Materials for Energy[PHY563]

Lecture VI: Durability of Energy Materials

(chemical and mechanical stability )

12/02/2018

Polina Volovitch (Chimie ParisTech, PSL University)

polina.Volovitch@chimieparistech.psl.eu

“For a product (article) “permanence” may be regarded as the most important aspect, whether this permanents relates the shape (dimensional stability), mechanical properties (tensile and impact strength, fatigue) or environment (resistance to ageing). Not enough is as yet known about the fundamental background of these permanence properties. “

D.W. Van Krevelen in “Properties of polymers, their correlation with chemical structure; their numerical estimation and prediction from additive group contributions”, Fourth, completely revised edition, Elsevier, 2009 1

2

Why? “Real” materails in “real” environment…

Chemical degradation(corrosion):

Irradiation (UV, …)

T°C

High Tension

Flow

Scratch

Wind…Shadow

Wetness

Oxidants (02, 03, H2O2, SO42-, …)

Dust (particles)

Pollutants (NaCl, (NH4)2SO4, …)

Physical degradation

Mechanical degradation

(under stress)

Why materials degrade? Because of thermodynamics… Most materials are not in thermodynamic equilibrium with their surroundings => Gibbs free energy (G) is not at a minimum

work entropy

G = U + PV - TSfree energy internal energy

reactions will occur if G <0

What can reduce G ? Chemical reactions – reduction in U (and S)

Physical reactions – increase in TS

Fracture – reduction in PV ( reduced strain energy)

….and their combinations

3

http://www.aimonitoring.com/

operations Pits in 16 inch Sheet Iron Water Main in San Francisco's East Bay that failed in one month due to leakage of current from electric streetcars. (Photo dated August 10, 1926, Courtesy of East Bay Municipal Utility District).corrosion-doctors.org/Why-Study/Historical-perspective.htm

corrodere (lat.) = gnaw through (ronger)

Chemical degradation (corrosion)

4

NACE International: The deterioration of a material, usually a metal, that results from a reaction with its environment

ISO 8044-1986: Physicochemical interaction between a metal and its environment which results in changes in the properties of the metal

and which may often lead to impairment of the function of the metal, the environment, or the technical system of which these form a part.

IUPAC: Corrosion is an irreversible interfacial reaction of a material (metal, ceramic, polymer) with its environment which results in

consumption of the material or in dissolution into the material of a component of the environment

Pliny ” It is a great evil that to enable death to reach human beings more quickly

we have taught iron how to fly”

…rusting of iron as a punishment of the gods?... - corrosion control by religious

ceremony (ex. used to protect the chains of a suspension bridge built for Alexander the

Great)

5

The approach is to try and eliminate the external marine environment as much as is feasible, rather than producing all components from corrosion resistant materials. This allows commonality between on-shore and off-shore systems (mechanical and electrical) but requires well designed nacelles and efficient internal environmental control

Composite (blade)

Multimaterials (nacelle)

Steel (tower)

The submerged area

The Splash Zone

Atmospheric Area It’s WET!

Example: wind energy

off-shore plant

6

Turbine nacelle - to inhibit the ingress of water. But still NaCl, wind, moisture…Additional protection in particular for Power Generation Units, Electrical Distribution Panel, Instrument and Electrical Isolation and Junction Boxes , Hydraulic Pump Drive Motors

Vapour Corrosion Inhibitor (VCI) supplied in self-adhesive strips - to protect the inside of electrical and instrument enclosures, coil windings, contacts, and transformers .

Paul Hogg “Durability of wind turbine materials in off-shore environments ”

Corrosion Protection of Wind Turbine Electrical System

7

Biofouling on an offshore jacket foundation (a) and access to a wind turbine foundation for maintenance (b)

https://link.springer.com/chapter/10.1007/978-3-319-51159-7_5

Types of fouling: seaweed, mussels, hydroids, clasters of mussels…

Types of corrosion: sever in splash zone, tidal zon, accelerated low water (under mussels),localized (under cluster of mussels, local (under soil)…

It’s WET!

Example: wind energy

off-shore plant

fungicides, coatings, …

catodic protection, coatings, …

Aqueous corrosion is an electrochemical process

with origins in the characteristics of electrical

conduction of the two phases: electronic conduction

in the metal phase and ionic conduction in

the aqueous phase, also called the electrolyte.

