Mohammed - Ali16PhD Sus 304
Mohammed - Ali16PhD Sus 304
Mohammed - Ali16PhD Sus 304
Stainless Steel
by
Haval Bashar Mohammed Ali
Atmospheric pitting corrosion of austenitic stainless steels 304L and 316L under
droplets.
304L stainless steel plate was investigated. The presence of retained delta
ferrite was found to influence the morphology of pits. Ferrite bands were
preferentially attacked, and pits were found to have layered attack morphology
dependent on the rolling direction and plane of the metal surface. Solution
pits.
Pits can grow with an initial shallow dish, which may propagate via “earring” or
small satellite pits. The size and morphology were seen to vary with exposure
humidity, chloride deposition density and distance from the droplet edge.
alloy and exposure time. Below this value much smaller pits were observed due
observed for MgCl2 droplets on stainless steel when pits form close to or at the
humidities.
i
conditions of intermediate level nuclear waste (ILW) in order to minimise the risk
of pitting corrosion.
ii
Dedicated to
Peshmerga of Kurdistan, those who fight terrorism on behalf of the whole world
All the Kurdish martyrs who sacrificed their lives for the sake of a bit of Freedom
iii
Acknowledgments
I would like to show my gratitude for the Kurdistan Regional Government (KRG)
for sponsoring me in the most intense economic crises it faces nowadays to do
my PhD in Birmingham University which is one of the top universities in the UK.
I would like to thank Professor Paul Bowen whose support has been very
evident in helping me and my family.
I also highly acknowledge the help from Steven and Angus for their priceless
suggestions, comments and support.
I would like to thank and name individually all my colleagues in the corrosion
group who have been there for sharing ideas and discussions: Majid, Ali, April,
Weichen, Andrew, Fei, Rowena, Georgia, Sarah, Sophie and George.
Last but not the least; I would like to especially thank my wife Noor and my
daughter Norjan for their endless support, love and patience along my study.
Endless thanks to my lovely parents, brothers and sisters for their
encouragement over a growing number of years.
iv
Table of Contents
1 Introduction ................................................................................................. 1
v
2.10.6 Effect of microstructure ................................................................. 43
experiments.......................................................................................... 45
stores ................................................................................................... 48
3.1 Materials............................................................................................... 61
vi
3.6.3 Electron backscatter diffraction (EBSD) ....................................... 65
3.8.1 Distance from the shallow dish and/or pit centre to the droplet edge
..................................................................................................... 73
3.8.2 Measurement of shallow dish region, active region and total pit
diameter ....................................................................................... 74
vii
4.2.4 Effect of elongated inclusions on pit morphology ......................... 92
morphology .................................................................................. 95
4.3.1 Pit initiation sites in stainless steels and the dependence of pit
................................................................................................... 102
4.3.4 Effect of annealing on the ferrite content of 304L stainless steel 109
5.2.3 Effect of the pit position on the diameter of the shallow dish for
5.2.4 Effect of pit position on pit diameter and depth for 316L stainless
viii
5.2.5 Effect of pit position on pit diameter and depth for 304L stainless
5.3.3 Effect of pit position on pit depth and diameter ........................... 155
5.3.4 The critical relative humidity for atmospheric pitting corrosion ... 156
pitting corrosion of Type 304L and 316L austenitic stainless steels ....... 162
ix
6.4.3 Threshold CDD for pitting ........................................................... 192
7.2 Chloride deposition density (CDD) threshold for pitting ...................... 206
8.2 Study the effect of sample position on the morphology of corrosion pits
212
x
9.2 Appendix 2, data logger diagrams ...................................................... 220
10 References.............................................................................................. 222
xi
1 Introduction
corrosion. The high Cr content gives good resistance against general corrosion,
steel surfaces, which can then draw in moisture during normal fluctuations in
level waste (ILW) is currently stored in stainless steel containers above ground
with no immediate plans for final disposal. Aerosol deposition over decades can
concern that this may lead to pitting corrosion, which could in turn develop into
containers.
The initiation and propagation of pitting corrosion has been widely studied in full
immersion conditions e.g. [1-3], but limited work has been done on atmospheric
affected by variations in relative humidity (RH) [4-17], salt density [6, 15, 17-20],
salt type [7, 15-17], droplet area [4, 21-23], temperature [15-17, 24, 25] and
microstructure [24-29].
1
Atmospheric corrosion experiments can be simulated using several methods.
In this study, droplets of MgCl2 solutions are deposited on sheets of 304L and
propagation and morphology of atmospheric corrosion pits. Key factors are the
effect of residual ferrite on pit morphology, the critical chloride density below
which pits do not propagate, and the critical relative humidity above which pits
2
2 Literature Review
According to the microstructure, stainless steels are divided into four main
groups: austenitic, ferritic, martensitic, and duplex which contain both austenite
and ferrite. Stainless steels possess strong passivity due to their chromium
aggressive ions [30-32]. Generally, stainless steels contain 11-30 wt.-% Cr, and
~1-25 wt.-% Ni, and other alloying elements such as Mn, Mo, N. The reason
chromium oxide film (passive film) to protect the metal from corrosion [3], while
2.0 - 3.0 wt.-% Mo improves the corrosion resistance of austenitic steels and
imparts hot strength characteristics [33]. The low-carbon grades, 304L and
316L, are usually selected for the application of intermediate level nuclear
ferrite (Figure 2-1) [35]. The mechanism by which the residual delta ferrite forms
3
Figure 2-1 Microstructure of Type 304 stainless steel; fully austenitic (left) and mixture
of austenite and residual delta ferrite stringers (right) [35].
of the solidification process [36-38]. They exhibit four solidification modes based
by austenite [37, 39-41]. Type 304L has a primary delta-ferrite phase, followed
in (Figure 2-2). However, under more rapid solidification conditions, ferrite can
be retained as a minor phase in the matrix of austenitic stainless steels [42] due
rolling [35]. Retained ferrite usually present in relatively low fraction volume (2-
5%).
4
Figure 2-2 Fe-Cr-Ni equilibrium ternary diagram for constant-Fe section (70%
Fe) [43].
Unstable austenitic stainless steels such us 304L and 316L stainless steels are
(cold work deformation) [44-46]. Two types of martensite may form; martensite ϵ
[44].
5
2.3 Formation of inclusions in steel
steel. Based on their chemistry, inclusions can be classified into two types:
(which usually have both oxide and MnS phases together). Figure 2-3 shows
Figure 2-3 SEM images of different types of inclusions, (a) an oxide phase
without MnS in clean 304 stainless steel; (b) an oxide phase surrounded by
MnS in standard commercial 304 stainless steel; and (c) primarily MnS in
standard 304 stainless steel [50].
sulphur is added to the steel by ore and coke supplies to improve the
impurities remaining in the crude steel. If sulphur remains unalloyed after the
bulk has solidified, hot shortness can take place during hot working processes
(such as rolling) after casting. Considerable efforts have been directed towards
reactions. This is commonly carried out by addition of CaO, which reacts with S
6
and forms CaS (which floats off as a buoyant slag) and O. Elements such as
Desulphurisation does not remove all of the sulphur from the melt, and sufficient
sulphur may still remain such that precipitation at grain boundaries may still lead
steel [52]. Added Mn to the steel can react with S to form MnS inclusions, which
Deng et al. [53] proposed a possible mechanism for the formation of multi-
Factsage software that were consistent with experiment results. They proposed
that during solidification, inclusions of high melting point, such as Al2O3 and
CaS, can work as nucleation sites for MnS. TiOx, SiO2, MnO, Al2O3, and
sulphide inclusions, such as MnS, can precipitate during cooling of the melt. As
oxide core encasing it, or it precipitates in embedded form with the oxide
inclusion. Once the MnS phase is formed, Mn in steel matrix will diffuse into the
inclusion.
7
Figure 2-4 The mechanism of the evolution of complex inclusions during
solidification [53].
Holappa et al. [54] and Lind [55] described a simplified system for inclusion
formation based on the presence of four components: Al, Ca, O, and S. They
derived ten key reactions governing the inclusion formation and transformation.
Calcium treatment is widely used to modify the shape, composition and size of
the inclusions [54-56]. Figure 2-5 shows the change in inclusion morphology
8
Figure 2-5 Schematic presentation of the development of inclusion shape after
Ca- treatment of steel, figure reference is cited in reference [55].
9
Figure 2-6 Schematic illustration of the effect of Ca-treatment on inclusions in
steel, figure reference is cited in reference [55].
According to the morphology, as classified by Sims and Dahle [57], there are
three specific types of sulphides which can be found in steels: (1) randomly
dispersed globular sulphides (Type I), (2) elongated sulphides (Type II), and (3)
10
2.4 Methods for delta ferrite determination in stainless steels
diagram [58], DeLong diagram [59], and WRC-1992 diagram [60, 61], which
require data of the chemical composition of the weld metal. Although these
diagrams are primarily intended for use in welds, it has been reported that they
can be used to estimate the ferrite content of a base metal, weld consumable or
a weld metal and the microstructure of stainless steels [62, 63]. Experimental
in [63, 66].
reduce the presence of delta ferrite from austenitic stainless steels [36, 42, 67,
68]. Rho et al. [42] found that the amount of delta-ferrite (measured by
ferritescope) in 304L stainless steel is decreased from 0.99% to 0.23% after the
was also observed that, based on the selected annealing temperature and time,
11
Figure 2-7 Optical micrographs of 304L stainless steel (a) as-received with
narrow strips of delta ferrite in black parallel to the rolling direction; and (b)
solution-treated at 1050 °C for 1 h then water quenched. The etchant solution
was made from mixing of two solutions in equal amounts, solution one
(methanol 100 ml, nitric acid 25 ml and hydrochloric acid 25 ml) and solution
two (distilled water 100 ml and sodium pyrosulphate 1.5 g). The sample was
immersed for 5–60 s at room temperature in the mixed etchant solution. The d-
ferrite turns black [42].
The delta ferrite volume fraction was observed to decrease with increasing the
exposure temperature [67, 68] and time [67]. Kim et al. [67] studied the
dissolution kinetics of delta ferrite in 304 stainless steel samples and found that
it is a function of annealing temperature and time. Figure 2-8 shows that the
and time [67]. They claimed that a rapid dissolution was observed at the early
stage of dissolution but came under diffusion control at the normal stage without
reporting any data between 0 and 50 s. They attributed the rapid dissolution of
delta ferrite to the large difference in Cr concentration between delta ferrite and
austenite phases.
12
Figure 2-8 Change in delta ferrite fraction with annealing (dissolution)
temperature and time for 304 stainless steel [67].
Toor [69], found that the corrosion rate decreased with increasing annealing
In general, austenitic stainless steels have good ductility and toughness and
ferrite in relatively low fraction volume (2-5%) in the microstructure of cast alloy
can reduce the ductility and the toughness [35]. The presence of such small
13
because ferrite phase improves hot workability [70-73]. However, it has been
reported that this retained delta ferrite has detrimental effects on pitting
corrosion resistance for fully immersed stainless steels [74, 75]. These effects
phosphorus along the austenite/ferrite interface [74, 76], the formation of Cr-
depleted zone [76-78], the lower pitting resistance equivalent number (PREN) of
It has been reported that different electrochemical behaviours arise from the
difference in the chemical composition of delta ferrite and austenite phases [83,
84]; the active, active-passive transition curve of the alloy shifts to more
negative potentials due to the high chromium and low nickel contents [84].
selective dissolution behaviour and found two distinct peaks in the active-to-
duplex stainless steel. The higher anodic peak was associated with the
14
Figure 2-9 (a) Potentiodynamic polarization curves of 2205 duplex stainless
steels (DSS) operated in the H2SO4/HCl solutions with different compositions at
ambient temperature; (b) contributions of ferrite (α) and austenite (γ) phases on
the anodic current density at active-to-passive potentials [83].
stainless steel, Örnek et al. [82] found that the formation of different failure
FeCl3:MgCl2 after exposure at 50°C and 30% RH for long term exposure (368
days); with attack of both phases at the anodic center but the austenite at the
periphery of the droplet and the ferrite between the center and the droplet
15
2.8 Pitting corrosion of stainless steels
Stainless steels are protected from corrosion by a naturally formed passive film
on the surface. In the presence of chloride ions, however, stainless steels are
the pit site (anodic dissolution reaction takes place), the metal will dissolve and
produce metal ions, and then metal ions will react with the water to form an
acidic environment within the pit (pH will decrease inside the pit) which will
promote further growth of the pit. Consequently, oxygen will be depleted inside
the pit. Chloride anions will diffuse into the pit to maintain the charge balance.
The oxygen reduction reaction will take place outside the pit (cathodic area).
The rust (corrosion product) will be precipitated. Usually, hydrogen bubbles well
16
Figure 2-10 Schematic diagram of the electrochemical mechanism of pitting
corrosion in steel in neutral chloride solution [88].
In the following sections, the pit initiation, propagation and pit morphology are
discussed in detail.
[1, 51, 88-90] in commercial stainless steels. It has also been observed that
other types of inclusions such as multi-element (Cr, Mn, Al, Ti, V) oxides and
mixtures of sulphides and oxides can cause pit initiation [50, 91-93]. It has been
17
reported that when both MnS and multi-element oxide inclusions exist together,
pitting is generally not associated with the multi-element oxide inclusions but is
In addition to inclusion, other features for example delta ferrite, alpha prime and
sigma phase may also enhance pitting corrosion [85]. Figure 2-11 schematically
MnS inclusions [88, 89, 97-99]. Eklund [89] proposed that MnS inclusions can
dissolve and led to the formation of aqueous sulphide (H2S), sulphur (S) and
sulphate (SO42-). Castle and Ke also reported the formation of the elemental
sulphure (S). The formation of elemental sulphur (S), sulphate (SO42-) and
sulphide (H2S) has also been reported by Wranglen [88]. Both Lott [98] and
18
Web and Alkire [99-101] suggest that the MnS dissolution products can be
thiosulphate (S2O32-) which has been reported to enhance Cl- induced pitting
Figure 2-12 illustrates schematically the initiation of pitting corrosion at the site
Ke [97]. They proposed the following reactions for MnS dissolution in neutral
chloride solutions:
A local acidic solution will form by the oxidation of MnS to HSO 3-. Then Sulphur
19
It has been reported that initial attack on the metal close to the inclusion can
take place due to the lower pH associated with inclusion dissolution process
[88, 89]. It has also been reported that the formation of sulphur layer can
prevent the passivation of the metal and decrease the strength of metal-metal
The geometrical shape [7, 18, 90, 99, 100, 106-108], size [51, 91-93, 109],
composition [50, 90, 109-113] and distribution [50] of inclusions have been
reported to significantly influence pit initiation (the onset of pitting) and growth
on stainless steels. It has been proven that to cause stable pitting a critical size
of inclusions is needed [92, 109]. Both the geometry and the orientation of
inclusions were found to play a prevalent role in the pit initiation and further
developments of pitting on stainless steels. Webb et al. [106] have studied the
effect of MnS inclusion geometry and depth on pits and reported that narrow yet
deep inclusions led to pits, but large yet shallow inclusions did not. The
[95] in commercial stainless steels was attributed to the rolling process during
casting process.
generally accepted that inside a pit, a concentrated acidic chloride solution will
develop that prevents the formation of the passive oxide film and allows
continued dissolution.
20
As mentioned above, when pitting corrosion initiates, metal ions (Mn+) such as
These metal ions (Mn+) increase the acidity inside the pit through the hydrolysis
To balance the electrical charge of positive metal ions inside the pit, negative
Cl- ions which are small and have high diffusivity, move into the pit [1, 88] as
shown in Figure 2-10. Therefore the concentration of metal ions [Mn+] from
dissolution of metal inside the pit controls the acidity inside the pit. When the
concentration of metal ions increases, the acidity inside the pit increases and
this favours dissolution over passivation, thus when the rate of metal ions
production inside a pit is higher than the rate of metal ions escape from inside
After a while, the O2 inside the pit will be consumed in cathodic reactions and
there will be the shortage of O2 inside of the pit. However, the area outside of
the pit has relatively good access to O2 compared to inside the pit, and the
anodic reaction can continue inside the pit after the O2 shortage while cathodic
reactions continue outside the pit. Thus inside the pit will be a net anode and
21
It has been reported that the rate of metal dissolution in a pit decreases linearly
with the reciprocal of the pit depth. This indicates that the current density of the
Isaacs et al. [121] reported from 1D artificial pit electrode studies that a resistive
layer forms on the electrode interface. This layer self regulates its thickness as
the potential is varied. They reported that because of the existence of this layer,
the current is not dependent of potential, and showed that the diffusion rate of
positive metal ions from the layer is the same as the moving rate of the metal
In order to study diffusion effects in artificial pits, Tester and Isaacs [122] used
the Nernst-Einstein equation (Equation 2-5) for expressing the anodic current
𝑖 𝜕𝐶 𝑛𝐹 𝜕𝛷
= −𝐷 [ +𝐶 𝐷 ] Equation 2-5
𝑛𝐹 𝜕𝑥 𝑅𝑇 𝜕𝑥
where 𝛷 is the potential drop, n is the average valance of metal ions, F and R
When the potential drop (𝛷 ) is insignificant and can be ignored, for diffusion-
𝑛𝐹𝐷∆𝐶
𝑖𝑙𝑖𝑚 = Equation 2-6
𝑥
22
Where 𝑖𝑙𝑖𝑚 is the diffusion-limited current density, and ΔC is the metal ions
concentration difference between the resistive layer and the pit mouth.
