methods of studying the corrosion process mechanismhome.agh.edu.pl/~grzesik/fhtc/3_methods of...

METHODS OF STUDYING THE

CORROSION PROCESS MECHANISM

http://home.agh.edu.pl/~grzesik

Methods of determining the contribution of individual

reagents in matter transport through the scale

• Marker method

• Two-stage oxidation method

• Determining self-diffusion coefficients

• Scratch method

• Pellet method

Schematic illustration of scale formation

according to Tammann (1920)

Metal Metal

OxidantOxidant

Scale12 2X Me MeX

Marker method – schematic illustration

of Pfeil’s classic experiment (1929)

Fe

Air

Scale

Fe

SiO2

SiO2

Air

Marker experiment in a Fe-O system

H. Engell, F. Wever, Acta Met. 5, 695-700 (1957)

Marker method – interpreting results

Oxidant

Metal

Marker

Scale

Oxidant

Metal

Marker Scale

Oxidant

Metal

Scale

Oxidant

Metal*

**

* reaction location: 12 2X Me MeX

*

Marker

Marker

Oxidant

Metal

Marker

Me1-yX

X2

Me

Marker

MeX1-y

X2

Me

Marker

Me1±yX1±z

X2

Me

MarkerMe1+yX MeX1+y

For temperatures above Tammann’s temperature

Marker method – interpreting results

in the case of predomiant lattice diffusion

Conditions for the correct

procedure of a marker experiment

• the surface of the examined metal or alloy is smooth

• the marker does not react with the metallic substrate, oxidant, or

with substances that are compounds constituting the scale

• before beginning the oxidation process, contact between

the marker and surface of examined material is maintained

• the formed scale is compact, single-phase and strictly adheres

to the metallic core

• oxidation time is selected so that the thickness of the scale

will be at least one order of magnitude larger than the marker

Methods of depositing markers

on metallic substrate surfaces

• even distribution of marker grains (Al2O3, SiO2, etc.)

• even distribution of a thin (10 mm) wire (Pt, Au)

with around 1 mm length

• welding a thin wire constituting a marker

• Depositing a marker (Pt, Au) through an appropriate mesh

(Cu – SEM; Al)

• electrolytic deposition of a marker layer

• decomposition of a noble metal salt, placed on the sample

surface

• photolytic method

• Covering the sample surface with diluted platinum paste

Examples of incorrectly using the marker methodusing a non-elastic wire around the metal

Me

Markers

Me

Me

MeX MeX

Markers

AA

A A

Me

MeX

Examples of incorrectly using the marker methodmarker does not adhere to the metal after deposition