Aqueous corrosion - (CEFRACOR) et le CFPC

8

Fe2++2e-=Fe E°=-0.440V

Fe3++3e-=Fe E°=-0.037V

Example of application of thermodynamics for Fe corrosion (Fe in deaerated solution)

Which species will be major if Fe electrode is in solution – Fe(II) or Fe(III)?

What will be the protection potential? (a conventional potential at which the quantity of corroded metal results in

the equilibrium solution concentration 10-6 M, at this and lower potentials the corrosion is considered very slow)

Asuming cathodic water (hydrogen) reduction: effect of pH on the corrosion resistance of Fe?9

Let us tell “Fe corrosion=Fe anodic dissolution”and try to see what can we know from the potential measurements

Fe2++2e-=Fe E°=-0.440V

Fe3++3e-=Fe E°=-0.037V

Example of application of thermodynamics for Fe corrosion (Fe in deaerated solution)

Which species will be major if Fe electrode is in solution – Fe(II) or Fe(III)? considering that both reactions are at equilibrium and taking into account that Fe electrode is the sameErev (Fe2+/Fe)=Erev(Fe3+/Fe)

using Nernst law one can show C(Fe2+)0.5/C(Fe3+)0.33107 hence Fe3+ will be very minor in deaerated solution in the presence of Fe°(will lead to dismutation reaction: 2Fe3++Fe+e-=3Fe2+)

What will be the protection potential? (a conventional potential at which the quantity of corroded metal results in

the equilibrium solution concentration 10-6 M, at this and lower potentials the corrosion is considered very slow) :

Asuming cathodic water (hydrogen) reduction: effect of pH on the corrosion resistance of Fe?10

Let us tell “Fe corrosion=Fe anodic dissolution”and try to see what can we know from the potential measurements

Fe2++2e-=Fe E°=-0.440V

Fe3++3e-=Fe E°=-0.037V

Example of application of thermodynamics for Fe corrosion (Fe in deaerated solution)

Which species will be major if Fe electrode is in solution – Fe(II) or Fe(III)? considering that both reactions are at equilibrium and taking into account that Fe electrode is the sameErev (Fe2+/Fe)=Erev(Fe3+/Fe)

using Nernst law one can show C(Fe2+)0.5/C(Fe3+)0.33107 hence Fe3+ will be very minor in deaerated solution in the presence of Fe°(will lead to dismutation reaction: 2Fe3++Fe+e-=3Fe2+)

What will be the protection potential? (a conventional potential at which the quantity of corroded metal results in

the equilibrium solution concentration 10-6 M, at this and lower potentials the corrosion is considered very slow) :E=-0.44+(0.059/2)*log(10-6)≈-0.62(V vs SHE) (verify Pourbaix diagram next slide)

Asuming cathodic water (hydrogen) reduction: effect of pH on the corrosion resistance of Fe?See Pourbaix diagram next slide, Fe is stable in alkaline pH (insoluble products), unstable at neutral and acid pH 11

Let us tell “Fe corrosion=Fe anodic dissolution”and try to see what can we know from the potential measurements

Fe2++2e-=Fe E°=-0.440V

Fe3++3e-=Fe E°=-0.037V

Each line at the Pourbaix diagram corresponds to a specific reaction with equilibrium potential described by Nernst law.

Comparing the “protection potential” of Fe at different pH (it will be independent of pH when no insoluble oxide is formed because H+ is not involved in the reaction) with the potential of water/hydrogen reduction, one can conclude that 1) in the presence of water Fe will be oxidized (anodic reaction);2) the cathodic reaction can be hydrogen/water or oxygen reduction. Hence, metallic iron will be unstable in the presence of water and will corrode, but we can’t tell anything about the corrosion rate.In alkaline solutions the formation of insoluble products can protect iron from rapid corrosion (passivation).

Fe3++e-=Fe2+ E°=+0.771V

12

Example of application of thermodynamics for Fe corrosion (2)

Fe2++2e-=Fe E°=-0.440V

2H++2e-=H2 E°= 0V

Conclusion from TD: iron in water should be described as a mixed electrode(cathodic and anodic reactions involve different species)

Espontaneous=Ecorr (Kinetic) ≠ Erev (TD)

EFe(2+)/Fe Ecorr EH(+)/H2

Polarization () =E-Ecorr

polarization0 I0 cathodicpolarization0 I0 anodic

13

Espontaneous=Erev

Surtension ( ) = E-Erev

For comparison: case of simple electrode (Fe in non-aqueous solution of Fe2+)- reversible potential is determined by thermodynamic constants.