Ghahari et al. [123] have studied 2D pits growing at 0.005, 0.01, 0.1 and 1 M
NaCl solutions under current and potential control using in situ synchrotron X-
ray radiography. Their results revealed that the pit depth is under diffusion
Pitting corrosion can be seen in different forms such as narrow deep, wide
Figure 2-13 A schematic diagram of the various shapes of pits shown in cross-
section [124].
23
The morphology of pits is a function of several parameters, such us metal
stainless steels and ferritic chromium steels, the formations of flat-walled pits
were evident only at potentials near to pit nucleation potentials [110, 125, 126].
Schwenk [125] reported the formation of “regularly etched pits, mostly hexagons
Cl- at low applied potentials (with low current density within pits). His analyses
of the etch pits lead to the conclusion that the pit bounding planes belong to the
type {111} having the lowest dissolution rate (the most closely packed crystal
hemispherical and isotropic pits with polished or dull interiors have been evident
to form. Schwenk also established that pits in their initial state are of
hemispherical form covered with a thin and porous layer of metal. He found that
the edges of the pits were almost, but not always, circular. Under hollowing and
24
Figure 2-14 regularly etch pits morphology formed in 18Cr-10Ni solution
annealed 1300 °C then water quenched steel surface [125].
dissolution of the 16Cr-Fe alloy in solutions containing CI- ions near the pitting
etch pits ( polyhedrons flat walled pits) formed of most closely packed {110} and
Similar observations were also made by Pickering and Frankenthal [127, 128],
they reported that pits are initially crystallographic in stainless steel and iron.
higher potentials (more noble). They observed two types of pits: (1)
crystallographic pits not associated with inclusions and (2) covered pits which
initiate at inclusions. The latter type was found to propagate radially underneath
the metal surface away from the inclusion and finally penetrate the surface from
25
2.9 Atmospheric corrosion of stainless steel
ocean waves [129]. Such aerosols then are carried by coastal wind patterns
and can be deposited on the metal surfaces. Later, these hygroscopic salt
particles will absorb moisture from the environment when the relative humidity
reaches a critical rate and form deliquescent salt droplets or thin electrolyte
temperature and RH they can greatly quicken the localised corrosion of metals
there can be easy access of oxygen to the surface. However, the chloride
deposition rate can be influenced by the design and location therefore changes
in chloride deposition rates are expected [130]. Besides, topography of land and
the intensity and direction of predominant winds can play a vital role in
26
2.10 Atmospheric pitting corrosion of stainless steel
number of authors under natural exposure conditions time [132, 133], under salt
droplets [4-6, 15, 21, 22, 134-139], and under other conditions e.g. inkjet tests
[22, 140].
There are several exposure factors which could affect the atmospheric pitting
level and relative humidity etc. The effect of these factors will be explored in
containing chloride ions [4, 5, 21]. It is widely accepted that MnS inclusions are
the most common sites for pit initiation in stainless steels under atmospheric
conditions [4, 5]. Figure 2-15 shows a schematic diagram of pitting corrosion
suggested by Tsutsumi et al. [4]. Hastuty et al. [5] have proposed a similar
27
Figure 2-15 Schematic drawing of pit initiation and growth under atmospheric
conditions [4].
researchers [4, 6, 8, 18, 21]. Tsutsumi et al. [4] investigated the pitting corrosion
of Type 304 stainless steel under droplets of MgCl2 solution for a certain time in
formation of a shallow type pit with always at least one small hole (~10-30 µm
diameter) in the centre of the corroded area (Figure 2-16), which proposed to be
the remnant of a MnS inclusion dissolved out. Consistent with the work of
Tsutsumi, Hastuty et al. [8] also reported the shallow type pit morphology using
stainless steel.
28
Figure 2-16 Laser-microphotograph and depth profile of pitting corrosion under a
MgCl2 droplet of 12 mm diameter 1.9 µm thickness exposed at 35% RH [4].
Maier and Frankel [21] also reported a shallow dish pit morphology while
droplets of MgCl2 using a Scanning Kelvin Probe (SKP). All of the above
mentioned researchers proposed the formation of a shallow dish pit type with a
small hole in the centre of the pit which assumed to be the initiation site. The
shallow dish was found to propagate in the lateral direction at the early stage of
initial attack. However, Maier and Frankel [21] further reported that after a
certain time the pit stopped growing in the lateral direction but continued to
“ear-shaped” pit as shown in Figure 2-17. The active region was considered to
29
Figure 2-17 Ear-shaped pit formed on 304 stainless steel under MgCl2 droplet
in atmospheric conditions[21].
Recently, Street et al. [6] observed a similar morphology under MgCl2 droplet on
304L stainless steel. They named it as “spiral pit”. More details about this
2.10.2.
2.10.2 Effect of RH
30
Figure 2-18 shows the relationship between the solution concentration (M) of
MgCl2 and (a) RH (%), (b) electrical conductivity (mhom-1), and (c) self-
[Fe2+] is lower.
Figure 2-18 Relationship between the solution concentration (M) of MgCl2 and
(a) RH (%); (b) electrical conductivity (mhom-1); and (c) self-diffusivity of Fe2+
(m2 s-1) in MgCl2 solutions at 30 °C. OLI Analyzer 9 used to calculate these
parameters. Humidities investigated in Street’s work are marked in (a) [6].
31
A number of studies were carried out to investigate the critical RH for pit
stainless steel beneath MgCl2 droplets in lab and reported that the critical
relative humidity for occurrence of pitting corrosion lies between 65% RH and
75% RH.
The critical RH for pit initiation was found to be related with the surface finish of
the material. The finer the surface finish, the lower the critical RH. For a study of
304 stainless steel with surface finish of 1000 grit carried out by Tsutsumi et al.
[141], the critical RH for pit initiation was found to be ~70% RH at 300 K. While,
with a surface finish of 2000 grit for 304 stainless steel, Nishikata et al. [142]
found that the critical chloride concentration for pit initiation was 7.5 M
with final polish surface (0.25 µm diamond suspension), the critical RH was
concentration [CL-]pit for onset of pitting of 304 stainless steel under MgCl2
droplets was in the range of 7.1to 8.4 M at 298 K. The critical chloride
concentration for onset of pitting has been reported by some other researchers
was used to record the corrosion potential beneath an electrolyte layer on 304
32
Cyclic RH fluctuation experiments were conducted by a number of researchers
[7-14]. In such experiments, it is widely accepted that pits will repassivate during
wetting stage (from low RH to high RH), while during the drying stage (from high
RH to low RH) new pits can initiate and grow. Recently Nam et al. [9] have
austenitic (SS304) and ferritic (SS430) stainless steels during the wet-dry
cycles beneath MgCl2 droplets. They reported that at 298 K, for SS304, the
RHpit was obtained to be in the range of 58%–47% while that for SS430 was
obtained to be in the range of 68%–48%. They also reported that both RHpit and
RHrep vary with varying temperature. Significant increase in both of them was
Chen et al. [10] reported that more meta-stable pitting occurred on the surface
of mild steel samples in the wet cycles than that in the dry cycles. Also, the
metastable pits were uniformly dispersed in wet cycles while locally distributed
in dry cycles. Hastuty et al. [8] investigated the pitting corrosion of 430 stainless
steel in drying process of chloride solution layer and concluded that pitting
increases.
Yadav et al. [11] investigated the corrosion behaviour of galvanized steel under
impedance technique and reported that the corrosion is accelerated in the early
33
RH was found to influence the pit morphology. Street et al. [6] have investigated
values (33% and 38%) satellite pits can form ( Figure 2-19(a, b)), while at higher
RH values (38%- 48%), spiral pits can form (Figure 2-19(c-d)). Both these pit
morphologies had a shallow dish region. Satellite pits were seen to be located
toward the edge of the droplet while spiral pits were seen to be located toward
the centre of the droplet. At 56% RH, However, pits were seen to have a
proposed that this variation in pit morphology is due to the difference in IR drop,
solution diffusivity and conductivity. Street et al. [6] also reported that the
diameter of the shallow dish region varies with the distance from the droplet
edge, with larger diameters seen towards the centre of the droplet while smaller
diameters seen towards the edge of the droplet (Figure 2-20). No significant
34
Figure 2-19 Typical examples of pit morphologies grown beneath MgCl2
droplets with initial chloride deposition density (CDD) of 750 µg/cm2 after
exposure at constants RH values, shown in the figure, for 1 day at 30 ºC; (a)
and (b) satellite pits; (b-d) spiral pits; (f) circular pit [6].
35
Figure 2-20 Shallow dish equivalent diameter versus the distance from the
droplet edge for pits grown under MgCl2 droplets with initial chloride deposition
density of 750 µg/cm2 on 304L stainless steel after 1 day of exposure at 30 °C
[6].
36
2.10.3 Effect of Chloride Deposition Density (CDD)
For a fixed droplet area and at controlled exposure RH, any change in the salt
deposition level will lead to a change in the height of the droplet because the
of that both the diffusion rate and resistance of the droplet can be different.
It has been reported that the electrolyte layer morphology can be affected by
the salt deposition density [18-20] under fixed exposure RH. Figure 2-21 are
a function of CDD (indicated in Figure 2-21) on 304 stainless steel pin samples
after exposure at fixed RH (45%) and 21±1°C, shown in Mi’s work [18]. She
beneath MgCl2 droplets of ≤100 µg/cm2, while at higher deposition levels (400–
solution height from Mi’s X-ray microtomography experiments shows that the
electrolyte layer height is increasing with increasing the salt deposition density
4000, 2000, 1000, and 400 µg/cm2 was 260, 120, 60, and 25 µm respectively.
For the lower depositing values (10, 20 and 100 µg/cm2), the electrolyte layers
were difficult to be measured because it was not clearly seen. For pitting
corrosion to take place, Mi found that a lower chloride deposition density within
37
Figure 2-21 Vertical tomography sections of MgCl2 salt solution morphologies
as a function of CDD on 304 stainless steel (2 mm pins) at 21±1 °C and 45%
RH [18].
Albores-Silva et al. [19] also reported that the continuity of the droplet was
interrupted at lower CDD values. They have seen the formation of ring-like
droplet morphology (Figure 2-22) under MgCl2 solution with a CDD of 10 µg/cm2
38
proposed that the interruption of film continuity was due to the droplet shrinking
Figure 2-22 General appearance of MgCl2 droplets with a CDD of 1000, 100,
and 10 µg/cm2 on 316L stainless steel following exposure for 12 weeks (left)
and chlorine EDX elemental mapping after equilibrium was reached (right) [20].
39
The chloride threshold issue is very important as, in the presence of stresses,
(AISCC) [15, 19, 20, 145]. Albores-Silva et al. [19] and Padovani et al. [20] have
reported that the lower threshold for the occurrence of AISCC is above 10 and
316L stainless steel. Figure 2-23 shows the chloride deposition density and the
temperature needed for the AISCC to take place under MgCl2 droplets on 316L
Figure 2-23 The influence of temperature and the chloride deposition density on
the occurrence of AISCC on 316L stainless steel under MgCl2 droplets at
~30% RH. Tests carried out at 30 °C and 50 °C were exposed for 12 weeks
while the exposure time for tests performed at room temperature (22 °C) was 52
weeks.
Padovani et al. [20] further reported that at higher RH (60% RH), there was
need to higher chloride deposition densities for AISCC to take place at 50 °C,
40
while at 30 °C, AISCC was not seen up to very high CDD values
(>10,000 µg/cm2).
Maier and Frankel [21] reported that under small droplets, pits initiated more
rapidly than under large ones in 33-34%RH due to faster evaporation rate and
however, can include more pitting initiation sites such as MnS inclusions.
Therefore, Maier and Frankel found that the probability of occurrence of pitting
corrosion was higher under a large droplet than under a small droplet at lower
chloride concentrations (high RH). They attributed this to the presence of more
susceptible defects within a large droplet than in a small droplet. Tsutsumi et al.
with decreasing droplet diameter and thickness, which was attributed to the fact
that the smaller exposed area can provide a less cathodic current. Similarly,
Wang et al. [23] found that as the droplet size increases the susceptibility of
stainless steel to localised corrosion increase as well. They suggested that the
increase of cathodic area and the three-phase boundary could be the reason of
this trend. Also, Mi et al. [22] found that the pit diameter increases with
increasing droplet diameter under constant RH. They attributed this to the lower
Near coastal areas the atmospheric environments contain relatively high levels
41
has a lower deliquescence point (33% DRH) [130] so at relatively low relative
MgCl2 [4, 5, 15, 21, 139, 146] pure NaCl [15, 23, 129, 146, 147] and artificial
sea water [4, 5]. Just recently, however, Guo [7] investigate the effect of mixed
Prosek and his co-workers [15] studied the corrosion resistance of eight
stainless steel grades, including type 304 and 316L, under manually deposited
salt solutions of pure MgCl2, CaCl2, and NaCl as a function of temperature (20-
50˚C) and RH (15-70%). They reported that both 304 and 316L were prone to
pitting in the presence of MgCl2 and CaCl2 at 20˚C and higher and at relative
humidity of 30% and 50% while at low ~15% RH and 30˚C they were
susceptible to SCC. However, no pits where observed under pure NaCl at 40˚C
concluded that pitting corrosion on 304 stainless steel only occur under a
study by Tsutsumi et al. [4], they studied the probability for pitting corrosion to
occur on 304 stainless steel under MgCl2 droplets in lab and reported that the
critical relative humidity for occurrence of pitting corrosion lies between 55% RH
42
and 75% RH corresponding to (8.5 M) and (4.9 M) chloride solution
recent work published by Davenport et al. [26] on 304L stainless steel plate, the
presence of retained delta ferrite, with the formation of circular layered pit
morphology on top surface of the plate and horizontal stripes on the end grain
and side grain of the plate. This layered morphology was attributed to the
In addition to residual delta ferrite, it is well agreed that pits in stainless steels
often initiate at inclusions under atmospheric conditions [4-6, 21]. It has been
reported that elongated inclusions play an important role in pit initiation and
propagation in austenitic stainless steels [7, 18]. Figure 2-24, from Mi’s work,
shows the growth of a pit with time on 304 stainless steel beneath a MgCl2
sign of pits, however, only a pre-existing defect parallel to the rolling direction.
After 17 hours, a pit was found at the site of the defect. The pit was found to
grow both radially and vertically with time. It was proposed that the propagation
of the pit down bottom was due to the pre-existing defect (probably a narrow
43
Figure 2-24 Time-dependent pit growth imaged with X-ray microtomography of
a pit on 304 stainless steel pin (2 mm diameter) beneath a droplet of MgCl2 of
CDD ~1000 µg/cm2 after exposure at 45% RH and 21±1 °C; top images are
horizontal sections and bottom images are vertical sections [18].
Similarly, Guo [7] found an elongated deep fissure growing together with the pit
(Figure 2-25) on a 304L stainless steel pin sample machined parallel to the
rolling direction while studying the effect of wet-dry cycling on pitting corrosion
that the deep fissure attack was because of the presence of elongated
inclusions or ferrite bands in the alloy. It is important to bear in mind that the
plate used in Guo’s work and the current work is from the same supplier.
44
Figure 2-25 Horizontal and vertical sections of a tomogram of a 304L stainless
steel pin sample beneath an MgCl2 droplet (CDD ~1000 µg/cm2) after wet-dry
cycling exposure at 21±1 ºC. Cyclic exposure conditions are indicated in the
figure [7].
corrosion experiments
In the literature, both indoor and outdoor exposure tests were carried out to
for a period of time [132, 133]. Aerosols salt particles present in the
45
environment can deposit on the sample surfaces and then leads to atmospheric
concentrated solution of salts in shape of droplets [4-6, 15, 21, 22, 134-139] or
salts particles [129, 148-152] can be deposited on the metal surfaces for
methods:
salt solution of a known concentration to the metal surface. This method is the
most common one in lab based experiments [4, 5, 18, 21, 136-138].
method the particle size, the distribution and the mass of salt can be controlled
by controlling the spray nozzle dimensions, the spraying time, and the settling
time (the time to keep the samples in the shielded environment after spraying)
[153].
(3) Inkjet printing technique can be used to deposit patterns of salt solutions
precision hydraulic system for deposition of arrays of solutions down to 0.5 µl,
46
In the present study, it is shown that Multiprobe II liquid handling system can be
corrosion.
using a probe for selecting salt particles with a chosen size then putting them on
metal surfaces [148]; Dissolving the salt in ethanol then depositing drops of that
solution on the sample surface and keeping them to dry out forming clusters of
salt particles [150, 151, 154]; exposing the sample to a salt fog produced by an
ultrasonic humidifier [129]. In the above methods the sample is kept in a humid
environment in order to allow the salt crystals to deliquesce and form droplets.
laboratory conditions such as Kelvin Probe (KP) [21, 155-159], Scanning Kelvin
[13, 14, 141, 142, 146, 160, 161], Electrochemical Impedance Spectroscopy
47
2.12 Overview of conditions within intermediate level nuclear
Figure 2-26 shows the air temperature and RH monitored in waste stores of
atmospheric conditions, the RH was found to fluctuate between 30% and 90%,
while the temperature changes between 0 and 50 °C [162] over a two year
period.
degree of corrosion was observed on the container base (including welds) than
on the other surfaces (centre images). However, very small pits observed on
horizontal ledges of the container (right hand images) but extensive rusting was
observed on areas of embedded iron or carbon steel (left hand images) [163].