Cross-section

Example of how a marker experiment can be

incorrectly interpreted due to the undercutting mechanism

Marker

Metal Metal Metal Metal

OxidantOxidantOxidantOxidant

Scale

Scale

Scale

Co

S9

8C

o

Pt

50 mm

Cross-section of a sulfide scale formed

on cobalt with an indicated marker location

Z. Grzesik, Ceramika, 87, 1-124 (2005).

T = 700 oC, p(S2) = 10-2 Pa

Co

Co

S4

3

Pt

50 mm

Z. Grzesik, Ceramika, 87, 1-124 (2005).

T = 860 oC, p(S2) = 10-1 Pa

Cross-section of a sulfide scale formed

on cobalt with an indicated marker location

Energia / MeV

0,0 0,5 1,0 1,5 2,00

2

4

6

8

10

12

Nb1,68MeV

Au1,84MeV

Lic

zb

a z

liczen / 1

03

Lic

zba z

licze

n / 1

03

0,0 0,5 1,0 1,5 2,00

5

10

15

20

25

Energia / MeV

Przed siarkowaniem

Po siarkowaniu

SNb/NbS2

SNbS /S2 2

NbNb/NbS2

NbNb/NbS

1,68 MeV2

Au1,48MeV

Z. Grzesik, K. Takahiro, S. Yamaguchi, K. Hashimoto and S. Mrowec, Corrosion Science, 37, 801-810 (1995)

RBS spectrum (4He - 2 MeV) of an Nb sample

marked with Au, before and after sulphidation

Before sulfidation

After sulfidation

Cross-section of an MoS2 scale formed on Mo

with an indicated marker location

correctly

performed

experiment

Z. Grzesik, M. Migdalska, S. Mrowec, High Temperature Materials and Processes, 26, 355-364 (2007)

incorrectly

performed

experiment

Cross-section of an MoS2 scale formed on Mo


Cross-section of an TaS2 scale formed on Ta


correctly

performed

experiment

metal

zewnętrznawarstwa zwarta

utleniacz

pośrednia warstwa porowatawewnętrzna warstwa porowata

X2

Me

e

+

-

MECHANIZM POWSTAWANIA TRÓJWARSTWOWEJ

ZGORZELINY W OBSZARZE POWIERZCHNI PŁASKICH

Marker

Cross-section of a triple-layer scale on a flat

surface with an indicated marker location

metal

External

compact layer

oxidant

Intermediate

porous layer

Internal porous layer

OBRAZ SZCZELIN DYSOCJACYJNYCH W ZGORZELINIE

SIARCZKOWEJ NA STOPIE Cu-9%Zn, UZYSKANEJ

W PROCESIE DWUETAPOWEGO SIARKOWANIA

Autoradiogram

S. Mrowec, An Introduction to the Theory of Metal Oxidation, National Bureau of Standards and National

Science Foundation, Washington D.C., 1982.

Picture of dissociation crevices in a sulfide scale

on Cu-9%Zn alloy, obtained in a two-stage

sulphidation process

Cross-section of a sulfide scale on copper,

obtained at 444 oC

reaction time: 25 s reaction time: 9 min

compact scale

layer

marker

porous scale

layer

Cu

S. Mrowec, An Introduction to the Theory of Metal Oxidation, National Bureau of Standards and National

Science Foundation, Washington D.C., 1982

Marker method in a system

with a ceramic substrate

a) „classic” results interpretation

b) point defect concentration in studied materials and the

result of the marker experiment

c) nuclei of the reaction product inside the substrate and

the result of the marker experiment

d) physical microdefects of the substrate and the result of

the marker experiment

Schematic illustration of using the marker method

in an oxide – oxidant system

X2

MeX

Markers

X2 diffusion Me diffusion Me and X2 diffusion

X2

MeX

MeX2

X2

MeX

MeX2

X2

MeX

MeX2

Marker experiment interpretation

in an oxide – oxidant system

X2

MeaXb

MecXdd

bMe X Me X

d a

bc Me n

d a

bc ea b c d

n

d a

bc Me

d a

bc n e

d d a b c

b cX

d a b c

b cMe Xn c d

2 2

Ratio of external thickness to the internal part of the MecXd oxide layer

d a b c

b c

Z. Grzesik, Termodynamika i kinetyka defektów w kryształach jonowych, WN Akapit, Kraków 2011

Methods of depositing markers on

the surfaces of ceramic substrates

• depositing a marker (Pt, Au) through an appropriate mesh

(Cu – SEM; Al)

• photolytic method

• covering the sample surface with a diluted platinum paste

Schematic illustration of the marker location

in Co3O4 resulting from CoO oxidation

Cation sublattice

of Co3O4 is defectedAnion sublattice

of Co3O4 is defected

4 3 4CoO Co O Co nen Co ne O Co On 23 2

13 3 4

CoO Co O3 4

Co+n

e-

25 %75 %

Markery Au

O2

Schematic illustration of marker location in Co3O4grown during CoO oxidation, in the case where the