Fe Fe2+

Fe Fe2+

Fe Fe2+

H+

H2

For comparison: case of mixed electrode (Fe in aqueous solutions) – the concentrations of species in anodic and cathodicpartial reactions are not related by TD constants and depends on the rates of partial reactions;the spontaneous potential depends hence on the rates of anodic and cathodic reactions (determined by kinetics and not TD)

Learning points for the application of TD in corrosion research

1. TD describes the possible species, nothing can be told about kinetics!

2. Use Nernst law to predict the electrode potential and construct simple Pourbaix diagrams

3. TD helps predict major species involved in anodic and cathodic reactions and the effect of solution composition on potentials

4. Standard potential is not the reversible potential (it determines different relative activity of metals in galvanic series with different electrolytes)

5. Understand the relativity of “protection potential”

6. Note the difference between simple and mixed electrode and the related potentials

7. Define polarization

14

Kinetic aspects: Polarization curves measurement I=f(E) and E=f(I)

Rappel: Faraday law: charge from the oxydation reaction (Q) is propositonal to the - quantity of corroded metal (moles), Faraday constant F and number of exchanged electrons (n)

Q = n F The corrosion rate [V= d /dt ] can be hence related to the corrosion current (I) I = n F [d /dt ]= nF V

15

Why ?

Possible rate limiting steps (RDS)

Fe Fe2+

O2

OH-

e-Charge transfer through the double layer Mass transfer (by diffusion, etc.)(potential dependent) (reactives or/and products)

Fe2+

O2

OH-

Fe Fe3O4 Fe2+

(simplified)

OH-

O2e-

Specific case of passive films: both charge and mass transfer are determined by the properties of a passive film

which depends on potential, chemistry, etc.

Attention – direct application of Faraday law for corrosion depth calculation is possible only for uniform corrosion!

What can determine the shape of the I(E) curves?

There are 3 main types of kinetics studied in corrosion: determined by charge transfer, by mass transfer or by the properties of passive films

i =k C exp (-G/RT)

C doesn’t depend on E => E effect = variation of G for charge transfer.

Ex. anodic reaction: E=E-Ereversible>0

=> the driving force for e- to reach the electrode G=G°-nFE/RT

i increases by a factor exp(nFdE/RT) =>

16

Shape of I=f(E) curves for reactions limited by charge transfer through the double layer

Note: deviation from the linear behavior at |E|>>|Eeq| could be due to the electrolyte resistance (Ohmic drop, R) or due to the mechanism change

Practical note: Ohmic drop increases with current and with distance from the reference electrode (RE) – minimize!

nFE/RT

(Butler Volmer)

fraction of E effective for the G decrease

Reaction rate

(current)

concentrationactivation energy

slope=-cnF/RT

log

(ic)

Ex. cathodic E=E-Ereversible<0

ic=i°exp(cnFE/RT)

ia=i°exp(anFE/RT)

current w/o polarizationcurrent

under polarization

Example: O2+4H++4e-=2H2O in acid solution on an inert electrode

the maximum currents can be described as the currents when the concentration of the oxygen at the interface is

zero - all is consumed as soon as it arrives.

Assuming the linear concentration decrease through the diffusion layer of thickness , the limit current can be

written as

iL=-4FD(O2)[C(O2)]/

(and in general the current can be i=-4FD(O2)[C(O2)-C(O2)surface]/ if If the maximum rate is not yet achieved and

some oxygen is still present at the surface (concentration C(O2)surface )).

2. Uner polarization E (at constant pH) can be correlated with the concentration gradient from the surface to the

bulk if the charge transfer is rapid and “steady state” is achieved (application of Nernst for steady state)

E=E-Erev=RT/4F ln[C(O2)surface/C(O2)]

Taking into account that [C(O2)surface/C(O2)]=1-i/iL one can obtain i=iL(1-exp(nFE/RT)

17

Shape of I=f(E) curves for diffusion limited reaction rates

Typical shape:i=iL(1-exp(nFE/RT)

Example of oxygen reduction O2+4H++4e-=2H2O

Figure adopted from M. Tachibana et al. J. Nucl. Sci. Tech. 49 (2012) 551 and D. Landolt book Fig. 4.23.