Swab samples collected from different locations at the storages showed the
anions (e.g. chloride, nitrate, and sulphate) [163]. The concentrations of such
ions deposited onto the container surfaces were found to be a function of the
surface orientation. For example, small amount of chloride ion was found on
vertical and overhanging surfaces but a greatest amount of such ion was found
48
Figure 2-26 The external air RH and temperature observed in the 4 meter box
program for two years (from Nov. 2000 to Nov. 2002) [162].
Figure 2-27 Condition of 304L prototype container after ~10 years of storage
within an inland warehouse in Culham, UK. (left) region of embedded iron;
(centre) bottom of container, which was exposed to road-salts during transport
(CDD of some measurements >100 µg/cm2); (right) horizontal ledge (CDD
~10 µg/cm2)[163].
49
2.13 Secondary spreading and micro-droplet formation
One phenomenon of interest which can take place during the course of
secondary spreading region [164]. With time, the gradual merging of micro-
steel and under different salts solutions including MgCl2 by many researchers
It has been reported that the main driving force for the formation of micro-
the centre of the main-droplet and its periphery [129, 151, 156]. Development of
anodic region within the droplet and a cathodic region at the edge. The cathodic
reaction takes place at the edge of the droplet due to differential aeration, with
50
established for stainless steel [136, 156] and for aluminium [167, 168] under
NaCl droplets.
According to both Zhang et al. [151] and Tsuru et al. [136], there are three steps
for the formation of secondary spreading: firstly the formation of a thin “invisible”
electrolyte layer that contains clusters of water molecules on active sites such
as surface defects and protuberant sites. Secondly, such water clusters work as
the reaction sites for the cathodic oxygen reduction reaction. Finally, migration
of cations from the main droplet into peripheral region leads to the formation of
51
Figure 2-28 The mechanism of micro-droplets formation proposed by Zhang et
al. [151]. The formation of micro-droplets was considered to take place in the
following three main stages: (a) formation of a thin water film with clusters of
water molecules; (b) synergistic action of the oxygen reduction and the water
molecule adsorption; and (c) migration of cations and formation of micro-
droplets.
52
There are various variables that influence formation of micro-droplets, such as
[136, 151, 156, 164-166, 169]. The effects of these variables showed that the
little spreading at 75% but no spreading for RHs lower than 75%. Tsuru et al.
[136] reported that the secondary spreading area is greater at 90% RH than that
Both Tsuru et al. [136] and Zhang et al. [151] reported that secondary spreading
steel. They suggested that the secondary separating might be blocked by the
For aluminium alloys, it has been shown that when a pit is formed at the edge of
the main droplet, secondary spreading can take place close to the pit [166-168].
Morton and Frankel [167] studied the corrosion behaviour of AA7075‐T6 under
chloride droplets and compared it to that in bulk solutions. Figure 2-29 shows
3.5 wt% NaCl solution after exposure for 20 hours at an equilibrium humidity of
85%. A secondary droplet can be seen at the left of the main droplet (Figure
2-29(b)). The SKP line-scan shows that the advancing edge of the secondary
droplet was at highest potential relative to both the secondary droplet and the
53
remaining area of the main droplet, indicating this site and a wide area
surrounding it acted as a cathodic region to support the attack. The area under
the secondary droplet was at a lower potential relative to the entire main
droplet, indicating the site of the “anodic half reaction” [167]. The area of the
secondary droplet, marked as “1” in Figure 2-29, was covered with large
the secondary droplet and the remaining area of the initial droplet was found
between the secondary droplet and the arc of discoloration. The potential
decreases as scan goes away from the secondary droplet to the main droplet,
until reaching an “intermediate potential” far away from the secondary droplet
[167].
Li et al. [168], found that secondary droplets formed close to a pit at the edge of
the MgCl2 solution droplets on Al–Mg–Si alloy exposed at 33% and 75%
54
Figure 2-29 (a) Scanning Kelvin probe (SKP) line-scan profiles for potential and
height of a droplet test with 0.03 mM cerous ions and 3.5 wt% NaCl after
20 hours of exposure at 85% RH by the use of open beakers of saturated KCl.
(b) optical image of the droplet upon completion of the test. The arrow indicates
the direction and location of the SKP line-scan. Corresponding sites (1, 2, and
3) are labelled numerically.
It has been reported that the secondary spreading area increases with
increasing exposure time and their growth take place at the early stages of the
influenced by the superficial grinding marks. Both Bian et al. [166] and Zhang et
al. [165] reported that the covering area of micro-droplets developed outwards
in a semi-circle or oval shape. They linked this with the direction of the polishing
marks.
55
2.14 Synchrotron X-ray Techniques
generally consists of four main components: injection system (The electron gun
and linear accelerator “linac”), the booster synchrotron ring, the storage ring and
the beamlines [170]. The electron gun produces the electrons which then are
the electrons accelerate by a linear accelerator before entering into the booster
ring. Electrons are accelerated again in the straight sections of the booster ring
and then bent around the curved sections via bending magnets (electrons travel
at nearly the speed of light at this stage). Thereafter, the electrons are released
into the storage ring (to be accelerated for the third time) which is actually a
polygon with bending magnets at the corners. Insertion devices found in third
the form of X-rays. Finally, the X-rays enter the beamline, where they are
collimated and refined. The two types of X-ray source used for tomography are
56
Figure 2-30 Schematic diagram of a synchrotron X-ray facility [174].
Figure 2-31 shows the setup of in situ synchrotron X-ray tomography. A series
of radiographs is collected as the sample is rotated through 180°, and these are
57
Figure 2-31 Schematic diagram illustrating data collection of a synchrotron X-
ray microtomography experiments [164].
on austenitic stainless steel [7, 18, 107, 175-177], aluminium [164, 178-180],
the micro-meter length scale [123, 175, 178-181]. For atmospheric corrosion
steel pin. Mi [18] studied the effect of chloride deposition density and time
dependent pit growth on SS304 pin samples under MgCl2 droplets. Guo [7]
58
performed in situ X-ray microtomographic tests to study the influence of RH
fluctuations and the effect of mixed salts on SS304 stainless steel pins. She
also monitored the pit growth with time on such samples. More recently, Burnett
using multiple techniques together to study the same three dimensional (3D)
region of interest which was hidden beneath the sample surface. In their work,
researchers [7, 18, 182, 183]. Both Mi and Guo have reported that a greater
number of pits were observed in tomography tests than that formed in lab-
based tests.
2.15 Summary
59
so far have not been focusing on the effect of microstructure, especially the
effect retained delta ferrite. Besides, none of the previous investigations have
studied such effect on the three plate orientations. Therefore there was need to
investigate the effect of delta ferrite on pit propagation on all plate planes. Lab-
based experiments carried out to study such effect cannot show the undercut
attack in 3D. Therefore, synchrotron X-ray microtomography was used for this
purpose.
Since in realistic conditions the deposited salt particles (which can form different
stainless steel under MgCl2 droplets was not reported previously. In the present
work, such micro-droplets were observed and briefly investigated using time-
lapse imaging.
60
3 Experimental Method
3.1 Materials
Austenitic stainless steel sheet of grades 304L and 316L was obtained from
Aperam-France and Acerinox, respectively. The thickness of the 304L and 316L
sheets was 3 mm and 6 mm, respectively. The alloys were cold-rolled, solution
composition of the both materials as taken from the mill certificates (Appendix
1).
Table 3-1 Foundry specification of composition of the both alloys used in this
study.
Element
(wt%) Cr Ni Mn S Mo C Si N P Fe
Material
18- 8-
304L 2 0.015 - 0.03 0.75 0.1 0.045 Bal.
19.5 10.5
16.5-
316L 10-13 2 0.015 2-2.5 0.03 1 0.11 0.045 Bal.
18.5
A solution annealing treatment was carried out for 304L stainless steel samples
sealed in a silica glass tube that had been evacuated and backfilled with Ar gas.
The samples were annealed for 1 hour at 1050±5 °C and water quenched.
61
3.3 Sample preparation for lab-based droplet experiments
304L sheet was cut using a Buehler Isomet 4000 SiC blade into 20-25 mm
plane (top surface) and LS plane (side grain) and ST plane (end grain) surfaces
(Figure 3-1). For studies of effect of RH, droplet diameter, and CDD, samples
were cut in larger sections. The 304L plate was cut into 25 mm x 70 mm
samples and 316L plate was cut into 25 mm x 60 mm samples. The top surface
of the plate was then ground using 800 grit SiC paper and then ultrasonically
cleaned in deionised (DI) water (18 MΩ cm) for 10-15 minutes. An empty
squeeze plastic bottle was used to blow air on samples to dry them. The
For synchrotron experiments, 2 mm pins were machined from the middle of the
plate with their axis parallel to the rolling direction (Figure 3-1). Then pin
samples were ground to 800 grit using SiC papers, washed with deionised (DI)
62
Figure 3-1 Plate planes and the orientation of machined pin samples with
respect to the plate.
304L samples were ground and polished to a mirror finish using oxide polishing
100 ml HCl, 100 ml ethanol) for 5 s at lab temperature (~22 °C). SEM JEOL
6060, Phillips XL-30, and JEOL 7000 machines were used for pit and material
(EDX) systems and INCA analysis software for elemental analysis. EDX
distance.
3.6.1 XRD
63
accelerating voltage of 40 kV and current 40 mA, angular range 20°-100°, to
detect the presence of delta ferrite on the LT plane (top surface) of the base
used to detect the presence of delta ferrite in the as-received and solution
annealed 304L stainless steel alloy by measuring the magnetic hysteresis loop.
Before the measurements, the VSM was calibrated with a standard sample of
sample was a ~3 mm cubes (~0.19 g) that was attached to the end of the
64
3.6.3 Electron backscatter diffraction (EBSD)
through successive grades of SiC paper to 4000 grit using the minimum
polishing time at each grade. They were then successively polished using
colloidal silica suspension (0.25 µm) for 5 minutes each. EBSD was used to
quantify crystallographic phases and grain sizes using an FEI Quanta 650 SEM
connected with a Nordlys EBSD detector from Oxford Instruments with AZtec
Version 2.2 acquisition software. An area of 150 x 131 µm2 was scanned with
The indexing rate was typically close to 99%. Data post processing was carried
out with Oxford Instruments HKL Channel 5 software. The phase database
used in this study (Table 3-2) was included in AZtec software. For successful
Table 3-2 Ferrite and austenite crystallographic geometry parameters for EBSD
phase identification given by HKL database.
The grain size distribution was determined by the mean linear intercept method
boundaries). The average grain size was found to be 9±6 µm (Figure 3-3).
65
Figure 3-3 The grain size distribution determined by the mean linear intercept
method as the mean of vertical and horizontal directions (Excluding twin grain
boundaries) for 316L stainless steel sample (LT plane).
A free online tool (MIG WELD GmbH International) [184] was used to predict
the ferrite content using a Schaeffler diagram, a DeLong diagram, and a WRC-
alloy. By entering the cast analysis (Appendix 1) and minimum and maximum
alloy specification, the Cr-equivalent (Creq) and the Ni-equivalent (Nieq) were
calculated and shown in a diagram. Each diagram has its own Creq and Nieq
equations.
66
3.7 Atmospheric Corrosion Tests
this study is shown in Table 3-3. Tests were performed to investigate the effect
depositing water droplets containing MgCl2 onto the surface of each specimen
used. All corrosion tests were performed on specimens with 800-grit surface
finish. The exposure time was from 1 day to 6 weeks. Further detailed test
67
After exposure, the samples were photographed and droplets were individually
photographed and then the samples were rinsed with deionised water. The
corrosion products were removed by immersing the samples in dilute nitric acid
(2.9 M HNO3) for 1 hour at lab temperature. The diluted HNO3 was made by
solution.
Note that the droplets were deposited at ambient RH and Temperature, which
ranged from (44±6%, 25±1 °C) in the summer to (35±5%, 22±0.7 °C) in the
winter. The full array of 70 droplets was deposited within 8 -10 minutes.
The amount of chloride deposited was known from the volume and
determined using FIJI/ImageJ image processing software [185]. For tests which
68
via serial dilution with DI water of a stock magnesium chloride solution. The
system was programed to wash the pipette tip after each deposition set.
Figure 3-4 Micropipette salt solution deposition on 304L stainless steel (a) LS
and ST plane; (b) LT plane.
69
Figure 3-5 MultiPROBE II Ex robotic liquid handling system.
The relative humidity of tests carried out in desiccators was maintained by using
70
In all experiments, RH and temperature was monitored using RH/TEMP data
temperature for the experiments carried out in desiccators (RH was controlled
by the saturated salts) and the controlled atmospheric chamber are shown in
Appendix 2.
Table 3-4 Saturated salts used for RH control according to ASTM E104-02
[186].
Effective Temperature
RH (%) Salt Type
Range(°)
33 MgCl2.6H2O 5 – 80
43 K2CO3 5 – 30
56 NaBr 5 – 80
70 KI 5 – 80
75 NaCl 5 – 80
software was used to carry out time-lapse tests. Two time-lapse tests were
carried out in this work. One was used to monitor the pit growth, while the other
performed when test conditions were not sustained, such as out of focus of
71
3.7.5 Synchrotron microtomography corrosion tests
2 mm pin stainless steel samples were used for in situ X-ray microtomography
tests, with the top surface ground to 800 grit, and then washed with DI water
and methanol. ~1.1 µL droplets of 0.4 M MgCl2 were deposited on the end of
pins. Figure 3-6 shows a schematic illustration of the design of the cell used for
this type of tests. The design of the sample cell was set up by covering the base
side of the 2 mm pin stainless steel sample with a short section of silicon tube of
2 mm internal diameter, then wrapped with some parafilm for better sealing.
Then a longer silicon tube section of 4 mm inner diameter was used to cover
the cell. The humidity inside the sample cell was controlled by inserting a piece
of filter paper soaked with a saturated salt solution (mentioned in Section 1-8)
into the silicone tube. Then an aluminium cap was used to seal the cell. An
extra seal with Parafilm was performed to the top of the sample cell for effective
The tests were performed at Beamline I12 at Diamond Light Source. The
energy of the beam and the pixel size were 70 keV and ~1.8 µm, respectively.
During a 180° rotation for each sample, 1800 projections were collected. An
algorithm (filtered back projection) was used to reconstruct the projections. The
exposure time was typically 1 s per projection and for each sample the full scan
takes 40–45 min. In order to make 3D rendered images, Avizo software was
used [26].
72
Figure 3-6 Schematic diagram of cell used for in situ X-ray tomography
experiments using 2 mm diameter 304L stainless steel pin.
3.8 Measurements
3.8.1 Distance from the shallow dish and/or pit centre to the droplet edge
The distance from approximately the centre of the shallow dish region to the
“straight line” tool in FIJI/ImageJ software [185] (Figure 3-7). For pits where no
shallow dish region was present, a line was drawn from the centre of the pit to
the nearest droplet edge. The uncertainty in such manual measurement was
typically ±5 µm. In general terms the errors were determined from the standard
deviations of 5 measurements.
73
Figure 3-7 Measurement of the distance from the shallow dish region observed
during pitting. Pit was grown under MgCl2 droplet (~2.5 mm equivalent diameter
with ~1000 µg/cm2 CDD at deposition) on 304L stainless steel surface (LT
plane) after 1 week exposure at 33% RH and 30 °C.
3.8.2 Measurement of shallow dish region, active region and total pit
diameter
A Leica DMLM optical microscope equipped with a Leica DFC 420 digital
camera of maximum 5 mega pixel resolutions was used for optical microscopy.
For pits that showed a “shallow dish region”, the diameter of the shallow region
was determined, by considering that the shallow dish region was circular, by
drawing an circle (using the “oval” tool in FIJI/ImageJ software) around the
remaining dish region after pitting corrosion test (Figure 3-8). The area of the
74
drawn circle was measured directly from FIJI/ImageJ software. The equivalent
diameter of the circle was obtained from the area of the circle using the
(Equation 3-2). The uncertainty in the area of the shallow dish measurement
was ±64 µm2. Consequently, the uncertainty in the equivalent diameter of the
shallow dish was ±0.4 µm. The standard deviations are from 5 measurements.
4𝐴
𝑑=√ Equation 3-2
𝜋
Figure 3-8 Measurement of the shallow dish region observed during pitting. The
pit was grown under MgCl2 droplet (~2.5 mm diameter with ~1000 µg/cm2 CDD
at deposition) on 304L stainless steel surface (LT plane) after 1 day exposure at
33% RH and 30 °C.
75
Pits that did not show a shallow dish region and had only circular morphology
(e.g. at ≥ 56% RH for 304L and ≥ 48% RH for 316L), the pit mouth area (total pit
area) was selected and measured using the semi-automatic “wand” tool
combined with the “Freehand selections” tool in FIJI/ImageJ software. Then the
equivalent pit diameter was obtained from (Equation 3-2) based on the
Figure 3-9 shows the methodology for selecting the active region and the total
pit area. The active area is defined as the deep region of the pit outlined in red.
The total pit area is defined as the area as outlined in yellow. The equivalent
area as a circle. The uncertainty in the active area and the total pit area
measurements was ±300 µm2 and ±15 µm2 respectively. Consequently, the
uncertainty in the equivalent diameter of the active region of pit and the total pit
area was ±2.1 µm and ±0.1 µm respectively. The standard deviations are from 5
measurements.