cation sublattice of Co3O4 exhibits predominant disorder

Au markers

4 3 4CoO Co O Co nen

Co ne O Co On 23 213 3 4

Co3O4 scale cross-section grown CoO


Au markers

Cu O2

CuO

Au-markers

50 mm

Cu O CuO Cu e22 2

Cu e O CuO 2 12 22

M. Migdalska, Z. Grzesik, S. Mrowec, Defect and Diffusion Forum, 289-292, 429-436 (2009)

Cross-section of CuO scale grown on Cu2O


Au markers

Cu O2 CuO

Cu+2

e-

50 %50 %

Markery Au

O2

Cu O CuO Cu e22 2 Cu e O CuO 2 12 22

Schematic illustration of marker location in CuO grown

during oxidation of Cu2O, in the case where the cation

sublattice of CuO exhibits predominant disorder

Au markers

NiS2 scale cross-section grown during NiS sulphidation


Au markers

Schematic illustration of marker location in NiS2grown during NiS sulphidation

disordered NiS2anion sublattice

disordered NiS2cation sublattice

Ratio of external thickness to the internal part of NiS2 equals

around 0,82



Au markers

Correlation between deviation from stoichiometry

in Ni1-yS and sulfur vapor pressure

10-1 100 101 102 103 104 10510-3

10-2

10-1

1073 K1023 K

973 K

923 K

873 K

Badania własne

H. Rau, 1975

1168 K

1110 K

1038 K

986 K

914 K

835 K

Ni1-y

S

y

pS

2 / Pa

ymax in Ni1-yS ≈ 0,09

Z. Grzesik, 2005

The experimentally determined ratio between external thickness and the internal

part of the NiS2 sulfide equal to 0,82 remains in accordance with the theoretical

value obtained under the assumption that deviation from stoichiometry is present in

Ni1-yS on the level of 0,91

e64,1Ni82,0NiSSNi2 2291,0

222 NiS82,0S82,0e64,1Ni82,0



Au markers

a

b

s

homogenization: 873 K, 10 Pa, 24 h

homogenization: 873 K, 200 Pa, 24 h

sulphidation: 873 K, 1000 Pa, 5 min

b

a

s

10-1 100 101 102 103 104 105 10610-3

10-2

10-1

H. Rau, 1975

1168 K

1110 K

1038 K986 K

914 K

835 K

Ni1-y

S

y

pS

2 / Pa

10 11 12 13100

101

102

103

104

1051073 1023 973 923 873 823 773 723

T-1.104 / K-1

NiS2

Ni1-y

S

T / K

pS

2 /

Pa

NiSh2VSNi Ni22

1xNi

Influence of NiS homogenization parameters

on marker location at the initial stage of NiS2 formation

Au markers

Ni coating Ni coating

Marker location after NiS homogenization

Au markers

Ni coating Ni coating

Influence of MeX homogenization parameters on

marker location at the initial stage of MeX2 formation

Au markers

Marker location after MeX homogenization in the case where

oxidant pressure is lower than MeX2 dissociation pressure

Au markers

Marker location after sulphidation

of CoS to CoS2 (SEI)

Au markers

CoS2

CoS

Au markers

Marker location after sulphidation

of CoS to CoS2 (BSE)