18

Shape of I=f(E) curves under mixed control: charge transfer activation and diffusion

i0 / iL =10-2

i0 / iL =10-5

a) Activation controlled region (charge transfer control)b) Diffusion controlled region

19

Mixed potential theory: idea1. The total current measured in the system is a sum of cathodic and anodic currents (grey curve).

2. If all anodic and cathodic reactions act at the same electrode, anodic and cathodic areas are not separated

physically, there is no mass transfer limitations and no potential drop related to the electrolyte, in the

absence of external current Ia, total + Ic, total=0. The potential at which anodic and cathodic current “annulate”

each other is a mixed potential (corrosion potential), anodic current value at this potential is an exchange

current which can be associated with corrosion rate..

Example of polarization behavior of iron

1 M HCl(charge transfer control, Tafel behavior)

0.5 M H2SO4

(passive film control)

23

immunity

activedomain

passive

transpassive

Eb- pitting potential

ip- passivation current

Properties of the passive film depend on the potential and solution composition – mechanism can be modified under polariztion

Learning points from kinetic aspects

1. Relation between corrosion rate and current measured in polarization curve (Faraday law)

2. Recognize some typical shapes of i(E) curves: charge transfer control (Butler Volmer, BV) and mass transfer control (diffusion limited)

3. Use the idea of mixed potential and exchange current to define corrosion potential and exchange current from experimental polarization curve

4. Know Tafel approximation and polarization resistance for corrosion rate measurement: methodology and applicability conditions

24

The approach is to try and eliminate the external marine environment as much as is feasible, rather than producing all components from corrosion resistant materials. This allows commonality between on-shore and off-shore systems (mechanical and electrical) but requires well designed nacelles and efficient internal environmental control

Composite (blade)

Multimaterials (nacelle)

Steel (tower)

The submerged area

The Splash Zone

Atmospheric Area It’s WET!

Example: wind energy

off-shore plant

25

26

Atmospheric corrosion

corrosion that occurs in materials exposed to the ambient air

W.H.J. Vernon – 1920s… first test with CO2 and SO2

1. Surface hydroxylation

2. Water ad /absorption

3. Gas deposition and solution chemistry modification

4. Surface chemistry modification

(H+ and ligand induced metal dissolution)

5. Nucleation of corrosion products

6. Coalescence of corrosion products

7. Ageing and thickening of corrosion products

27

1 ps 1µs 1s 1 min 1h 1 day 1 month 1 year 100 years

Charge

transfer

Surface

film

formation

Nucleation

of corrosion products

Mass transfer

Steady state rate

Relative timescale

28

Pollutants in evaporating surface layer

I = 1/2 mi[Zi]2

10-3 10

Rain River Fog Ocean Aerosol Particle

Species:

CO2, O3, NO2, H2S→SO2, Cl-, Ca2+, Mg2+, Na+, NH4+, H+

,

CH3COO-, CHCOO-

29

Time of Wetness

(RH >80%)

P(SO2)

P(Cl-)

Parameters for classification

RH

ML

of

wate

r T1, T2

Industrial atmosphere

Urban atmosphere

Rural atmosphere

Corroded mass (mg m-2)

Exposition time (year)

30

Exercise 1: DiscussT°C effects on Corrosion rate (CR)?

Reasonable T°C interval for testing?

T°C

CR

?

31

Atmospheric corrosion sites (ISO9323)

c

C2: Atmosphere with low degree of pollution. Rural areas in particular.

Unheated buildings where condensation may occur.

Examples: storage areas, sports centers

C3: Urban and industrial atmospheres, moderate SO2 pollution.

Coastal areas with low salt content.

Production area with high humidity and some pollution.

Examples: food industry, laundries, breweries, dairy plants etc

C4: Industrial and coastal areas with a moderate salt content

Examples: chemical factories, swimming baths, shipyards on the coast etc.

C5: C4 with high salt or pollution

[Corrosion Rate] in g m-2 year-1

32

How to proceed?

