although the less noble metal remains passive in the considered environment, for instance aluminium joinery fixed by stainless-steel screws does not cause corrosion after disappearance of the protective polymer washer. We must also consider that bi-metallic corrosion is not limited to only the galvanic couple: an active metal in a given environment may lead to corrosion of a metal which should normally be passive in this environment, simply because the products of corrosion of the former acidify the solution!!!

Diapositive 20 INPG. Oct. 2003 Corrosion in brief

Galvanic Corrosion of Aluminium alloys •

Rest potentials measured vs SCE in sea water (mV)

• • • • • • • • • • •

Graphite SS 316 titanium Copper Brass Fe (Steel) Al 2024-T4 Al1050 Al 7072 Cd Zn

+100 +10 -100 -150 -200 -600 -610 -750 -880 -800 -1100

3e-

Al 3+ 3/2 H2

Al Cl-

O 2 +H

Fe

O2

1/2 2OH-

3H+

20

Lorsque deux métaux différents comme l ’Aluminium et l ’Acier sont mis en contact électrique dans un même milieu conducteur (ionique), ils forment une pile qui débite du courant en consommant le métal le plus anodique (voir schéma). Comme l ’Aluminium est anodique par rapport à la plus part des métaux usuels, il est souvent la victime de ces assemblages. La corrosion galvanique nécessite trois conditions : -deux métaux différents -un contact électrique(électronique) -un électrolyte conducteur (contact électrolytique) La suppression de l ’une de ces trois conditions supprime le phénomène. Bien que certains éléments comme le cuivre déplace le potentiel de l ’Aluminium vers les valeurs plus nobles, le choix de l ’alliage d ’Aluminium ne permet pas d ’éviter ce problème. Si l ’on s ’en tient a la série galvanique ci-dessus, on pourrait croire qu’il suffit de remplacer un assemblage aluminium (1050)/acier inoxydable par un assemblage aluminium (2024)/acier ordinaire, pour régler le problème. Malheureusement l ’expérience montre que si on diminue effectivement le courant de corrosion galvanique avec ce nouvel assemblage, on n ’augmente pas pour autant la pérennité de l ’assemblage. La solution utilisée remplace deux éléments résistants à la corrosion par deux éléments corrodables, ce qui est gagné sur la corrosion galvanique est perdu par l ’auto-corrosion des éléments du couple. L ’expérience montre que dans l ’air, l ’assemblage de tôles Aluminium par des vis en acier inoxydable est préférable (cf.. les assemblages de mâts en Aluminium avec vis inox. sur les voiliers). Par contre s ’il y a un électrolyte; cas des parties immergées par exemple, il faut trouver d ’autres solutions.

Diapositive 21 INPG. Oct. 2003 Corrosion in brief

The consequences of galvanic coupling • The Galvanic corrosion – – – –

General rules: the electrochemical series Examples The effect of passivity Breakdown of noble coatings. Local corrosion

• The Galvanic protection – Sacrificial anodes. Examples – Sizes and distances – Galvanic coatings

21

Remarks: 1) In most cases the most noble metal is protected but accelerates the corrosion of the less noble metal (galvanic corrosion). In other cases nothing happens, since the less noble metal remains passive in the considered environment, for instance aluminium joinery fixed by stainless-steel screws does not cause corrosion after disappearance of the protective polymer washer. 2) We must also consider that bi-metallic corrosion is not limited to only the galvanic couple: an active metal in a given environment may lead to corrosion of a metal which should normally be passive in this environment, simply because the products of corrosion of the former acidify the solution!!!

Diapositive 22 INPG. Oct. 2003 Corrosion in brief

Active and passive Iron Active (in HCl)

Passive (in H2SO4)

In some circumstances, the dissolution current drastically decreases in a 22 potential range (passivity) whereas it remains large elsewhere (activity)

In very aggressive media (hydrochloric acid in the example), the anodic current (iron dissolution) increases continuously when the electrode potential increases In less aggressive media (sulphuric acid for instance) the polarisation curve of iron exhibit several different domains. First (active behaviour) the anodic current increases up to a maximum (The critical passivation current). Second, it decreases down to a very low intensity plateau (passive region) before increasing again for high potential (transpassive region). The onset of passivity in a potential range is due to the presence at the metal surface of a very thin but protective oxide layer (the passive film). This film is unstable at too high potentials (transpassive domain).