substrate

product

Au markers

Marker location at the initial stage of CoS2 formation

on the surface of CoS homogenized

at the CoS2 dissociation pressure

CoS cross-section sulphidized at a sulfur vapor pressure

that enables CoS2 formation

Interior of CoS sulphidized at a sulfur vapor pressure

that enables CoS2 formation

CoS

Co

Cross-section of a Co sample covered with CoS

CoS2

CoS

Initial stage of CoS2 formation inside CoS

CoS2

CoS

Au markers

Marker location in CoS sulphidized to CoS2

Possible marker locations in a CoS-CoS2 system

depending on the predominant direction

of reagent diffusion

Au markers

Schematic illustration of marker location

in CoS sulphidized to CoS2

neCoCoSCoS2 n2

22n CoSSneCo

Schematic illustration of marker location

in CoS sulphidized to CoS2

Interior of NiS sulphidized at a sulfur vapor pressure

that enables NiS2 formation

NiS2

NiS

20 mm

Interior of NiS sulphidized at a sulfur vapor pressure

that enables NiS2 formation

Co3O4

CoO

Cross-section of a CoO sample covered with Co3O4

Co3O4

CoO

Au markers

Marker location in Co3O4growing on CoO

Surface of CoO formed as a result of Co oxidation

at 900oC and at 1000 Pa oxygen pressure

Surface of NiS formed as a result of Ni sulphidation

at 700oC and at 1000 Pa sulfur vapor pressure

Surface of CoS formed as a result of Co sulphidation

at 700oC and at 1000 Pa sulfur vapor pressure

Marker location on a developed CoS surface

Marker method – oxidation of a high entropy oxide

(Mg, Co, Ni, Cu, Zn) O

O2


Markers

O2 diffusion Me diffusion Me and O2 diffusion

O2 O2 O2

(Mg, Co, Ni, Cu, Zn) O(Mg, Co, Ni, Cu, Zn) O (Mg, Co, Ni, Cu, Zn) O



Presence of predominant disorder inside the cation sublattice

O2

(Mg, Co, Ni, Cu, Zn) O g2'

i

x

X

x

MX

2

1e2MXM

xX

''Mg2 Xh2VX

2

1

reaction location:h

mdAh

where:

h – thickness of the oxide formed on the marker layer

A – proportionality coefficient

d – oxide density

m – oxide mass change during oxidant pressure change



h = 12 μm

Conclusion:

In the case where predominant disorder exists in the cation

sublattice, markers should be located 12 micrometers below the

oxide surface.

Necessary conditions for obtaining reliable

results using the modified marker method

• Before marker deposition the oxide sample must be heated at

oxygen pressure lower than that used during oxidation of the

sample covered with markers.

• The change in deviation from stoichiometry of the studied oxide

must be large (order of a few percent) during the marker

experiment.

• Sample thickness should exceed 1 mm.

• Reequilibration time of defect concentration during oxygen pressure

change should be many time greater than around 1 minute (which

means a small chemical diffusion coefficient for defects and/or a

large sample thickness).

Conclusion:

Complete interpretation of a modified marker experiment is possible

after finishing both marker studies, as well as point defect

concentration and mobility studies in a given oxide.

Marker experiment – summary

Obtaining reliable results during predominant disorder studies

using the marker method in ceramic systems, especially porous

systems, is a much more difficult task compared to metal-oxidant

systems. In order to interpret results, precise analysis of the reaction

product growth location in the substrate is necessary. In the case of

ceramic substrates with high point defect concentration, the

materials should be homogenized before the “marking” process at a

highest oxidant pressure, at which the chemical compound

constituting the substrate remains stable. In the case of compounds

exhibiting large deviation from stoichiometry, this should be taken

into account when writing the appropriate chemical reactions, on the

basis of which marker locations inside the reaction product can be

foreseen.

Two-stage oxidation method

Me MeX

C(*X)

OUTWARD LATTICE METAL DIFFUSION

Ideal case(X and *X atoms do not mix together)

Me MeX

C(*X)

Real case(X and *X atoms mix together)

C(*X) – concentration of the oxidant isotope (tracer)

Me MeX

C(*X)

INWARD OXIDANT DIFFUSION ALONG GRAIN BOUNDARIES

Ideal case(X and *X atoms do not mix together)

Me MeX

C(*X)

Real case(X and *X atoms mix together)



SIMULTANEOUS DIFFUSION OF BOTH REAGENTS

Slow mixing together of X and *X atoms

Me MeX

C(*X)

Me MeX

C(*X)

Fast mixing together of X and *X atoms



Cu, O16 and O18 concentration profiles inside a sample obtained

during Cu2O oxidation at 1273 K and 105 Pa oxygen pressure

0 500 1000 1500 2000 25000

10

20

30

40

50

60

70

Cu2O

concentr

ation

/

at.