76
Figure 3-9 Technique used for measuring the active region and the total pit
area. The pit was grown under a MgCl2 droplet (~2.5 mm diameter with
~1000 µg/cm2 CDD at deposition) on 304L stainless steel (LT plane) after 1 day
exposure at 43% RH and 30 °C.
A Leica DMLM optical microscope with fine focus resolution of 1 µm was used
to characterize the pit size for lab-based experiments. The pits depth was
measured using the fine focus method [124], where the pit depth was estimated
as the difference between optical focus on the surface of sample and the optical
focus on the base of the pit (the lowest observable point). The uncertainty in pit
77
4 Effect of microstructure on the morphology of
4.1 Introduction
It is widely accepted that small quantities of residual ferrite can be found in “fully
austenitic” stainless steels such as 304L and 316L [36, 42, 68, 70, 187, 188].
However, its effect on pitting corrosion has not been studied under atmospheric
annealing is also outlined. Both MnS and mixed oxide inclusions were
characterised.
4.2 Results
4.2.1 Microstructure
The ST plane of the 304 stainless steel plate has a microstructure that is
4-1(b)). Bands can be seen parallel to the surface of the plate (in the plane of
the rolling direction). In the EBSD map (Figure 4-1(b)), it can be seen that these
elongated features are red, indicating that they are ferritic (point 1). However,
there are other regions that appear to have a ferritic structure that are not
78
parallel to the plate surface, such as point (2 and 3) in Figure 4-1(b). These
regions are not obviously distinct on the SEM image, and might possibly be
diffraction (XRD) (Figure 4-2). There are strong austenite peaks and one
smaller peak that correlates with the ferrite (110) peak. The presence of ferrite
magnetic field, for 304L samples before and after solution annealing. It was
found that the alloy behaves in a soft magnetic manner in the presence of a
magnetic field, consistent with the presence of the ferrite phase. After solution
79
Figure 4-1 304L stainless steel, ST plane, rolling direction normal to the page
(a) SEM image showing the microstructure; (b) EBSD of ST plane.
80
Figure 4-3 Hysteresis loops of 304L stainless steel obtained by VSM: base
material shown in solid red and solution annealed shown in dashed red.
A Schaeffler diagram (Figure 4-4), a DeLong diagram (Figure 4-5), and a WRC-
1992 diagram (Figure 4-6) were used to predict the ferrite content [184]. The
predicted volume fraction of ferrite content was 12%, 6.2% and 8% from the
81
Figure 4-4 Schaeffler diagram showing the predicted prevalence of austenite
(A), ferrite (F) and martensite (M); the blue point represents the theoretical
ferrite content of the 304L alloy used in this work using the procedure described
in Section 3.6.4; the red box outlines the range of standard analysis based on
composition range.
82
Figure 4-5 DeLong diagram showing the predicted prevalence of austenite (A),
ferrite (F) and martensite (M) as a function of composition; the blue point
represents the analysis of ferrite content of the 304L alloy used in this work
using the procedure described in Section 3.6.4, the red box outlines the range
of standard analysis .
83
Figure 4-6 WRC-1992 diagram shows the predicted prevalence of austenite (A),
ferrite (F) and martensite (M) as a function of alloy composition; the blue point
represents the analysis of ferrite content of the 304L alloy used in this work
using the procedure described in Section 3.6.4, the red box outlines the range
of standard analysis.
Figure 4-7 shows SEM images of the three plate orientations (LT, LS and ST) of
304L stainless steel after 5 s etching with Kalling's 2 reagent following polishing.
The bands along the rolling direction were preferentially dissolved. These bands
correlate with ferrite bands, as observed in Figure 4-1. Wider bands were
84
Figure 4-7 SEM images of the microstructure of the LT, LS and ST planes of
304L stainless steel plate after 5 s etching with Kalling’s 2 reagent.
The effect of plate orientation on pit morphology was investigated for 304L
stainless steel. Two sets of specimens were tested; one set was exposed for
1 week while the other one exposed for 6 weeks. The exposure RH was 33, 43,
~1000 µg/cm2 CDD) were deposited manually onto the three plate planes (LT,
LS, and ST) using a micropipette. At least 10 droplets were deposited on the LS
and ST planes while 12-20 droplets were deposited on LT plane (see Figure
85
3-4). Summary of the results of the pitting behaviour of these tests is given in
Table 4-3.
Figure 4-8 SEM images of pits grown on 304L stainless steel plate (LT, LS, and
ST planes) under droplets of MgCl2 (chloride deposition density, CDD, of
~1000 µg/cm2), 30 °C, after 1 week of exposure at 33±2% ,43±2%, 56±2% RH.
The corrosion products were removed by immersing the samples in dilute nitric
acid (2.9 M HNO3) for 1 hour at room temperature.
Figure 4-8 shows the SEM images of pit morphologies on the three plate
orientations at 33±2%, 43±2%, and 56±2% RH. The LT plane (top surface of
the plate) shows layered attack morphology. Both LS and ST planes show a
rolling direction. This was consistent at all exposure RHs. However, the effect of
86
Figure 4-9(a) shows a droplet of MgCl2 solution on the LS plane that has been
corroding for 6 weeks. Figure 4-9(b) shows the surface following removal of the
droplet and washing in deionised water. The primary pit is elongated in the
rolling direction. Figure 4-9(c) shows the sample after sectioning through the pit
along the line indicated in Figure 4-9(b), and Figure 4-9(d) shows an SEM
image of the sectioned sample following cleaning in dilute nitric acid at room
temperature for 1 hour and polishing with OPS. The section on the left side
(ST), shows narrow fissures of attack deep into the surface, and the pit surface
on the right side (LS) shows bands of attack in the pit surface. Figure 4-9(e)
shows an SEM image of the fissures in the ST plane, and (Figure 4-9(f)) shows
a higher resolution image of the tip of a fissure. The tip of the lower fissure is
located in a band of narrow grains that was confirmed by EDX mapping (Figure
than the adjacent austenite grains. Table 4-1 shows the chemical compositions
elemental composition of both phases were measured for Fe, Cr, Ni, Si, Mn and
S content in weight percent. 5 point scans were measured over the centre of the
ferrite band at different regions, and a similar number of scans were measured
over an austenite region. The average chromium and nickel content of ferrite
phase was 24±1% and 4±0.5% respectively, while that for austenite phase was
87
Figure 4-9 304L plate (LS) with a droplet of MgCl2 (CDD ~1000 µg/cm2); (a) after 6
weeks at 33±2% RH, 30 °C; (b) washed with DI water (c) section of pit; (d) SEM of
sectioned sample, tilted (45°), showing both end grain (ST) and side grain (LS) sides of
the pit; (e) polished with OPS end grain side (ST); (f) magnified part of (e); (g) EDX
maps of region outlined in (f). The rolling direction (RD) is indicated.
88
Table 4-1 The average elemental composition with standard deviation of ferrite
and austenite phases in weight percent (wt.-%) for 304L stainless steel, ST
plane.
Weight % Cr Ni Mn S Si Fe
Figure 4-10(a) shows a droplet of MgCl2 solution on 304L stainless steel (LS
plane) that has been corroding for 6 weeks at 30 °C and 33±2% RH. A single
pitting site had developed at the edge of the droplet (Figure 4-10(a)). After being
washed (Figure 4-10(b)), the pit was lightly polished to a mirror finish (Figure
along the pit surface was seen to follow narrow grains (Figure 4-10(e)) that
were higher in Cr and lower in Ni than the adjacent austenite (Figure 4-10(f)),
89
Figure 4-10 (a) a 304L stainless steel plate, LS plane, with a droplet of MgCl2 on
its surface (CDD ~1000 µg/cm2) after 6 weeks of exposure at 33±2% RH and
30 °C; (b) SEM image of pit following gentle washing with deionised water; (c)
following gentle polishing with OPS; (d) magnified region of (c); (e) magnified
region of (d); and (f) EDX line scan through the ferrite band shown in (e).
is shown (Figure 4-11). The layered attack is clear from the surface (Figure
4-11(a)), with corrosion not only propagating down into the plate but also
parallel to the rolling direction. After focused ion beam (FIB) milling adjacent to
the pit mouth (Figure 4-11(b)), it is clear that dissolution parallel to the rolling
direction occurs under the metal surface in a planar morphology consistent with
attack on the ferrite bands. The walls of the pit also show this preferential
attack.
90
Figure 4-11 (a) SEM image of a pit grown on 304L stainless steel plate, LT
plane, with a droplet of MgCl2 (CDD ~1000 µg/cm2), after 1 week of exposure at
30 °C, 56±2% RH; (b) after FIB milling, sample tilted at 52°.
(shown in red) again shows propagation of the pit along the rolling direction
(Figure 4-12(b and c). An SEM image of the pit surface after 79 hours (Figure
4-12(d) shows that the grinding direction on the surface of the pin is
In total, six tomography pin samples were examined, two samples at each
exposure RH (33, 43, and 56% RH). It should be noted that there were always
multiple pits per droplet ‘’sample’’ (a greater number of pits per droplet
compared with lab-based tests). This was attributed to the effect of beam
damage, which has been reported previously [7, 18, 182, 183, 189].
91
Figure 4-12 In situ X-ray tomogram of a stainless steel pin, ST plane, exposed
to an MgCl2 droplet (CDD ~1000 µg/cm2) at 21±1°C and 59% RH for 32 hours;
(a) horizontal section of the tomogram with the surface of the pit shown in red
(segmentation using Avizo); (b) top and (c) side 3D view of the pit; (d) SEM
image after 79 hours.
In Figure 4-13(a) (LS plane), an elongated inclusion can be seen along the
ferrite bands parallel to the rolling direction of 304L base material. Figure
4-13(b) shows an elongated pit on the same plane formed under a droplet of
33±2% RH and 30 °C. Such elongated pits were also observed at 43±2% and
56±2% RH (see Table 4-3). The inclusions on three plate planes were
92
characterised with EDX for the alloy used for these experiments, and found to
be either MnS or mixed oxides. SEM and EDX maps of MnS and a mixed
oxides inclusion are shown in Figure 4-14 and Figure 4-15, respectively. EDX
inclusions, a typical example of each type and the chemical composition of the
Figure 4-13 (a) 304L base material, LS plane; (b) an elongated pit on LS plane
formed under a droplet of MgCl2 (chloride deposition density ~1000 µg/cm2),
after 1 week of exposure at 33±2% RH and 30 °C. Following removal of
corrosion products by immersing the sample in dilute nitric acid (2.9 M HNO 3)
for 1 hour at room temperature.
93
Figure 4-14 SEM and EDX maps of an MnS inclusion on a mirror polished 304L
stainless steel sample (LS plane); the arrows indicate the location of a ferrite
band.
Figure 4-15 SEM and EDX maps of a mixed oxide inclusion on a mirror
polished 304L stainless steel sample (LS plane).
94
Table 4-2 The chemical composition of typical MnS and mixed oxide inclusion
observed in 304L stainless steel detected with EDX.
Weight % O Mg Al Si S Ca Ti Cr Ni Mn Fe
MnS-
0 0 0 0.2 8.3 0 0 16.3 5.0 19.4 balance
containing
mixed
18.7 4.7 4.3 11.9 0.1 11.5 0.9 10.7 2.8 2.00 balance
oxides
Base
0 0 0 0.6 0.1 0 0 18.6 8.2 1.9 balance
metal
morphology
The 304L plate was solution treated at 1050 °C for 1 hour before quenching in
water. Figure 4-16 and Figure 4-17 shows the microstructure of 304L stainless
steel (LS plane) that has been etched with Kalling’s 2 reagent for 5 s. It can be
seen that there is no preferential attack of ferrite bands, suggesting that the
ferrite has transformed to austenite. It can also be seen that only certain grains
have been attacked. Furthermore, Figure 4-17(b) shows that the annealing twin
is free from the attack. It suggests that the attack is related to the
crystallographic orientation.
Figure 4-18(a) shows a pit from an atmospheric corrosion test on the LT plane
(top surface) of the plate without heat treatment, and Figure 4-18(b) shows a pit
on the plate after heat treatment. Figure 4-18(a) shows the layered morphology
usually seen for the plate resulting from attack on the ferrite phase (see e.g.
plate, which shows faceted crystallographic attack with no sign of layers. The
difference in pit morphology is even more striking on the LS plane (Figure 4-19).
The layered attack is seen for the as-received plate, whereas crystallographic
95
facets are again seen following heat treatment. This suggests that the heat
treatment has greatly decreased the amount of residual ferrite in the plate. This
Figure 4-16 (a) Microstructure of the solution annealed 304L stainless steel
plate (LS plane) after 5 s etching with Kalling’s 2; (b) single grain with
preferential attack.
Figure 4-17 (a) Microstructure of the solution annealed 304L stainless steel
plate (LS plane) after 5 s etching with Kalling’s 2; (b) A grain with preferential
attack with annealing twin unattacked.
96
Figure 4-18 SEM images of pits on 304L stainless steel plate (LT plane); (a)
base material; (b) solution annealed at 1050 °C for 1 hour followed by water
quenching. Pits grown on 304L stainless steel plate under droplets of MgCl2
with a chloride deposition density of ~1000 µg/cm2, for 1 week at 43±2% RH
and 30 °C.
Figure 4-19 SEM images of pits on 304L stainless steel plate (LS plane); (a) as
received; (b) solution annealed at 1050 °C for 1 hour followed by water
quenching. Pits grown on 304L stainless steel plate under droplets of MgCl2
with a chloride deposition density of ~1000 µg/cm2, for 1 week at 43±2% RH
and 30 °C.
97
Table 4-3 summarises the results of the corrosion tests carried out in this
section on the different plate orientations of 304L stainless steel. The pit
morphologies and the number of droplets pitted at each exposure RH and time
are presented. It can be seen that the faceted pit morphology (marked with red)
was associated with the solution annealed material. A layered pit morphology
(marked with blue) was found on the LT plane of the steel while a striped pit
with violet) were usually observed on both LS and ST planes but more often on
LS plane. Details of the test conditions can be taken from Table 4-3. The ratio
x/y in the table represents the number of droplets with pits/total number of
droplets investigated. In total, 274 droplets were investigated in this section, 261
98
Table 4-3 Summary of the results of pitting behaviour on the three plate
orientations of 304L stainless steel under MgCl2 droplets (~2 mm equivalent
diameter, ~1000 µg/cm2 CDD at deposition) at different exposure time and RH.
Number of
Exposure Material Pit
RH (%) Orientation droplets
time condition morphology
pitted
LT As-received 14/16 Layered
9 Striped
33% LS As-received 11/12
2 Elongated
10 Striped
ST As-received 12/12
2 Elongated
As-received 26/30 Layered
LT
Solution
20/20 Facet pits
annealed
1 week 6 Striped
43% As-received 12/12
LS 6 Elongated
Solution
9/14 Facet pits
annealed
ST As-received 12/12 Striped
LT As-received 16/16 Layered
56% LS As-received 12/12 Striped
ST As-received 12/12 Striped
LT As-received 12/12 Layered
33% 5 Striped
LS As-received 10/10
5 Elongated
ST As-received 10/10 Striped
LT As-received 12/12 Layered
6 weeks 43% 9 Striped
LS As-received 11/11
2 Elongated
ST As-received 11/11 Striped
LT As-received 12/12 Layered
7 Striped
56% LS As-received 11/12
4 Elongated
ST As-received 10/10 Striped
32 hours 33% ST (pin) As-received 2/2 Striped
Tomography 43% ST (pin) As-received 2/2 Striped
pin samples 56% ST (pin) As-received 2/2 Striped
99
4.3 Discussion
4.3.1 Pit initiation sites in stainless steels and the dependence of pit
In the 304L plate used in the current work, extended inclusions (10-20 µm in
length) of both MnS and mixed oxides (Figure 4-14 and Figure 4-15,
chloride deposition density ~1000 µg/cm2 at 33±2% RH. It is likely that these
inclusions are the initiation sites for these very long pits.
to MnS inclusions [4-6, 18, 21]. Li and Hihara [129], however, found evidence
that slag based inclusions under relatively large NaCl droplets can initiate
pitting.
parallel to the rolling direction [95]. The alloy used in the current work is cold
Some authors have reported the initiation and growth of pits along the extended
needle shaped inclusions [7, 18, 95, 107, 190]. Mi [18] investigated pitting
100
corrosion on the end of a SS304 rod under MgCl2 droplets using X-ray
microtomography technique and found that a deep pit can develop with time
stainless steel in NaCl solutions and found that a pit can grow deeply into the
steel matrix beneath the inclusion resulting in the formation of a large occluded
cavity.
The rod used in Chiba et al.’s work was cut into coupons parallel to the rolling
direction (i.e. inclusions will be parallel to the rolling direction). The pre-existing
defect (probably a MnS inclusion) observed in Mi’s work was also parallel to the
rolling direction. The elongated pit morphology shown in Figure 4-13(b) was
observed on the side grain of the plate (elongated inclusions will be parallel to
the rolling direction and perpendicular to the grain of the plate). Therefore, it is
It has been reported that pit initiation can take place at the delta ferrite
(δ)/austenite (γ) interface in stainless steel alloys [43, 74, 77]. Tomashov et al.
oxide film would form over the interface, resulting in a more susceptible portion
of the surface for pit initiation. Manning et al. [43, 74], however, disproved the
101
segregation of impurities such as sulphur and/or phosphorus to the boundary
increases the dissolution tendency of protective film over the δ/γ interface. In
addition, S might tend to inhibit the repassivation kinetics at the δ/γ interface.