Diapositive 23 INPG. Oct. 2003 Corrosion in brief

Iron Chromium alloys

23

The figure above shows the typical behaviour of Fe-Cr alloys at different Cr contents in sulphuric acid (1M). The fall in passive current with the Chromium content increase is particularly noticeable up to 12% Cr which is generally considered as the practical limit for a steel to be stainless.

Diapositive 24 INPG. Oct. 2003 Corrosion in brief

The Passive film of stainless steels Thickness some 10 A° = some interatomic distances Typicallyt L=30 A° (for Fe,Cr, Ni, S.S...) Composition (oxi-hydroxide) oxidisables (Fe, Cr, etc...) -Oxygen (O ) Water and its derived forms ( H20 , H30+ , OH- ) Electrical properties F ~ 106 V/cm =10mV/Å

Vf ~ 300 mV pour L=30 Å Ip ~1µA à 1nA/cm2

~10µm/an (si métal divalent)

24

The composition and the thickness (around a few nanometres) of a passive film can be measured by different surface analysis techniques. The film is rich in the most oxidizable elements of the alloy (in this case chromium). It is also hydrated and often heterogeneous (the parts furthest towards the outside being more hydrated than the inner part, closer to an oxide). In addition, the thickness and composition depends largely on the metallurgical history of the surface. They also evolve in time, generally ensuring an increasingly protective nature with regard to the different forms of corrosion.

Diapositive 25 INPG. Oct. 2003 Corrosion in brief

Passivable materials: summary Certain materials, among which the chromium alloys, protect themselves by the formation of their own layer of corrosion in the form of a very thin protective film called a passive film. These materials are described as "stainless", quite simply (but this paradox is only apparent) because their oxidation is fast enough and intense enough to inhibit any subsequent corrosion. For these materials, the question of corrosion resistance no longer arises in terms of rate of dissolution, but in terms of the stability of the protective oxide. 25

Diapositive 26 INPG. Oct. 2003 Corrosion in brief

Some Remarks of practical interest • Effect of the surface condition: Different surfaces conditions (2B, BA, mechanical polishing, chemical passivation,etc…) produce different film characteristics and corrosion resistance • Aging of passive films. Passive films change with time (on periods of the order of days, weeks, months). Films generally grow, enrich in oxidisable elements and dehydrate (going from poorly protective hydroxides to more protective oxides). Localized corrosion resistance is generally improved • Ion release troughout the film: it may significantly affect only the elements present in the film (Fe, Cr, in the case of stainless steel). The nickel of austenitic stainless steel can only be released in case of local film breakdown, leading to the corrosion of the base metal. This does not occur if the metal has been properly chosen (i.e. with a stable passive film). 26

Diapositive 27 INPG. Oct. 2003 Corrosion in brief

Destabilising elements and classification of corrosive environments Acidity + 2H+

H20

MO

CORROSION RISK

Quasi- neutral

Acidic

Chloride free

NO

UNIFORM CORROSION

Chloride containing

LOCALISED CORROSION

DANGER!!

MCl+ + ClChlorides

M++ + Cl27

The factors involved in the potential destabilisation of a passive film are the acidity (which consumes oxygen and hydroxil ions) and the chloride ions (or all other similar ions) which combine with the cations without forming a protective species. The chloridized acid environments are particularly dangerous. Neutral non-chloridized or quasi-neutral environments (by this we mean where the pH is greater than a critical pH called the depassivation pH) present no risk, the non-chloridized acid environments can lead to uniform corrosion of the surface by disappearance of the passive film if the critical current is higher than the capacity of the oxidising agent to consume the electrons produced. As we shall see later, the neutral chloridized environments (pH above the depassivation pH determined in the presence of chlorides) can lead to a local form of corrosion which only affects a small part of the surface, the remainder remaining passive. The relevant notion is the critical chloride content, but it is difficult to implement since many other parameters (such as the potential reached) may intervene and other quality criteria will generally be preferred.