J. Braz. Chem. Soc., Vol. 24, No. 3, 449-458, 2013

Example of climat parameters

(Valparaizo, Chilie)

33

Corrosion tests

+ mass balance + surface characterization 33

The approach is to try and eliminate the external marine environment as much as is feasible, rather than producing all components from corrosion resistant materials. This allows commonality between on-shore and off-shore systems (mechanical and electrical) but requires well designed nacelles and efficient internal environmental control

Composite (blade)

Multimaterials (nacelle)

Steel (tower)

The submerged area

The Splash Zone

Atmospheric Area How to protect?

Example: wind energy

off-shore plant

34

Anticorrosion protection

Passive (barrier effect)

limits access of aggressive species (oxygen, Cl-, etc…), water and/or charge transfer

Inhibitors(inorganic or organic …)(natural passive films on Al, stanless steel,

anodized films on Al, thin films, organic paints,

temporary protection by oil, etc…)

Active

external anode(cathodic protection)

Galvanic coupling

sacrificial Me coating(Zn galvanizing, etc.)

Mechanisms can be very complex, different for different inhibitors (in-situ film formation, local pH change, modification of cathodicreaction, potential shift, etc…)

but inhibitor does not form a steady state film in advance, but it acts “on request” and both kinetic issues and stability are important

“+” - unlimited (if no damage!): no material renewal necessary

“-” - if damaged: strong localized corrosion

“-” - time limited (renewal of the active agent can be necessary)“+” no enhancement of localized corrosion

Usually: combination of mechanisms!!!

35

36

Galvanic protection from Evans diagram: current definition*

Igalvanic =abs(Ic+Ia)=jc Aa+jaAa=- jc Ac-jaAc

whereA – surface area, I –total current and j – current density;

Indices c and a indicate cathode and anode.

Eg Fe (c)Zn (a)

log

I /

A

E / V

2H22H+

(on Fe)

ZnZn2+

2H22H+

(on Zn)FeFe2+

logIg

Igalvanic =(Eg-Ea)/Rpa

Assuming small potential variation at Me/electrolyte interface, one can expect

linear variation of current via polarization resistance R=E/Iand hence

* electrolyte with no resistance (Rsolution=0)

Old example : galvanizing

Sacrificial protection by galvanic coupling + barrier protection by formed corrosion products of Zn ( if stable in the atmosphere)

37

-1.6 -1.4 -1.2 -1 -0.8 -0.6 -0.4

104

103

102

10

1

10-1

Potential / V vs. SCE

steel

Curr

en

t d

en

sity |j| /

µA

. cm

-2

O2 reduction rate

on uncovered steel

on steel covered with Zn corrosion products

38

Effect of environment on corrosion potential

-3

-2

-1

0

1

2

3

4

-1,5-1,4-1,3-1,2-1,1-1-0,9-0,8-0,7

E / V vs. SCE

Log(j) / µA.cm

-²

Corrosion inhibitors

Ecor

Ex. cathodic inhibitor: Ex. anodic inhibitor:

E / V

Ecor

E / V

log

I /

A

Action by1. Increasing the anodic or cathodic polarization behavior (Tafel slopes);2. Reducing the movement or diffusion of ions to the metallic surface;

3. Increasing the electrical resistance of the metallic surface…

Inhibitor Efficiency (%) = 100·(CRuninhibited- CRinhibited)/ CRuninhibitedCR – corrosion rate

Typically good inhibitor would give at least 95% inhibition at a concentration of 0.008% and 90% at a concentration of 0.004%

log

I /

A

iaic

w/o inhibitor

with inhibitor

39

Scholar approach to organic (mixed) inhibitors: adsorption and blocking access of aggressive species to the surface

Standard approach (not every time correct!) is to

relate inhibition efficiency with surface coverage ()and calculate adsorption isotherms.

Modelling of adsorption in such an approach permitsto screen potential inhibitor molecules. DFT calculations of surface – inhibitor interactions become more and more popular.

http://www.beilstein-journals.org/bjnano/single/articleFullText.htm?publicId=2190-4286-5-258

Surface coverage

Concentration 40

Physical degradation (irradiation)

41

Physical degradation (irradiation)

Examples from presentation of L. Malerba

VOIDS

42

Mecanical Degradation

43

FF

Tensile

F

Bending

FF

Compressive

L=L-L0

F

Shear

F

A-section

EYoung modulus

MODULUS = “UNIT STRESS” / ”SHARE”