In the present study, it can be seen that elongated inclusions (MnS and mixed
oxide inclusions) were found to be aligned with the residual delta ferrite
stringers (Figure 4-13a and figure 4-14), which might had a direct role in pit
e.g. 304L and 316L [35, 36, 42, 44, 67, 68, 70, 187, 191-194].
In fully austenitic stainless steels, the presence of residual delta ferrite (2-5%)
68, 70]. The formation of delta ferrite in austenitic stainless steels has been
processes such as cold rolling and hot rolling. Consequently, ferrite stringers
are usually aligned along the rolling direction [35, 191]. The alloy used in this
work was cold-rolled and solution treated at 1040-1100 °C followed by forced air
102
cooling. Therefore, it is reasonable to expect the formation of residual delta
In austenitic stainless steels, the cold working process can lead to the formation
of strain (deformation) induced martensite α' (bcc) [192]. It has been reported
and grain boundaries [44, 193]. In this work, the occurrence of non-austenitic
phase along the twin boundaries (Figure 4-1(b region 2)) and as equiaxed
grains (Figure 4-1(b region 3)), which are not orientated along the rolling
been observed previously in similar alloys [194]. The trace presence of this
working [194].
The amount of ferrite predicted from the 304L alloy used in this research is
estimated via predictive methods; the Schaeffler diagram (Figure 4-4), DeLong
Moreover, these diagrams do not consider the effect of the cooling rate upon
solidification of the residual delta ferrite. The Schaeffler diagram has not been
verified as accurate for 304L but is known to be valid for a wide range of other
103
EBSD was used to quantify the presence of ferrite experimentally. In this
technique, it is not possible to distinguish between delta ferrite (bcc) and strain-
induced martensite α′ (bcc) so their volumes are combined. The volume fraction
correlates with the amount obtained from DeLong diagram within a reasonable
XRD can be used to detect and quantify delta ferrite and both martensite
phases. However, XRD has limitations in use that include a minimum volume
needed to achieve sufficient signal (quoted as 5% vol [45, 65]). XRD is also
sensitive to orientation effects [195]. The small peak of delta ferrite observed in
It has been reported that VSM can be successfully used to determine the
amount of residual ferrite in stainless steel alloys [64], as well as stress induced
martensite [45, 192]. However, the accuracy of VSM method depends on the
Moreover, if more than one ferromagnetic phase is present in the sample the
making identical (in weight) samples for comparisons. In addition, in this work,
due to system limitations it was not possible to completely saturate the samples
(i.e. it was not possible to reach the linear parts of magnetisation) to obtain the
ferromagnetic phase and can be obtained from intercepting the linear parts of
magnetisation curves back to the zero applied field). Hence, the amount of
residual ferrite was not quantified. However, Figure 4-3 shows that there is a
104
decrease in magnetisation after annealing which suggests a decrease in ferrite
content.
Since strain- induced martensite α′ (bcc), and residual delta ferrite (bcc) are
very similar in 304L stainless steel, certainly in terms of XRD and magnetic
metallography, then do EDX in the SEM. If there are regions of delta ferrite they
should have lower Ni content (significantly different) than austenite and strain
induced martensite. An EDX map (Figure 4-9) and line scan (Figure 4-10)
across the elongated micro grains successfully showed Cr-enriched and Ni-
Figure 4-8 show that the morphology of corrosion pits was affected by the
orientation of residual delta ferrite in the plate. Layered attack morphology was
planes, which appears as horizontal sheets parallel to the rolling direction. In all
cases, the ferrite was preferentially attacked, as can be seen, for example, in
MnS inclusions, for stainless steel. However, pit initiation can also take place at
the delta ferrite (δ)/austenite (γ) interface in Type 304L [43, 74] and 18Cr-14Ni
stainless steels [77]. In this work, it seems the growth of atmospheric corrosion
105
pits takes place preferentially in the residual delta ferrite. The sheets of
undissolved austenite inside the pit might act as a barrier for diffusion of ions
delta ferrite inside the pit. To propagate, pits require the diffusion of metal ions
away from the corroding surface to be less than the generation of metal ions
[118]. In a relatively broad pitting surface, such as on the LT plane, the required
area requiring a sufficiently high current density to maintain a salt layer is much
higher than that inside the crevices of a LS/ST plane pit. As such, the cathodic
current needed to be drawn on the LT plane is higher. This would cause these
higher RH. Pits on the LS/ST planes would require much less total cathodic
current to be drawn to keep the pit alive as only the narrow region in between
the remaining austenite ridges would need to remain active which has a
Figure 4-9 and Figure 4-10 clearly show that ferrite bands have been
morphology to the pit. Pitting on the LT plane (top surface of the plate) showed
a circular layered structure. FIB milling next to a pit on the LT plane showed that
attack undercut the surface and propagated along the ferrite bands (Figure
4-11). From the tomography results (Figure 4-12), it is evident that attack on the
ST plane after 32 hours also shows propagation parallel to the plate surface
and thus parallel to the ferrite bands. This is confirmed by SEM after the
106
experiment. Overall, the orientation of residual ferrite in the plate strongly
Prosek et al. [15, 16] studied the corrosion resistance of eight stainless steel
grades, including type 304 and 316L austenitic stainless steels, under salt
and RH (15-70%). They reported that, for duplex stainless steels exposed to
different conditions, the corroded phase was ferrite, which contained more Cr
(identified by the use of Energy Dispersive X-ray (EDX) analysis. However, they
didn’t give information about austenitic stainless steels type 304 and 316L.
often observed that one phase, either ferrite or austenite, corrodes preferentially
[15, 16, 24, 25, 27-29]. This depends on the exposure conditions and material
nitrogen (N), which is one of the strongest austenite stabilizers, in highly alloyed
duplex stainless steels. They found that the corroded phase in the tested
stainless steels with nitrogen alloying was ferrite while those without nitrogen
alloying were austenite. They attributed this to the higher dissolved nitrogen in
austenite phase than in ferrite phase which might increase the pitting resistance
Based on this work and the results of the energy dispersive X-ray (EDX)
analysis (Figure 4-8, 9, 10), it can be seen that the ferritic phase was
107
steel was exposed. Based on the cast analysis given in mill certificate (see
Appendix 1), the 304L stainless steel had 0.072 wt. % nitrogen, therefore it is
possible nitrogen had the similar effect shown in Lothongkum’s work [196]. It
was not possible to detect the nitrogen content of either phase using the EDX
and austenite were not obtained. The PREN of the phases is a function of the
of duplex stainless steels, 2205 and 2507, under MgCl2 salt deposits exposed
to 80 °C and 40% RH for 7 days. They found that ferrite is more susceptible to
also found that heat treatment (at 750 °C for 3 hours followed by water
Grade 2205, while in Grade 2507, mainly primary and secondary austenite was
attacked.
a narrow range of potentials (Figure 2-9), with the active peak for ferrite being
slightly lower in potential for that of austenite [83], so it is possible that small
changes in potential can change the active phase under particular conditions. It
108
should be noted that Donohoe et al. [75] were conducting experiments in full
More recently, Örnek et al. [82] found that corrosion attack of ferrite and
exposure to 50°C and 30% RH for approximately one year. At the centre of the
droplet, both phases were found to be attacked while at the rim of the droplet,
austenite was found to be attacked. In between of these two regions, ferrite was
the preferential attack of ferrite was not a function of pit location within the
droplet, for example in Figure 4-9 the pit is near the centre of the droplet and in
Figure 4-10 the pit is at the edge of the droplet. This difference in results might
be due to the different material and experiment conditions such as salt solution
[36, 42, 67, 68]. This is consistent with the results obtained in the present work.
The preferential attack of ferrite bands seen in Figure 4-7(b) was not observed
after the solution annealing treatment (Figure 4-16 and Figure 4-17). Although
the volume fraction of ferrite was not quantified after solution annealing, VSM
109
results (Figure 4-3), also indicate a significant reduction in ferrite content in this
work.
4-19 shows that following solution annealing of the stainless steel at 1050 °C for
observed. Instead a faceted pit has formed. This implies that solution annealing
In full immersion condition, faceted pits have been previously reported on both
austenitic stainless steels and ferritic chromium steels [110, 125, 126]. Schwenk
[125] found regularly etched pits, mostly hexagons and squares (Figure 2-14),
containing Cl- at low applied potentials. Based on the examination of the etch
pits Schwenk [125] reported that pit bounding planes belong to the type {111},
having the lowest dissolution rate (the most closely packed crystal plane in the
case of austenite).
In this study, it was not possible to confirm the crystallographic orientation of the
110
4.4 Conclusion
propagated downwards into the steel, but also parallel to the rolling
direction along the ferrite bands. This has been shown using FIB.
to equiaxed etched faceted pits. The facets may be {111} planes. Such
111
5 Effect of relative humidity on atmospheric pitting
steels
5.1 Introduction
The atmospheric pitting corrosion of stainless steel under Cl- rich droplets is
concentration of the solutions that form on the metal surface depends on the
Pit morphologies have been found to be affected by relative humidity [6] and
microstructure [18, 26]. It is important to obtain large data sets in order to draw
pitting corrosion of commercial alloys. In this study and some recently published
studies [6, 134, 135, 198], large data sets were obtained.
The work presented in this chapter will explore the effect of relative humidity on
the pit morphology of atmospheric pitting corrosion of 304L and 316L austenitic
stainless steels. The effect of pit position as well as the exposure time on the pit
The plates were used “as received” (see Appendix 1) and did not go through
112
5.2 Results
Figure 5-1 Time sequence of optical images of a pit grown under a MgCl2
droplet (~1000 µg/cm2 CDD) on 304L stainless steel (LT plane) in ambient
conditions (RH 37±3%, temperature 24±2 °C).
~1000 µg/cm2 CDD MgCl2 in ambient conditions (RH 37±3, temp 24±2 °C) on
with a deeper hole in the centre (Figure 5-1(a)). The shallow attack continues to
grow radially for another hour (Figure 5-1(b)), with the shallow dish region
shallow enough to show work-hardening effects from grinding. The shallow dish
region stops growing radially after 1 hour. Instead, both “ear-shaped” attack [21]
into one edge of the shallow dish and localised “satellite” pitting [6] outside the
113
dish begin simultaneously (Figure 5-1(c)). Both ear-shaped attack, which
produces an “earing” morphology (Street et al. [6] call them “spiral” pits), and
satellite attack continue to grow at the same time (Figure 5-1(d) and (e)). After
15 hours, the earing attack has ceased to operate and the earing morphology
does not change with continued pitting (Figure 5-1(f)). However, the satellite
attack does continue to grow until the end of the experiment at 46 hours.
The shallow dish diameter was measured by using the FIJI/ImageJ software,
observed in the shallow dish diameter after the first hour of exposure. After
1 hour (Figure 5-1(b)), the shallow dish diameter was found to be about 117 µm.
After 46 hours (Figure 5-1(f)) it was found to be 120 µm. This is consistent with
the results observed by Street et al. [6], who carried out the time-lapse
~750 µg/cm2) . They reported that the shallow dish diameter had stopped
increasing after 2 hours of exposure, but further etching was seen. Street
carried out the. They also observed that the diameter of the shallow dishes
values and at 1 day or 1 week of exposure (See Table 5-4 ) on both 304L and
114
5.2.2 Pit covers
Figure 5-2(a) shows satellite pitting attack found on 304L after 1 day at 35% RH
under a MgCl2 droplet with a CDD of ~1000 µg/cm2. The shallow dish is
surrounded by several satellite pits at different distances (Figure 5-2 b and c).
Each of these pits was found to have a cover as shown in the SEM image
(Figure 5-2(c)). The cover was found to have grinding marks that matched with
those on the metal surface Figure 5-2(d). Three pit covers were analysed with
EDX and found to be enriched in Cr, Mn, S, O and Cl. Table 5-1 shows the
chemical compositions of typical pit covers formed on 304L and 316L. Similar
pit covers were also found on pits after 1 week exposure at 45% RH.
In tomography experiments, pits were also observed to have covers. From the
vertical section in Figure 5-3, it can be seen that there is a bright line just above
the pit, consistent with the presence of a high density cover. The pit was
observed on 304L pin sample under MgCl2 droplet of CDD ~1000 µg/cm2 after
115
Figure 5-2 Satellite pitting attack found under a droplet of MgCl2 of CDD
~1000 µg/cm2 on 304L stainless steel (LT plane) after 1 day exposure at 35%
RH and 30 °C; (a) after washing with DI water; (b) magnified section of (a); (c)
SEM image of selected region of (b) showing the shallow dish region and the
satellite pits with caps; and (d) magnified image of a satellite pit showing the pit
cover.
Table 5-1 typical EDX analysis of pit typical covers observed on 304L and 316L
after 1 day exposure at 35% RH and 30 °C.
116
Figure 5-3 X-ray tomographic horizontal and vertical sections of a pit found on a
304L pin (2 mm diameter, ST plane) under an MgCl2 droplet (CDD
~1000 µg/cm2) after exposure at 45% RH for 31 hours at 21±1 ºC.
5.2.3 Effect of the pit position on the diameter of the shallow dish for
towards the centre or edge of the droplet. However, close inspection showed
that the diameter of the shallow dish region depends on the distance of the pit
from the edge of the droplet. The technique used to measure the shallow dish
117
5.2.3.1 Effect of the pit position on the diameter of the shallow dish for
Figure 5-4 Macro photo of 304L plate surface (LT plane) deposited with arrays
of MgCl2 droplets (~2.5 mm equivalent diameter with ~1000 µg/cm2 CDD at
deposition), after 1 day exposure at 35% RH and 30 °C.
Figure 5-4 shows a 304L stainless steel plate after pitting at 35% RH for 1 day
under droplets of MgCl2 with ~1000 µg/cm2 CDD. 66 out of 70 droplets pitted.
in [6]. Typical examples of these morphologies are shown in Figure 5-5. In all
cases attack appears to start with a shallow dish region that develops into
deeper attack at the dish’s edge. The satellite morphology is generally found for
pits close to the droplet edge, and shows several small pits that mostly develop
around the perimeter of the shallow dish region, with some small pits initiating
inside the shallow dish and some far away in a few instances. The earing
morphology is generally found for pits closer to the centre of the droplet, and
always shows an ear-shaped attack at one side of the shallow dish with no
appearance of any small pits around or close to the perimeter of the dish region.
118
Figure 5-5 Typical pit morphologies formed on 304L stainless steel (LT plane)
under MgCl2 droplets (of ~2.5 mm equivalent diameter with ~1000 µg/cm2 CDD
at deposition), after 1 day exposure at 35% RH and 30 °C; (a) near the droplet
edge; (b) near the droplet centre.
Propagation of satellite pits were often seen to be influenced by the local crystal
structure and orientation. Figure 5-6 shows pits that have both earing attack and
satellite pits, and it is the satellite pits that have crystallographic facets.
119
Figure 5-6 Typical pit morphologies of satellite and earing pits affected by the
local crystal structure of 304L stainless steel (LT plane) under MgCl2 droplets of
equivalent diameter of ~2.5 mm with ~1000 µg/cm2 CDD at deposition after 1
day exposure at 35% RH and at 30 °C.
35% RH at ~1000 µg/cm2 CDD. All 70 droplets pitted. 21 of the pits were found
to be associated with micro-droplets. SEM of the pits (Figure 5-8) show that the
satellite pits continued to grow over one week compared to 1 day exposure
(Figure 5-5). Where more than one satellite pit continued to grow, these were
often similar in size. Earing morphology was also observed along with deeper
attack. The continued material loss after 1 week meant that in 13 out of 70
cases, the shallow dish region was heavily attacked and was not possible to
120
Figure 5-7 Macro photo of 304L plate surface (LT plane) with arrays of MgCl2
droplets of ~2.5 mm equivalent diameter with ~1000 µg/cm2 CDD at deposition,
after 1 week exposure at 35% RH and 30 °C.
Figure 5-8 Typical pit morphologies formed on 304L stainless steel (LT plane)
under MgCl2 droplets (~2.5 mm equivalent diameter with ~1000 µg/cm2 CDD at
deposition), after 1 week exposure at 35% RH and 30 °C; (a) near the droplet
edge; (b) near the droplet centre.
121
Figure 5-9 Typical pit morphology formed on 304L stainless steel (LT plane)
under MgCl2 droplets (~2.5 mm equivalent diameter with ~1000 µg/cm2 CDD at
deposition), after 1 week exposure at 35% RH and 30 °C; (a) droplet with a pit
near the droplet edge; (b) SEM of pit with heavily attacked shallow dish region.
The position of the initiation site inside the droplet affects the morphology that
develops during corrosion. Figure 5-10 shows that the occurrence of satellite
pitting morphology is increased when pits initiate towards the edge of the
droplets, and earing morphology is more common towards the centre of the
droplets for both 1 day (Figure 5-5) and 1 week (Figure 5-8) exposure tests. The
size of the shallow dish region is also affected, and is usually smaller towards
the edge of the droplet and larger towards the centre for both 1 day and 1 week
exposure tests. Where the shallow dish could be seen after 1 week exposure, it
was of a similar size to those seen after 1 day of exposure (Figure 5-10),
confirms the idea that the shallow dishes stop growing at an early stage
(~1hour).
122
Figure 5-10 Shallow dish equivalent diameter versus the distance from the
droplet edge in MgCl2 droplets (~2.5 mm equivalent diameter, ~1000 µg/cm2
CDD at deposition) on 304L (LT plane) exposed at 35% RH and 30 °C. 1 day
exposure is marked with red circles and squares while 1 week exposure is
marked with black circles and squares. The equivalent diameter is calculated as
the diameter of a circle with the same area as the dish.