Diapositive 28 INPG. Oct. 2003 Corrosion in brief

Passivity in acidic media: Typical polarisation curve i

Activity

icrit

V

Vcrit Activity Pre-passivity

Vtp Passivity Transpassivity

28

On this typical polarisation curve, we can clearly distinguish 3 potential ranges: 1) The peak of anodic dissolution (also called activity range) with an anodic current which may reach several mA/cm²) 2) The passive range where the current is les than 1 µA/cm² and often of the order of 1 nA/cm² ) 3) The transpassive range where the dissolution current increases again. The passive domain corresponds to a quasi absence of dissolution, i.e. in practice nonoxidisability of the material (1 µA/cm² is approximately 10 µm/year). This slowing of the anodic dissolution is due to the presence of a thin film of oxide (a few nanometres) called passivating film or passive layer or PASSIVE FILM which considerably slows down the kinetics of the ionic transfer (by a factor often greater than 1000).

Diapositive 29 INPG. Oct. 2003 Corrosion in brief

Passivation by an oxidising agent ( e.g.: Fe+++→Fe++)

29

The electrochemical behaviour may differ in the presence of an oxidising agent other than the proton. In the example shown in the figure, the oxidising power is high enough for the intersection of the anodic and cathodic curves to be found in the passive range of potential (Point P). This situation occurs when the critical passivation current is below the current which can be supplied by the cathodic reaction. If not (for instance when oxidation is due to the single protons as described earlier), the system operating point is situated in the active domain and there is no protection by the passive film.

Diapositive 30 INPG. Oct. 2003 Corrosion in brief

The critical passivation current

i

Anodic Icrit

Cathodic A

I P

V 30

The left-hand figure shows how behave different alloys with increasing passivation currents, when immersed in an electrolyte containing an oxidising element Following the case, the cathodic and the anodic curves intersect in 1 or 3 points: For high critical currents, the 2 curves intersect in the active range (point A) which corresponds to a stable stationary state For low critical currents, they intersect in the passive range (point P) which also corresponds to a stable stationary state For intermediate critical currents, they intersect in 3 points: the active state (point A) which corresponds to a stable stationary state, the passive state (P) which corresponds to a metastable stationary state , and at point I, which represents an unstable state The critical passivation current represents therefore the capability of the steel to be to maintained steadily in the passive state . Finally, better than the corrosion current measured at free potential on the polarisation curve, the critical current constitutes the best quality criterion for the passivation aptitude of a stainless steel. The right hand figure shows the same thing that the left one, but for a single steel and different oxidising powers .

Diapositive 31 INPG. Oct. 2003 Corrosion in brief

Oxidising acids

• Nitric acid and other oxidising acids (such as phosphoric ..etc..) encourages anodic reaction, but overall the cathodic one • Then, adding HNO3 in sufficient amount in an electrolyte tends to passivate and not to corrode the stainless steels • This is why these acids are generally used for passivation treatment of stainless steels

31

Diapositive 32 INPG. Oct. 2003 Corrosion in brief

The effect of pH (depassivation pH ) (430 Ti in deaerated Na2SO4 1M)

-2.5

pHd

-3.5 -4.5 -5.5 -6.5 0

2

6

4

8

10

pH 32

The effect of the pH can be measured by plotting the polarisation curves and by measuring the critical passivation currents in relation to pH. Above a certain pH, called the depassivation pH, the activity peak is no longer observed and the metal is therefore passive whatever the potential (provided it stays below the pitting potential as we shall see later, and of course provided we avoid the high potentials corresponding to the transpassive domain, or the decomposition of water, or the change in the degree of oxidation of certain cations).