Compliance = 1 / MODULUS

Typical stress-strain behavior

44

Exercise: Examples of compounds which behavior can be described by different graphs (before and after Tg, different ways of synthesis affecting M, etc…)

Physical degradation leads to mechanical changesExamples from presentation of L. Malerba

Radiation embrittlement is usually consequence of hardening(migration of defects result in pinned dislocations) 45

Examples from presentation of L. Malerba

Not only hardening but also loss of hardening in defect-free zones

Physical degradation leads to mechanical changes

46

Chemical degradation leads to Mecanical changes(Stress Corrosion Cracking, SCC)

Mechanical Stress + Corrosive

EnvironmentSCC Failure

47

48

SOLID + LIQUID + STRESS DURABILITY (time)?

MECHANICAL CHEMICAL

Fracture Rehbinder effect Dissolution Corrosion

adsorption induced decrease in strength of solid

cr1/2

( embitterment, plasticization,…)

OBJECTIFS:

« reversibility » vs drying!

PHENOMENA:

Rehbinder effect

P. A. Re(h)binder

The Rehbinder Effect, E. N. DA C. ANDRADE & R. F. Y. RANDALL , Nature 164 (1949) 1127New applications of the Rehbinder effect in tribology. A review, V.I. Savenko, E.D. Shchukin, Wear 194 (1996) 86

Surface effects in adhesion, friction, wear, and lubrication, Donald H. Buckley - 1981 - Science

surface energy

49

Liquid Metal Embrittlement (0)

Aluminium in presence of Ga

Aluminium in presence of Ga

50

0

500

1000

1500

2000

2500

0 1 2 3 4 5 6

AirLead

ST

RE

SS

(M

Pa

)

STRAIN (%)

T91 revenu à 500°C5µm

T91 - Fe9Cr1Mo0.25V0.07Nb

ductilefragile

Liquid Metal Embrittlement (1/3)

Aluminium in presence of Ga

(illustration of Vogt, Université de Lille)

51

Moomba Australia - January 1, 2004 (Liquid Metal Embrittlement,

Gas Processing Plant, USD 5,000,000)

The gas was released that led to vapor cloud

explosion. The gas released was caused by the

failure of a heat exchanger inlet nozzle in the

liquids recovery plant. The failure of the inlet

nozzle was due to liquid metal embrittlement of

the train B aluminium heat exchanger by

elemental mercury.

Skikda Algeria - January 19, 2004

(Liquid Metal Embrittlement, LNG Plant, 27

killed 72 injured, USD 30,000,000)

A report noted that the explosion was the

consequence of a catastrophic failure in one of

the cold boxes of Unit 40, which led to a vapour

cloud explosion of either LNG or refrigerant.

The most probable source of ignition was the

boiler at the north end of Unit 40. The report

concluded that the escaped gas was from the

cryogenic heat exchanger.

“The 50 Major Engineering Failures (1977-2007)”

Liquid Metal Embrittlement (3/3)

52

Min at 0.2% of oleic acid for all

metals (tin, lead and copper).

Why?

This characteristic concentration corresponds to the equilibrium saturated adsorption layer, and does not depend on the nature of

the metal, but depends only on the nature of surfactant molecules. By analogy with the Ducleaux-Traube rule, this

concentration decreases with the increase in the length of hydrocarbon chain for a homologous series.

- large decrease in the surface energy,

- hard stressed state

- kinetic opportunity for penetration

“Softening” (plasticization) vs embrittlement (1/4)

small

0 %

0.2 %

0.1 %

1 %

0.5 %

Monocrystalline metals in Vaseline oil + OA

53

Extension to SOLID - GAS system: Strength decrease of catalysts in reactive media

tf -time to fracture (durability) in function of compressive stress PLg(tf /s)Ln(tf /min)

P / MPaP / MPa

In reaction

Inert atm

Co-Mo catalyst, 200°C

4123

MgO catalyst, 395°C

AcroleineN2

CO2

Acetaldehyde

54

Why? “Real” materails in “real” environment…

Chemical degradation(corrosion):

Irradiation (UV, …)

T°C

High Tension

Flow

Scratch

Wind…Shadow

Wetness

Oxidants (02, 03, H2O2, SO42-, …)

Dust (particles)

Pollutants (NaCl, (NH4)2SO4, …)

Physical degradation

Mechanical degradation

(under stress)

Interplay

55