123
5.2.3.2 Effect of the pit position on the diameter of the shallow dish for
Figure 5-11 Macro photo of 316L plate surface (LT plane) after 1 week
exposure at 35% RH and 30 °C under MgCl2 droplets (~2.5 mm equivalent
diameter, CDD of ~1000 µg/cm2 at deposition); plate has been washed with DI
water.
Figure 5-11 shows a 316L plate after exposure for droplets of MgCl2 with a CDD
pits showing earing morphology (usually formed towards the centre of the
droplet) and 22 pits showing satellite pitting formed towards the edge of the
droplet. Typical examples of both these morphologies are shown in Figure 5-12.
The satellite morphology shows an attack at one edge of the shallow dish
region in addition to the formation of small pits at the perimeter of the dish
region. In some cases the small pits were found at a distance close to the dish
region. While the earing morphology only shows an ear-shaped attack at one
124
As described above for 304L at the same RH, the shallow dishes have the trend
of formation of smaller diameters towards the edge of the droplet and larger
diameters towards the centre of the droplet. The shallow dishes were
Among the total number of pits (52 pits), 24 of them were found to be
Figure 5-12 Typical pit morphology formed on 316L plate (LT plane) at 35% RH
under MgCl2 droplets (~2.5 mm equivalent diameter with ~1000 µg/cm2 CDD at
deposition), after 1 week exposure at 30 °C; (a) near the droplet edge; (b) near
the droplet centre.
125
Figure 5-13 Shallow dish equivalent diameter as a function of distance from the
droplet edge for MgCl2 droplets (~2.5 mm equivalent diameter, ~1000 µg/cm2
CDD at deposition) on 304L and 316L (LT plane) exposed at 30 °C and
35% RH for 1 week. The equivalent diameter is calculated as the diameter of a
circle with the same area as the dish.
126
5.2.3.3 Effect of the pit position on the diameter of the shallow dish for
Figure 5-14 Macro photo of 304L plate surface (LT plane) with arrays of MgCl2
droplets (~2.5 mm equivalent diameter, ~1000 µg/cm2 CDD at deposition) after
1 day exposure at 45% RH and 30 °C.
Figure 5-14 shows results of pitting 304L stainless steel at 45% RH after 1 day
for droplets of MgCl2 with ~1000 µg/cm2 CDD. 74 out of 75 droplets pitted. 36
droplets appear to have crystallised, but this happened during photography after
the experiment was completed. The monitored lab RH was 25% therefore it is
expected MgCl2 salt will crystallise because the deliquescence relative humidity
(DRH) of MgCl2 is 33% below this value the salt will crystallise. 39 of the pits
previous sections, a similar trend was observed between pit position and
morphology as at 35% RH, with more ear-shaped pits occurring towards the
centre of the droplets, and more satellite pits occurring towards the edge
127
(Figure 5-16). Shallow dish regions were also larger towards the centre of
droplets.
Figure 5-15 Typical pit morphologies formed on 304L stainless steel (LT plane)
under MgCl2 droplets (~2.5 mm equivalent diameter, ~1000 µg/cm2 CDD at
deposition), after 1 day exposure at 45% RH and at 30 °C; (a) near the droplet
edge; (b) near the droplet centre.
128
Figure 5-16 The shallow dish equivalent diameter versus the distance from the
droplet edge in MgCl2 droplets (~2.5 mm equivalent diameter, ~1000µg/cm2
CDD at deposition) on 304L (LT plane) after 1 day of exposure at 45% RH at
30 °C. The equivalent diameter is calculated as the diameter of a circle with the
same area as the dish.
5.2.3.4 Effect of pit position on the diameter of the shallow dish for 304L
Figure 5-17 shows results of pitting 304L stainless steel at 48% RH after 1 day
for droplets of MgCl2 with ~1000 µg/cm2 CDD. 64 out of 75 droplets pitted, with
all droplets containing only one main pit. 25 out of 64 pits were found to be
(Figure 5-18), satellite pits were not observed. A small hole can be seen in the
centre of the shallow dish region (Figure 5-18(a)), which is possibly the cavity
129
that developed from the dissolving of an inclusion during the pit initiation
process. Pits formed at the droplet edge were usually with one side significantly
more attacked (Figure 5-19), leaving a small trace of the shallow dish. The trend
in shallow dish region diameter is similar to that seen at 35% and 45% RH, with
Figure 5-17 Macro photo of 304L plate surface (LT plane),after washing with DI
water, with MgCl2 droplets (~2.5 mm equivalent diameter, ~1000 µg/cm2 CDD
at deposition) after 1 day exposure at 48% RH and 30 °C.
130
Figure 5-18 Typical pit morphology formed on 304L (LT plane) under MgCl2
droplets (~2.5 mm equivalent diameter, ~1000 µg/cm2 CDD at deposition), after
1 day exposure at 48% RH at 30 °C; (a) near the droplet edge; (b) near the
droplet centre.
Figure 5-19 Typical pit morphology, at the droplet edge, formed on 304L (LT
plane) under MgCl2 droplets (~2.5 mm equivalent diameter, ~1000 µg/cm2
CDD at deposition), after 1 day exposure at 48% RH at 30 °C; (a) optical image
of the droplet after washing with DI water; (b) SEM of pit.
131
Figure 5-20 Shallow dish equivalent diameter versus the distance from the
droplet edge in MgCl2 droplets (~2.5 mm equivalent diameter, CDD of
~1000 µg/cm2 at deposition) on 304L stainless steel (LT plane) after 1 day of
exposure at 48% RH at 30 °C. The equivalent diameter is calculated as the
diameter of a circle with the same area as the dish.
Table 5-2 summarises the results of the average shallow dish diameters as a
function RH and exposure time. It can be seen that the shallow dish diameter
does not change with RH and exposure time, and alloy type when comparing
304L and 316L after 1 week of exposure. For 316L at 35% RH after 1 day
1 day exposure.
132
Table 5-2 average diameters with standard deviations of shallow dish diameters
formed on 304L and 316L (LT plane) under MgCl2 droplets (~2.5 mm equivalent
diameter, CDD of ~1000 µg/cm2 at deposition) after exposure at 30 °C. The
equivalent diameter is calculated as the diameter of a circle with the same area
as the dish.
5.2.4 Effect of pit position on pit diameter and depth for 316L stainless
steel at 48% RH
Figure 5-21 shows results of pitting 316L stainless steel, LT plane, at 48% RH
after 1 day exposure for droplets of MgCl2 with ~1000 µg/cm2 CDD. 60 out of 65
droplets pitted and in all cases, only one main pit was observed. 19 out of 60
morphology with a rough interior with no obvious shallow dish (Figure 5-22).
Only two droplets had pits that seemed to be growing along the rolling direction
133
Figure 5-21 Macro photo of 316L plate surface (LT plane) with arrays of MgCl2
droplets (~2.5 mm equivalent diameter, ~1000 µg/cm2 CDD at deposition) after
1 day exposure at 48% RH and 30 °C.
Figure 5-22 Typical pit morphology formed on 316L (LT plane) at 48% RH
under MgCl2 droplets (~2.5 mm equivalent diameter, ~1000 µg/cm2 CDD at
deposition) after 1 day exposure at 30 °C; (a) near the droplet edge; (b) near
the droplet centre.
134
Figure 5-23 Morphology of a pit growing along the rolling direction on 316L (LT
plane) surface under MgCl2 droplet (~2.5 mm equivalent diameter, ~1000
µg/cm2 CDD at deposition), after 1 day exposure at 48% RH at 30 °C; (a)
optical image of droplet; (b) SEM of the pit.
Figure 5-24 shows the pit diameter and depth as a function of distance for pits
relatively larger diameters are found towards the centre of the droplet. Pits close
to the edge (<200 µm from the droplet edge) had an average pit diameter of
80±6 µm while pits >200 µm from the droplet edge had an average diameter of
90±8 µm. The equivalent pit diameter was obtained from Equation 3-2 based on
the assumption that the pit mouth area is circular; the procedure was detailed in
Section 3.8.2. Pit depth, with an average of 45±10 µm, was invariant with
distance. The fine focus method used to measure the pits depth was described
in Section 3.8.3.
135
Figure 5-24 Equivalent pit diameter and depth as a function of distance for pits
shown in Figure 5-21 formed on 316L stainless steel (LT plane) in MgCl2
droplets (~2.5 mm equivalent diameter, ~1000 µg/cm2 CDD at deposition)
following exposure at 48% RH for 1 day at 30 °C.
5.2.5 Effect of pit position on pit diameter and depth for 304L stainless
steel at 56% RH
Comparisons were made between pitting at 56% RH after 1 day (Figure 5-25)
and 1 week (Figure 5-26) for droplets of MgCl2 with ~1000 µg/cm2 CDD. After
observed in any of the pits in either data set. Pits showed neither earing nor
136
satellite morphologies. They instead showed circular pit morphology after both 1
Figure 5-25 Macro photo of 304L plate surface (LT plane) deposited with arrays
of MgCl2 droplets (~2.5 mm equivalent diameter, ~1000 µg/cm2 CDD at
deposition), after 1 day exposure at 56% RH and 30 °C.
Figure 5-26 Macro photo of 304L plate surface (LT plane) deposited with arrays
of MgCl2 droplets (~2.5 mm equivalent diameter, ~1000 µg/cm2 CDD at
deposition), after 1 week exposure at 56% RH and 30 °C.
Figure 5-27 shows pit morphologies formed on 304L at 56% RH after 1 day
exposure. The pit morphology was affected by the position of pits in the droplet.
Pits near the centre of the droplet (Figure 5-27(a)) and near to the droplet edge
(Figure 5-27(b)) showed a hemispherical attack with approximately the same pit
137
mouth diameter. Pits formed at the edge of the droplet had larger pit mouths
than those grown towards the centre. 15 out of 70 pits occurred at the edge of
the droplet, with 14 growing fully inside the droplet showing a deep attack into
the pit towards the droplet centre (e.g. Figure 5-27(c)). Only 1 pit found to be
Figure 5-27 Typical pit morphologies formed on 304L (LT plane) in MgCl2
droplets (~2.5 mm equivalent diameter, ~1000 µg/cm2 CDD at deposition), after
1 day exposure at 56% RH at 30 °C; (a) near the droplet centre; (b) near the
droplet edge; (c) at the droplet edge; (d) bisected by droplet edge.
After 1 week at 56% RH, the position of the pits in the droplet affects the
morphology which can be categorised in 4 ways (Figure 5-28). Pits near the
centre of the droplet (Figure 5-28 (a)) shows a deep, hemispherical attack into
the metal. When pits occur closer to the edge (Figure 5-28 (b)) the attack
appears to be less hemispherical and more layered, although the mouth of the
pit is approximately the same size as that for pits found at the centre.
138
11 out of 60 pits occurred at the edge of the droplet, with 6 growing fully inside
the droplet (Figure 5-28 (c)) and 5 pits being bisected by the droplet (Figure
5-28 (d)). Pits at the edge of the droplet have larger pit mouths than those
grown towards the centre (Figure 5-29). When the pit grows fully in the droplet,
the deep attack into the pit always occurs towards the droplet centre (Figure
5-28 (c)). However, when pit is bisected by the droplet edge, 3 out of 5 were
seen to have deep pitting in the part of the pit that has a very thin solution layer,
outside the bulk of the droplet (Figure 5-28 (d)). This was not clear for the other
2 pits.
Figure 5-28 Typical pit morphologies formed on 304L (LT plane) in MgCl2
droplets (~2.5 mm equivalent diameter, ~1000 µg/cm2 CDD at deposition), after
1 week exposure at 56% RH at 30 °C; (a) near the droplet centre; (b) near the
droplet edge; (c) at the droplet edge; (d) bisected by droplet edge.
139
The pit mouth diameter was generally larger at the edge of the droplet than
towards the centre in both one day and one week experiments (Figure 5-29),
with an average diameter of 89±15 µm for one day and 98±12 µm for one week.
There was no significant change in pit diameter when initiation occurred further
than 200 µm from the edge of the droplet in both 1 day (65±6 µm) and 1 week
(68±7 µm) experiments. The procedure used to measure the pits diameter was
Although there was very little variation in pit diameter between 1 day and 1
week experiments, there was significant variation in pit depth (Figure 5-30).
Within 200 µm of the droplet edge, the average pit depth after 1 day was
50±9 µm and after 1 week was 75±11 µm. For pits >200 µm from the droplet
edge, the average depth was 47±8 µm after one day and 61±9 µm after 1 week
(Table 5-3). The fine focus method was used to measure the pits depth as
140
Figure 5-29 Equivalent pit diameter as a function of pit position for pits shown in
Figure 5-25 and Figure 5-26, formed on 304L stainless steel (LT plane)
deposited with arrays of MgCl2 droplets (~2.5 mm equivalent diameter, ~1000
µg/cm2 CDD at deposition) following exposure at 56% RH and 30 °C for 1 day
(red circles) and 1 week (black squares). The equivalent diameter is calculated
as the diameter of a circle with the same area as the pit mouth.
141
Figure 5-30 The pit depth as a function of pit position for pits shown in Figure
5-25 and Figure 5-26, formed on 304L stainless steel (LT plane) deposited with
arrays of MgCl2 droplets (~2.5 mm equivalent diameter, ~1000 µg/cm2 CDD at
deposition), following exposure at 56% RH and 30 °C for 1 day (black
diamonds) and 1 week (red squares).
Table 5-3 The pit diameter and depth as a function of exposure time and
distance from the droplet edge. Pits formed on 304L stainless steel (LT plane)
under MgCl2 droplets (~2.5 mm equivalent diameter, ~1000 µg/cm2 CDD at
deposition), following exposure at 30 °C.
142
At 56% RH, it was often found that more than one pitting site was present under
a droplet. Examples of these are shown in (Figure 5-31). For droplets containing
two or more pits after 1 week exposure (lower row images in Figure 5-31) the
rust deposited by the smaller pits, which tended to initiate at the edge of the
droplet, overlays the rust formed by the larger pit, however the rust from the
smaller pits is not affected by the rust from the larger pit, suggesting that the
smaller pit(s) occurred after the main pit had stopped producing rust. Therefore
Figure 5-31 Typical examples of droplets containing 1 pit, 2 pits, and 3 or more
pits at 56% RH, 30 °C of ~1000 µg/cm2 CDD MgCl2 droplets on 304L stainless
steel (LT plane) after exposure for 1 day (upper optical images) and 1 week
(lower optical images).
143
Figure 5-32 shows the number of droplets containing no pits, 1 pit, 2 pits, and 3
or more for both 1 day and 1 week exposure at 56% RH. After 1 day, 4 out of 70
(6%) droplets had no pit, 27 out of 70 (39%) droplets had only one pit, while 36
out of 70 (51%) droplets had 2 pitting sites, and 3 (4%) droplets had 3 or more.
After 1 week, the number of droplets with a single pitting site was much lower,
with 9 out of 60 (15%) droplets having only one pit, 20 (33%) droplets having 2
pitting sites and 28 (47%) droplets having 3 or more pitting sites, 3 out of 70
Figure 5-32 Number of pits per droplet of MgCl2 droplets (~2.5 mm diameter,
~1000 µg/cm2 CDD at deposition) for pits shown in Figure 5-25 and Figure 5-26
on 304L stainless steel plate, following exposure at 56% RH and 30 °C for 1
day (in blue) and 1 week (in brown).
144
5.2.6 Atmospheric pitting corrosion behaviour of 316L stainless steel at
56% RH
Pitting of 316L stainless steel at 56% RH after 1 day and 1 week exposure are
shown in (Figure 5-33) and (Figure 5-34) respectively, under droplets of MgCl2
with ~1000 µg/cm2 CDD. After 1 day, only 2 out of 65 droplets pitted (1 circular
pit and 1 elongated pit) while after 1 week, 9 out of 60 droplets pitted (all
Circular pits with a rough interior with possible grain facets morphology that
correlates with the grain size of the alloy shown in Figure 3-3, without any
obvious shallow dish region, were observed after both 1 day and 1 week
exposure (Figure 5-35). It can be seen that there is a significant increase in pit
give a range of accuracy) for 1 day but 110±10 µm for 1 week exposure. The pit
depth was 75±5 µm after 1 week. Single pits per droplet were observed after 1
day of exposure while after 1 week both single pit and multiple pits (2, 3 or
more) were observed. The effect of pit position on pit diameter and depth is not
considered in these two data sets because of the small number of pitted
droplets compared with those formed on 304L at the same RH. However,
examination of individual optical micro images showed that most of the pits
145
Figure 5-33 Macro photo of 316L (LT plane) with MgCl2 droplets (~2.5 mm
equivalent diameter, ~1000 µg/cm2 CDD at deposition), following exposure at
56% RH for 1 day at 30 °C; pitted droplets are selected with dashed square.
Figure 5-34 Macro photo of 316L plate surface (LT plane) with MgCl2 droplets
(~2.5 mm equivalent diameter, ~1000 µg/cm2 CDD at deposition), following
exposure at 56% RH for 1 week at 30 °C.
146
Figure 5-35 Typical pit morphologies formed on 316L stainless steel (LT plane)
under MgCl2 droplets (~2.5 mm equivalent diameter, ~1000 µg/cm2 CDD at
deposition), after exposure at 56% RH and 30 °C; (a) after 1 day; (b) after 1
week.