Diapositive 33 INPG. Oct. 2003 Corrosion in brief

Pitting in neutral chloride media (430 Ti in deaerated NaCl pH6.6)

33

In a chloride containig neutral environment, increasing the potential can lead to a sudden increase in the current long before the transpassivation potential range is reached. Micrographic examination after testing shows that local failure of passivity has arisen and that pitting has developed. The above polarisation curves have been obtained with a 430Ti steel in a NaCl aqueous solution with 6.6 pH. The pitting potential measured as described above diminishes linearly with the logarithm of the chromium content. It is also subject to wide scatter and this leads to the use of statistical methods. This scatter seems inherent in the pitting phenomenon, which appears to be probabilistic, at least insofar as initiation phase is concerned.

Diapositive 34 INPG. Oct. 2003 Corrosion in brief

Pitting in oxidising media I

K1

The pitting potential is a quality criterion for the corrosion resistance

K2

A V Vrest

Vpit Vredox1

Vredox2 34

Si le milieu contient un oxydant permettant une réaction cathodique, le potentiel d ’abandon devrait s ’établir à lintersection de la courbe anodique déterminée à l ’aide du potentiostat et de la courbe cathodique (supposée connue) correspondant à la réduction de cette oxydant. Suivant le pouvoir oxydant de cette réaction (courbes cathodiques K1 ou K2), le potentiel d ’abandon s ’établira en dessus ou en dessous du potentiel de piqûration et il y aura ou non corrosion localisée. Le potentiel de piqûres est donc bien un critère de qualité pour la résistance à la corrosion par piqûres et les alliages ayant un potentiel de piqûre élevé sont ceux qui résistent le mieux à la piqûration.

Diapositive 35 INPG. Oct. 2003 Corrosion in brief

Acidic chloride containing media

• The fall in pH and the increase in the Cl- content reduces the passive range 35

Chloride acid environments cumulate trends to passivation failure. Diminishing the pH increases the critical current and increasing the content of chlorides brings down the pitting potential. This assembly considerably reduces the passivity range Such environments must really be considered constituting the most hazardous.

Diapositive 36 INPG. Oct. 2003 Corrosion in brief

Localized corrosion in neutral chloride containing solutions solution surface (passive) metal

occluded Corrosion zone

CORROSION RISK

Quasi- neutral

Acidic

Chloride free

NO

UNIFORM CORROSION

Chloride containing

LOCALISED CORROSION

DANGER!!

A small local active region is surrounded by a large passive zone. This active region is partially occluded, then diffusion is slowered and corrosion products accumulate. This strongly modify the local chemistry and agressiveness. Local pH decreases continuously and Cl- concentration increases, which self maintains the active behaviour 36

Diapositive 37 INPG. Oct. 2003 Corrosion in brief

The successive stages of localised corrosion

• Initiation (different specific modes) • Propagation (common mode) – Chloride enrichment – Hydrolysis of dissolved cations – High propagation rate

37

Diapositive 38 INPG. Oct. 2003 Corrosion in brief

Chloride enrichment • Les cations produits par la dissolution anodique sont peu mobiles (car hydraté)s et restent confinés à l’intèrieur de la zone occluse, ce qui crèe un champ électrique orienté vers l’extèrieur • Entrainés par le champ électrique les anions majoritaires de la solution les plus mobiles migrent vers l’intérieur de la zone • dans le cas d ’une solution chlorurée cela provoque un enrichissement local en Ions Cl- et dans certains cas la précipitation d ’un film salin (chlorure du métal de base)

38

Diapositive 39 INPG. Oct. 2003 Corrosion in brief

Hydrolysis of dissolved cations produces Local Acidification M++ + H20 → MOH+ + H+ This acidification is limited by the migration and the diffusion of reaction products outside the zone. The fall in pH and the increase in the Cl- content reduces the passive range

39

Diapositive 40 INPG. Oct. 2003 Corrosion in brief

Large cathodic over anodic areas ratio produces High propagation rate cathodic area SK

cathodic area SK

Anodic (A) and cathodic (K) Current intensity (I) and density (J=I/S) IA +IK =0

Dissolution Mn+

IA =- IK = Icor

JA /JK =SK/SA >>1

Anodic area SA