%

Time / s

Cu

O16

O18

CuO

M. Migdalska, Z. Grzesik, S. Mrowec, Defect and Diffusion Forum, 289-292, 429-436 (2009)

O16 and O18 concentration profiles in an Ni-Cr sample

after two-stage oxidation at 1323 K

I stage: Ar-20%16O2; 0,5 h

II stage: Ar-20%18O2; 2 h

I stage: Ar-4%H2-2%H216O2; 0,5 h

II stage: Ar-4%H2-2%H218O2; 2 h

David J. Young, „High temperature oxidation

and corrosion of metals”, Elsevier, Sydney 2016

Cr, Fe, O16 and O18 concentration profiles

in 13CrMo4-4 steel after oxidation at 823 K

Ar / 1% 16O2 / 1% C18O2; 2 h

T. Olszewski, „Oxidation mechanisms of materials for heat exchanging components in CO2/H2O-containing

gases relevant to oxy-fuel environments”, Forschungszentrum Jülich GmbH, Jülich, 2012

Ar / 1% 16O2 / 2% H218O; 2 h

Concentration profile of elements in a P92 steel

sample after oxidation in Ar / 1% 16O2 / 1% C18O2

atmosphere at 923 K



Concentration profile of elements in an INCONEL 617

sample after oxidation in Ar / 1% 16O2 / 2% H218O

atmosphere at 1173 K



Self-diffusion coefficient studies

Distribution of the concentration of a tracer inserted into the near-

surface layer of a MeX crystal (lattice diffusion)

x x20 0

c ln c

cc

Dt

x

Dt

0

2

2 4exp ln lnc

c

Dt

x

Dt

0

2

2 4

c – tracer concentration at distance x from the crystal surface,

c0 – tracer concentration on the surface before beginning the heating process,

t – heating time

D – self-diffusion coefficient (tracer)

Distribution of the concentration of a tracer inserted into the near-

surface layer of a MeX crystal (intergranular diffusion)

x x0 0

c ln c

ln

/c

D D d

D tx const

V g

V

2

4


t – heating time

DV – lattice diffusion coefficient

Dg – intergranular diffusion coefficient


Distribution of tracer concentration in a MeX crystal, when tracer

concentration on the crystal surface is constant (lattice diffusion)

x0

c

c c erfx

Dt

0 1

2


c0 – tracer concentration on the surface before beginning the heating process,

t – heating time

D – self-diffusion coefficient (tracer)


Scratch method

Me

MeX

Me

MeX

Me

MeX

Me

MeX

Me

inward diffusion outward diffusion mutual diffusion

Wagner’s pellet method

H. Rickert, Z. Phys. Chem. Neue Folge, 23, 356 (1960)

The range

of metal

loss

S S

AgAg

pellet

pelletAg S2Ag S2

scale

Ag S2Ag+

e-

S

Ag S2

Ag

pellet

Ag S2

scale

Ag+

e-

S. Mrowec, An Introduction to the Theory of Metal Oxidation, National Bureau of Standards and National Science

Foundation, Washington D.C., 1982

Modified Wagner’s pellet method

S

Ag

pellet

Ag S2

Ag S2

Ag S2

S

Ag

pellet

Ag S2

35S

Ag

pellet

Ag S2

Ag S2

Ag S2

Ag S2 *

Ag S2 *

Ag S

stop Zr-Cu ZrC W WC

5 mm

temperature: 1400 oC

time: 1,5 h

Cross-section of a WC sample that reacted with Zr

Z. Grzesik, M. B. Dickerson, K. Sandhage, of Materials Research, 18, 2135-2140 (2003)

THE END

methods of studying the corrosion process mechanismhome.agh.edu.pl/~grzesik/fhtc/3_methods of...

Documents