304L and 316L plates were exposed to 65% RH for 1 day for droplets of MgCl 2
with ~1000 µg/cm2 CDD. Figure 5-36 shows results for 304L, where 7 out of 75
droplets pitted. All droplets contained only one main pit. All pits show a circular
pitting morphology and all pits were bisected by the edge of the droplet with an
average pit diameter of 100±20 µm. All pits were found to be associated with
micro-droplets. At this RH, no pits were found on the 316L plate. At 70% RH
and above there was also no pitting for either 304L or 316L. Table 5-4
summarises the results of the tests carried out on both 304L and 316L stainless
steel at the two exposure times at all RH values used in the present work.
147
Figure 5-36 (a) macro photo of 304L plate surface (LT plane) with MgCl2
droplets (~2.5 mm equivalent diameter, ~1000 µg/cm2 CDD at deposition), after
1 day exposure at 65% RH and 30 °C; (b) optical microscope images of pits
position relative to droplet edge; (c) SEM of pits.
148
Table 5-4 Summary of the results of the pitting behaviour at different RH and
exposure time for 304L and 316L stainless steels (LT plane)under MgCl2
droplets (~2.5 mm equivalent diameter, ~1000 µg/cm2 CDD at deposition). The
pit morphologies and the number of droplets pitted also the number of pits
associated with micro-droplets (shown in red) at each RH and exposure time for
both materials are presented.
149
5.3 Discussion
both 304L and 316L stainless steels (Figure 5-5, Figure 5-8, Figure 5-12, Figure
5-15, Figure 5-18), which has been previously reported [5, 6, 14, 21]. This is the
initial stage of attack, confirmed by time-lapse images (Figure 5-1). Once the
shallow dish has reached its full size, radial growth stops and other forms of
attack continue. Street et al. [6] suggest that this region is caused by passive
current dissolution due to the high passive current densities sustained in highly
concentrated MgCl2 solutions: the volume of material loss in the shallow dish
Maier and Frankel [21] proposed that the dish “radial” type of attack is caused
by the adsorption of sulphur species on the oxide layer. Hastuty et al. [5] have
hydrolysis caused by metal cation release drives growth radially away from the
sulphur-rich initiation site. This growth continues until current density is so low
the surface passivates. Recently, however, Street et al. [6] suggested that the
presence of sulphur species could not be the only controlling factor based on
the observation of the variation of the shallow dish diameter with distance from
150
In agreement with Street’s observation, the diameter of the shallow dish region
is found to be affected by position of the pit inside the droplet, with diameters
being generally smaller towards the edge and larger towards the centre of the
droplet. This is the case for 35% RH (Figure 5-10 and Figure 5-13), 45% RH
(Figure 5-16) and 48% RH (Figure 5-20), and there is no significant difference in
work, this was attributed to the IR drop increase towards the edge of the
droplet, reducing passive current at the initiation site which results in smaller
dish sizes.
Section 6.2.1, it was found that the shallow dish diameter as well as the pit
diameter was increased with increasing the droplet diameter. As the area of
electrolyte coverage affects the available cathodic area (with larger droplets
having more cathodic area, and so more available cathodic current) this
supports the idea that growth size of the shallow dish is strongly influenced by
cathodic area.
At 35% and 45% RH, two pit morphologies dominated after 1 day: “satellite”
pits, with several small pits surrounding the shallow dish region, and “earing”
pits, with increased growth of part of the shallow dish region, which may over
time extend beyond it. At 48% RH, only earing pits were seen for 304L (Figure
5-18), while for 316L almost circular pits were seen (Figure 5-22). Satellite
by the thickness of the droplet above the pit, which influences both the
151
and the rate of diffusion of ions away from the pit. Low diffusion allows metal
This agrees with observations that less satellite pitting occurs at higher RH,
where solution droplet thicknesses are larger and diffusivity is higher [6].
Satellite pits are covered with caps (Figure 5-2) enriched in Cr, Mn, S, O and Cl.
analysis) of pit covers obtained on 304L and 316L (Table 5-1). Pit covers are
also observed in tomography experiments (Figure 5-3), with a bright line just
above the pit, suggesting the metallic composition of the pit cover. Similar pit
covers were reported previously by Guo [7] on 304L stainless steel under MgCl2
droplets (CDD ~1000 µg/cm2) after wet-dry cycling exposure test examined in
The earing pits seen at 35%-48% RH were formed by ear-shaped attack into
the shallow dish region, as described by Maier and Frankel [21]. They attribute
this to the gradual passivation of the shallow dish region, concentrating current
in a smaller area. This earing attack is also seen to occur in the current work on
pits with satellite morphologies but is significantly smaller (Figure 5-1). This
indicates that both the satellite pitting and the earing attack start at the same
time (see time-lapse in Figure 5-1), with the satellite pitting eventually
dominating at lower RH values. It should be noted that Maier and Frankel did
not see satellite pitting although they used an RH of 33–34%, which may be due
to the different surface grinding (4000 grit instead of 800 grit). Preliminary
unpublished work carried out in the current project suggests that satellite pits
152
may not form on mirror-polished surfaces. This would remove occluded initiation
sites, allowing the earing mechanism to continue into the metal and form pits
For 304L exposed at 56% RH for 1 day (Figure 5-27) and 1 week exposure
(Figure 5-28) and 316L exposed to the same RH and exposure times (Figure
5-35), circular pits with no shallow dish region were observed, with radial growth
of the pit proportional to growth into the metal, resulting in smaller diameter pits.
This is in agreement with Street et al. [6], who suggested that shallow dish
regions could be caused by passive current dissolution (Figure 5-37 [18]). It can
density, with the only major difference being a reduction in breakdown potential
153
Figure 5-37 Current against potential polarisation curves for 250 µm diameter
304 stainless steel wires in MgCl2 solutions with concentrations of: (a) 0.05–
3 M; and (b) 3–5 M. The potential sweep rate was 0.2 mV s-1 [18].
The circular morphology is also seen at 65% RH (Figure 5-36) for 304L, with all
the pits forming at the droplet edge (bisected by the droplet edge). A possible
mechanism for this could be due to the concentrations needed for initiation to
occur. Recent work by Nam et al. [9] show that MgCl2 droplets on 304L require
under the droplets. In the current work pitting was also seen to rely on chloride
and 70% RH for 304L. The difference between this work and that of Nam et al.
is likely due to the conditions of exposure. Whereas the current work used a
fixed humidity for 1 day exposure and 800 grit surface finish, Nam et al.
154
When pitting occurs in 65% RH in the present work, initiation may be occurring
just outside the edge of the droplet, under monolayers of water. These very thin
layers would limit diffusion away from initiation sites and allow aggressive
Also, the initiation at the edge allows access to a strong cathodic region
supplied by the droplet, driving the corrosion process. As such, these pits look
“bisected” by the droplet edge, as pit growth can continue radially from the
initiation site. This explains the relative rarity of these pitting events (7 out of 75)
as specific conditions would need to be met for this to occur i.e. a susceptible
In the present study, for the pits found on 304L at 56% RH, it is clearly observed
that the diameter and depth of pits formed at or close to the droplet edge were
greater than those of pits close to the droplet centre (Figure 5-29) and (Figure
5-30). This can be attributed to the presence of a three phase boundary, where
there is particularly easy access of oxygen for the cathodic reaction [23, 199-
201]. Morton and Frankel [167] found that large pits only occurred at or very
close to the droplet edge in droplets deposited on aluminium. They carried out
diameter) with and without inhibitors and exposed at 85% RH (using saturated
KCl) for 20 hours. Morton and Frankel proposed that preferential pitting
generated close to the droplet edge will result in the formation of an acidic zone
155
that develops more pit initiation and growth for aluminium. A similar mechanism
Pits formed at the droplet edge were always surrounded by secondary droplets
(this is shown in Section 6.4.4), and it is more likely that the area covered by the
therefore wider and deeper pits were seen. Using SKP, Morton and Frankel
[167] observed that the outer edge of the secondary droplet and a wide area
aluminium.
While the pit depth increases between one day and one week, pit mouth
diameter does not (Table 5-3). This can be attributed to pit depth being under
only circular pits were found at 48% RH for 316L (Figure 5-24), pits were found
to be smaller in diameter close to the droplet edge than those formed towards
the centre. This is likely to be associated to the increased IR drop under thinner
layers since the water activity in the air and water phases is equal at equilibrium
156
droplet, leading to a decrease in the probability of occurrence of pitting
corrosion [4].
For 304L, The critical relative humidity for occurrence of atmospheric pitting
Pits were not observed at 70% RH and higher (Table 5-4), while for 316L, pits
were observed up to 56% RH (Figure 5-35), but were not observed at 65% RH
and higher (Table 5-4 ). The findings for 304L are in agreement with the results
corrosion of 304 stainless steel beneath droplets of MgCl2 solution and reported
that the critical relative humidity for occurrence of pitting corrosion lies between
For 316L, the critical relative humidity obtained in this study was in the range of
Padovani et al. [20]. These authors found pitting in MgCl2 solution droplets up to
concentration (Figure 5-37). For 304L, it has been shown that in de-aerated full
157
humidities close to deliquescence. Near saturation, passive current densities of
It is important to bear in mind that all the experiments carried out in this chapter
were performed on as received 304L and 316L alloys surface (LT plane). Pits
were often seen to be influenced by the local crystal structure. Pits were often
seen to have a small hole in the centre of the shallow dish region e.g. (Figure
5-1) and (Figure 5-18(a)), which is possibly the cavity that developed from the
Pits can initiate and propagate along pre-existing defects, such as elongated
inclusions [7, 18]. Figure 5-23 shows a pit grown on the alloy surface parallel to
the rolling direction of the alloy. This is most likely to be corrosion attack along
However, it might also be due to a surface defect present at the alloy surface. In
this chapter, among all the investigated pits, elongated pits were only observed
Pits can also propagate along residual delta ferrite [26]. Pits with layered
morphology were observed on the alloy surface of 304L (e.g. Fig.5-8 and 5-36).
This was due to the preferential attack of ferrite phase, as described previously
in Chapter 4.
158
Crystallographic facets were also observed in pits for the as received 304L
stainless steel (e.g. Figure 5-6). This is most likely due to the dissolution of
unattacked flat walls might be belong to the {111} close-packed crystal planes
morphology was also observed on 316L stainless steel exposed for both 1 day,
and 1 week exposure (Figure 5-35). The facet size of a given grain correlates
was found that the dissolution depth, after a given time, depends on the crystal
159
5.4 Conclusions
values ≤48%, in each droplet, the first corrosion attack is in the form of a
At lower RH values (35% and 45%), the shallow dishes develop small
satellite pits. Such pits are covered with caps enriched in Cr, Mn, S, O
and Cl. At 48% RH, for 304L only earing pit morphologies were observed
(no satellite pits were seen) as a result of continuous attack into the
shallow region while for 316L circular pits were observed. Shallow dishes
were not observed at ≥56% RH for 304L but for 316L were not observed
at ≥48% RH, and instead circular, deeper and rougher pits were seen.
2. For both 35% RH and 45% RH, a greater number of pits towards the
centre of the droplet, where the droplet height is higher, form an earing
attack while pits nearby the droplet edge, where the droplet height is low,
3. For 304L at 56% RH, the diameter and depth of pits formed at or nearby
the droplet edge was greater than that of pits close to the droplet centre.
160
This can be explained by the influence of three phase boundary where
4. Pit size significantly increased with increasing exposure time, while the
shallow dish diameter, after reaching a certain size, does not change
with increasing exposure time and RH, and was in same range for both
alloys.
5. For 304L, pits were observed up to ~65% RH (pits were not observed at
70% RH and higher), while for 316L, pit were observed up to ~56% RH,
161
6 Effect of droplet diameter and chloride deposition
6.1 Introduction
Both the size and deposition density of salt deposits can affect the corrosion
2.10.4 and 2.10.3. The salt concentration of such deposits is determined by the
relative humidity to which they are exposed [6, 141]. The two most common
methods of salt deposition on the metal surfaces for lab exposure experiments
recent work [6], it is shown that an automatic deposition system can be used to
different salt solutions and materials as described in Section 2.13. However, the
observed for MgCl2 droplets on stainless steel. In this work, these have been
162
6.2 Results
The effect of droplet diameter on pit diameter was investigated for both 304L
and 316L austenitic stainless steels; the results for each will be described
Figure 6-1 shows as deposited arrays of MgCl2 droplets of the same chloride
droplet diameter, for fixed chloride deposition density, on pit diameter and
depth.
Figure 6-1 Macrograph image of a 304L stainless steel plate surface (LT plane)
with as-deposited arrays of MgCl2 salt solution droplets of constant CDD
(~1000 µg/cm2) with various droplet diameters (“Dia”) and the correspond
droplet volumes (“Vol.”)
163
Figure 6-2 (a) a 304L stainless steel plate surface (LT plane) deposited with
arrays of MgCl2 droplets of fixed CDD (~1000 µg/cm2) shown in (Figure 6-1)
following exposure at 43% RH and 30 °C for 1 week; (b) typical droplets after
washing with DI water; (c) micrographs of pits shown in (b); (d) SEM images of
pits shown in (c).
Figure 6-2 (a) shows the macrograph of the 304L stainless steel plate surface
deposited with arrays of MgCl2 droplets shown in Figure 6-1 following exposure
at 43% RH and 30 °C for 1 week. It can be seen from the macrograph that the
droplets are still present and evidence of rust can be seen in all droplets. A
single pit per droplet can be observed. Only one droplet of the ~2.2 mm
164
diameter set contained an elongated pit that had grown along the rolling
direction of the alloy (not shown); all other pits were circular. The rust (corrosion
products) was washed away with deionised (DI) water (Figure 6-2 (b)) to
observe the pit morphologies shown in the lower optical micrographs (c). The
SEM images show more clearly that all the pits observed had a shallow dish-
shaped region with one side significantly more attacked (Figure 6-2 (d)).
Figure 6-3 shows an array of droplets designed to study the influence of droplet
diameter on both pit diameter and depth for deposits of constant chloride
Figure 6-3 Macrograph image of a 316L stainless steel plate surface (LT plane)
deposited with arrays of MgCl2 droplets of same chloride deposition density
(~1000 µg/cm2) with various droplet diameters.
165
Figure 6-4 (a) a macrograph of the 316L stainless steel plate surface (LT plane)
deposited with arrays of MgCl2 droplets of fixed CDD (~1000 µg/cm2) shown in
Figure 6-3 following exposure at 43% RH and 30 °C for 1 week; (b) typical
droplets after washing with DI water; (c) micrographs of pits shown in (b);
(d) SEM images of pits shown in (c).
Figure 6-4 (a) shows the macrograph of the 316L stainless steel plate surface
deposited with arrays of MgCl2 droplets of fixed CDD (~1000 µg/cm2) shown in
Figure 6-3 following exposure at 43% RH and 30 °C for 1 week. It can be seen
from the macrograph that droplets are still visible and evidence of rust can be
166
seen in all pitted droplets, which contained only one main pit. A significant
number of droplets did not show any sign of corrosion, in contrast to the results
for 304L stainless steel, where all droplets pitted. To reveal the pit morphologies
shown in Figure 6-4 (c), the rust was washed away with deionised water (Figure
6-4 (b)). In all pits, a shallow dish-shaped region with one side significantly more
attacked was observed more clearly with SEM images (Figure 6-4 (d)). In some
cases no shallow dish region can be observed, this may be because the entire
Figure 6-5 shows the relationship between the equivalent pit diameter (active
region only, as described in Section 3.8.2) and the deposited droplet diameter
for both 304L and 316L stainless steels. It can be seen that the pit diameter
chloride deposition density for both 304L and 316L stainless steels. It is also
evident that pits with larger diameters are formed on 304L than those that
formed on 316L stainless steel. The pit depth in 304L and 316L did not vary
alloys. The variation in pit depth in 316L appeared to be greater than the
variation in 304L. The fine focus method used to measure the pits depth and the
167
Figure 6-5 The equivalent pit diameter (active region without the shallow dish)
as a function of the droplet diameter for pits shown in Figure 6-2 and Figure 6-4;
the plates were deposited with arrays of MgCl2 droplets of fixed CDD
(~1000 µg/cm2) and exposed at 43% RH and 30 °C for 1 week. 304L is shown
in red circles and 316L is shown with blue squares, and solid symbols represent
pit diameter whereas open symbols represent pit depth.
The effect of chloride deposition density (CDD) for fixed droplet diameter on pit
diameter and depth was investigated for both 304L and 316L austenitic
stainless steels. The effect of droplet diameter and exposure time on pitting
168
Figure 6-6 shows an array of droplets on 304L stainless steel designed to study
the influence of chloride deposition density on pit size for droplets of constant
solution concentration, from 3.7 M to 185 µM MgCl2. This gave a range of CDD
Figure 6-6 Macrograph image of a 304L stainless steel plate surface (LT plane)
deposited with arrays of MgCl2 droplets of fixed deposition diameter (~2.5 mm)
with different CDDs ranging from ~0.5 to 10000 µg/cm2. The droplet marked
with red star had two droplets superimposed.
169
Figure 6-7 A 304L stainless steel plate surface (LT plane) deposited with arrays
of MgCl2 droplets of fixed deposition diameter (~2.5 mm) with different CDDs
ranging from ~0.5 to 10000 µg/cm2, as shown in Figure 6-6 following exposure
at 43% RH and 30 °C for 1 day. The droplet marked with a red star had two
droplets superimposed.
Following exposure at 43% RH and 30 °C for 1 day, (Figure 6-7), droplets with
CDDs of 0.5 and 1 µg/cm2 did not pit after 1 day exposure. 5 out of 6 of the
droplets with a CDD of 5 µg/cm2 pitted, and 4 out of 6 of the droplets with a
CDD of 10 µg/cm2 pitted. Both sets droplets with CDDs of 5 and 10 µg/cm2
showed a “ring-like” shape after 1 day exposure, with the central hole being
bigger in the 5 µg/cm2 set (Figure 6-8). For CDD values ranging from 50
to10000 µg/cm2, all droplets pitted. Droplets with a CDD of 10000 µg/cm2 had
170
Figure 6-8 Typical examples of MgCl2 droplets of fixed deposition diameter
(~2.5 mm) with different CDDs ranging from ~0.5 to 10000 µg/cm2 on the 304L
stainless steel plate surface (LT plane), shown in Figure 6-7 following exposure
at 43% RH and 30 °C for 1 day.
levels. Pits that grew at 10000 µg/cm2 CDD were round and deep with an
average diameter of 85±10 µm. In most cases this totally consumed any
shallow dish region, if indeed there was one. Pitting attack at a CDD of
5000 µg/cm2 was deep and followed the approximate boundary of the shallow
dish region. In the CDD range from 1000 to 50 µg/cm2, an earing morphology
was observed, with attack into the metal on one side of the shallow dish region.
At CDD values of 1 and 0.5 µg/cm2, there was a faint pattern of rust but no
At CDD of 10 and 5 µg/cm2, pitting was seen but pits were irregular in shape
and could not easily be described in terms of having a “shallow dish region”. At
5 µg/cm2, usually a single cluster of very small (<5 µm in diameter) pits was
formed.
171
At a CDD of 10000 µg/cm2, one instance of a shallow dish region was seen that
had been heavily attacked by the continuing corrosion but had an apparent
diameter of less than 50 µm. The rest of the pits, 5 out of 6, didn’t show any
sign for the shallow dish and had a hemispherical pit shape. The shallow dish
110±20 µm, which is much larger than for those with a CDD of 10000 µg/cm2,
and the shallow dish diameters were reduced for a CDD of 5 µg/cm2.
Figure 6-9 Typical pit morphologies of pits grown under MgCl2 droplets of fixed
deposition diameter (~2.5 mm) with different CDDs ranging from ~0.5 to
10000 µg/cm2 on the 304L stainless steel plate surface (LT plane), as shown in
Figure 6-7 following exposure at 43% RH and 30 °C for 1 day.
The total pit diameter, active region diameter (as defined in Section 3.8.2), and
pit depth, for a fixed deposition area, is plotted against the initial chloride
deposition density in Figure 6-10. The results show that the diameter of the
At a CDD of 10000 µg/cm2 there is a high amount of scatter. Similarly, the total
pit diameter increases with increasing CDD up to 1000 µg/cm2, after which the
172
high amount of scatter makes determining a trend difficult. The pit depth initially
increases rapidly as the CDD increases from 5 to 50 µg/cm2, but for CDD
Figure 6-10 Pit size as a function of CDD for pits grown under MgCl2 droplets of
fixed deposition diameter (~2.5 mm) on 304L stainless steel plate (LT plane)
following exposure at 43% RH and 30 °C for 1 day. The total pit diameter and
active region diameter are defined in Section 3.8.2).
173
Figure 6-11 shows the results obtained for 304L following 1 week exposure at
45% RH under MgCl2 droplets of fixed deposition diameter of ~3.3 mm. Almost
all droplets pitted. A single pit per droplet was observed at CDDs of 10 and
8 µg/cm2, while at ≤6 µg/cm2 both single pits and multiple pits were observed. At
central hole increasing with decreasing CDD. Pit morphologies are shown in
Figure 6-12; it can be seen that with decreasing CDD the pits become smaller,
dish with attacked region while at deposition levels ≤1 µg/cm2, pits were
Figure 6-11 A 304L stainless steel plate surface (LT plane) deposited with
arrays of MgCl2 droplets of fixed deposition diameter (~3.3 mm) but with
different CDDs ranging from ~10 to 0.1 µg/cm2 following exposure at 45% RH
and 30 °C for 1 week.
174
Figure 6-12 Typical morphologies of pits grown under MgCl2 droplets of fixed
deposition diameter (~3.3 mm) but with different CDDs ranging from ~10 to
0.1 µg/cm2 on 304L stainless steel plate surface (LT plane) shown in Figure
6-11 following exposure at 45% RH and 30 °C for 1 week.
Figure 6-13 (upper image) shows the results obtained for 304L following 1 week
~2.5 mm. The electrolyte layer was found to be discontinuous. Although there
was a faint pattern of rust at CDD values of 0.1 and 0.2 µg/cm2, no obvious pits
were observed. Pits were observed at CDD values of 0.5-10 µg/cm2. Single pits
per droplet were not the only case observed; pits were associated with very
small pits (<5 µm in diameter) that often appear in clusters at lower deposition
levels such as 0.5 µm/cm2 (Figure 6-14). Morphologies of the largest diameter
pits are shown in Figure 6-13 (lower SEM images). Figure 6-15(a) shows a
typical example of a cluster of very small pits seen under an MgCl2 droplet with
43% RH and 30 °C for 1 week. Figure 6-15(b) shows an SEM image at the
same magnification of an area site of 130x120 µm2 of 304L stainless steel (LT
175
plane) final polished with OPS, which contains seven inclusions of ≤2 µm in
diameter. Two types of inclusions were observed: MnS inclusions and mixed
oxide inclusions. Table 6-1 shows the chemical composition of typical examples
Figure 6-13 Typical morphologies of pits grown under MgCl2 droplets of fixed
deposition diameter (~2.5 mm) with different CDDs ranging from ~10 to
0.1 µg/cm2 on 304L stainless steel plate surface (LT plane) following exposure
at 43% RH and 30 °C for 1 week.
176
Figure 6-14 Typical example of MgCl2 droplet of diameter (~2.5 mm) with a
CDD of 0.5 µg/cm2 (at deposition) on 304L stainless steel plate (LT plane),
following exposure at 43% RH and 30 °C for 1 week. Multiple pits grown under
the interrupted electrolyte layer.
Figure 6-15 (a) Typical example of a cluster of very small pits shown in Figure
6-14, observed under an MgCl2 droplet of diameter (~2.5 mm) with a CDD of
0.5 µg/cm2 (at deposition) on 304L stainless steel plate surface (LT plane),
following exposure at 43% RH and 30 °C for 1 week; (b) an SEM image at the
same magnification of an area of 130x120 µm2 of 304L stainless steel (LT
plane, surface is final polished with OPS) showing the distribution of inclusions:
1-4 were MnS inclusions while 5-7 were mixed oxide inclusions.
177
Table 6-1 The chemical composition of typical MnS and a mixed oxide inclusion
observed in 304L stainless steel (LT plane) detected with EDX.
Weight % O Al Si S Ti Cr Ni Mn Fe
MnS
0 0 0.3 5.2 0 17 7.0 8.7 balance
inclusion
mixed oxide
7.3 2.4 0.2 0.5 3.4 25 4.1 13 balance
inclusion
Figure 6-16 A 316L stainless steel plate surface (LT plane) deposited with
arrays of MgCl2 droplets of fixed deposition diameter (~2.5 mm) with different
CDDs ranging from ~0.5-10000 µg/cm2 following exposure at 43% RH and
30 °C for 1 day.
Figure 6-16 shows the results obtained for 316L following exposure at 43% RH
and 30 °C for 1 day. Compared to 304L exposed to the same conditions, fewer
droplets were pitted. Similar to 304L stainless steel, pits were not observed
under droplets of CDD of 0.5 and 1 µg/cm2 after 24 hours. The 5 µg/cm2
droplets show a faint pattern of rust, but with no clear evidence of pitting. Similar
to 304L stainless steel, both sets of droplets with a CDD of 5 and 10 µg/cm2
178
showed a “ring-like” shape, with the central hole being bigger in the 5 µg/cm2
set (Figure 6-17). For droplets with a CDD of 10 to 1000 µg/cm2 not all the
droplets pitted, but all of the droplets with a CDD of 5000 and 10000 µg/cm2 did
pit. Droplets under the largest CDD had spread significantly after 1 day, similar
The pit morphologies obtained under different CDD values are shown in Figure
6-18. No shallow dish regions were observed at 10000 µg/cm2 CDD; pits were
circular in shape, with an average diameter of 110±15 µm. In the CDD range of
5000 to 500 µg/cm2, pits with a shallow dish region were seen with an area of
deeper attack into the metal surface on one side. In the CDD range of 100 to
10 µg/cm2, pits were approximately circular with no clear trace of a shallow dish.
The shallow dish diameters were reduced as CDD was reduced to 5 µg/cm2
CDD.
179
Figure 6-18 Typical morphologies of pits grown under MgCl2 droplets of fixed
deposition diameter (~2.5 mm) with different CDDs ranging from ~0.5 to
10000 µg/cm2 on the 316L stainless steel plate surface (LT plane) shown in
Figure 6-16 following exposure at 43% RH and 30 °C for 1 day.
Among all the pitted droplets, only one droplet of the1000 µg/cm2 set contained
an elongated pit that had grown along the rolling direction of the alloy (Figure
6-19).
Figure 6-19 Morphology of a pit grown along the rolling direction on 316L, LT
plane, under MgCl2 droplet (~2.5 mm equivalent diameter, ~1000 µg/cm2 CDD
at deposition), after 1 day exposure at 43% RH and 30 °C; (a) optical image of
the droplet; (b) optical image of the pit after washing with DI water.
180
Figure 6-20 shows the pit diameter (total pit diameter and active region
diameter) and pit depth as a function of CDD for a fixed droplet area (volume). It
can be seen that the total pit diameter increases with increasing CDD with the
value at 5000 µm/cm2 being within a high range of scatter which might be
attributed to the thicker droplet in this set of droplets, as they did not show
significant spreading (Figure 6-16 and Figure 6-17(b)). Similarly, the active
region diameter increases with increasing CDD up to 500 µm/cm2, then shows a
high amount of scatter for 1000 and 5000 µm/cm2 before increasing at
10000 µm/cm2. For pit depth, the trend was similar to that observed for 304L:
there is a rapid increase in pit depth at the beginning with increasing CDD, then
181
Figure 6-20 The equivalent diameters of the active region and the total pit, and
the pit depth as a function of the CDD for pits formed in MgCl2 deposits of fixed
deposition diameter (~2.5 mm) on 316L stainless steel (LT plane) shown in
Figure 6-16 following exposure at 45% RH and 30 °C for 1 day.
Table 6-2 shows that for 2.5 mm diameter droplets exposed for 1 day at
45% RH, pitting is observed down to a CDD of 10 µg/cm2 for both 304L and
316L stainless steel, although fewer droplets pit for 316L compared with 304L.
At a CDD of 5 µg/cm2, 304L shows pitting, whereas 316L shows signs of rust
staining but no obvious pits. No pits were observed at CDDs of 1 or 0.5 µg/cm 2
182
For ~2.5 mm diameter droplets on 304L exposed at CDD values of 1 and
0.5 µg/cm2, no pits were observed after 1 day of exposure, but pits were
threshold after a longer exposure time. For the 1 week exposures, pitting was
not observed at CD values of 0.2 and 0.1 µg/cm2. However, larger droplets
(~3.3 mm diameter) did show pitting even at a CDD value of 0.1 µg/cm2 (Figure
6-12). The table also include the number of pits associated with micro-droplets.
183
6.3 Secondary spreading and micro-droplet formation
observed only when the pit initiated close or at the droplet edge. It can be seen
Figure 6-21 Typical example of micro-droplet formation near a pit grown under a
MgCl2 droplet (~2.5 mm diameter, ~1000 µg/cm2 CDD at deposition) on 304L
stainless steel (LT plane) after 1 day exposure at 56% RH and 30 °C; (a)
droplet with pit at right edge; (b) magnified section of (a) showing the pit and
micro-droplet region; and (c) magnified sections of (b), showing the distribution
of micro-droplets.
184
Figure 6-22 shows the time-lapse results of atmospheric pitting corrosion of
temperature 24±2 °C) for 48 hours under a droplet of MgCl2 with a CDD of
~1000 µg/cm2. It can be seen that as the pit mouth grows with time, the size of
the micro-droplet region increases. Close inspection showed that the size of
outwards in a semicircle or oval shape near the pit following the grinding marks.
A corroded droplet on 304L (LT plane) after 1 day exposure at 35% RH with a
pitting site developed near the droplet edge is shown in Figure 6-23 (a).
these micro-pits is shown in Figure 6-23 (c), with bigger micro-pits (≤7 µm in
185
diameter) formed close to the pit (which formed at the primary droplet edge)
diameter). Figure 6-24 shows the formation of the similar features on 316L
186
Figure 6-24 Typical example of small pits formed under micro-droplets close to
the droplet edge. The pit was grown under a MgCl2 droplet (~2.5 mm diameter,
~1000 µg/cm2 CDD at deposition) on 316L stainless steel (LT plane) after 1
week exposure at 35% RH and 30 °C; the sample is washed with DI water.
65% RH is shown in Figure 6-25 (a). The secondary micro-droplets are clearly
seen at the bottom right edge Figure 6-25 (b). Also visible is formation of hemi-
Larger micro-droplets formed close to the pit. However, Figure 6-25 (c) and (d)
187
Figure 6-25 Typical example of micro-droplet formation near a pit grown under a
MgCl2 droplet (~2.5 mm diameter, ~1000 µg/cm2 CDD at deposition) on 304L
stainless steel (LT plane) after 1 day exposure at ~65% RH and 30 °C; (a)
droplet with pit at the droplet edge; (b) magnified section of (a) showing the pit
and micro-droplets distribution around the pit; (c) SEM image after washing with
DI water; (d) magnified section of (c), showing no evidence of micro-pits.
188
6.4 Discussion
In Figure 6-5, it is evident that the pit diameter increases with increasing droplet
diameter for both 304L and 316L stainless steels (for a fixed chloride deposition
density and relative humidity), which is in agreement with the idea that the pit
growth rate increases with the available cathodic current; the larger the cathodic
area the more the cathodic current [4, 22, 23, 147, 199, 208, 209]. The
the oxygen reduction reaction. The increase in droplet area gives area for the
stainless steel foils by Mi [18, 22] who reported that the pit diameter was found
to increase when the original salt deposit diameter was increased for a fixed
2.8 mm) after exposure at 45% RH and 300 K for 24 hours. Mi attributed this to
There was no significant change in pit depth with droplet area, or alloy type, for
a fixed exposure time (Figure 6-5). This is consistent with the idea that metal
dissolution under the salt layer at the bottom of the pit is under diffusion control.
This agrees with Ghahari [107], who reported that dissolution rate at the bottom
189
He attributed this to the presence of a salt layer, which regulates the interfacial
potential, consistent with previous work, e.g. [121, 122, 210, 211]. Recently, the
presence of salt layer at the bottom of the pit was confirmed by using X-ray
In Figure 6-5, it is also evident that slightly wider pits form on 304L compared
with those that form on 316L. The sides of the pit are not considered to have a
salt layer, as they are closer to the mouth of the pit where the solution is more
dilute as metal ions can escape into the bulk solution. As such, the pit sides are
under activation control and can experience much more materials loss. The
decrease in active dissolution rate for 316L compared with 304L [212] would
result in a lower current density on the pit walls and therefor a smaller pit mouth
[213].
Figure 6-10 and Figure 6-20 for 304L and 316L, respectively, show that, in
general, for the same initial deposition area and constant relative humidity, the
pit diameter increases with increasing the chloride deposition density (amount
of salt deposition). This finding is consistent with Mi’s [18] observation of pit
the effect of chloride deposition density on pit size (volume) under constant
relative humidity and temperature in both lab based tests and in X-ray
190
stainless steel foil and stainless steel pins exposed to lab temperature and
experiments (Figure 2-21) shows that the electrolyte layer height increases with
(Figure 6-17) suggests that this is the case in the present study as well. Since
the droplets are deposited with same initial deposited area, changes in the
amount of salt deposition will result in changes of the droplet height for
humidity.
There are two possible influences that can result from the droplet height: the
length of oxygen (the cathodic reactant) from the atmosphere [21]; as the
diffusion path increases, the oxygen reduction reaction rate will decrease,
predicting the formation of smaller pits. A thicker electrolyte, on the other hand,
can lead to a lower approach resistance between the pit and the cathodic area,
resulting in a higher pit current density, and anticipating that larger pits would
form [5, 18, 22, 147]. The observation of larger pits with higher CDD values
The pit depth as a function of CDD for 1 day exposure (Figure 6-10 and Figure
6-20) shows approximately the same trend and depth range in both 304L and
191
316L stainless steel. As described above, the growth of pits is controlled by the
diffusion of metal ions from the bottom of the pit, which will not be greatly
Pits propagate at CDDs of 10 µg/cm2 and above with diameters >30 µm, and
lower deposition levels (<10 µg/cm2), much smaller pits are formed (<30 µm). At
lower CDD values, ring-like droplet morphologies where observed e.g. for 1 day
exposure of constant droplet diameter of ~2.5 mm for both 304L (Figure 6-8)
and 316L (Figure 6-17), at 10 and 5 µg/cm2, with the central hole being larger in
the 5 µg/cm2 deposition level. Ring-like droplet shapes were also observed after
6-11) at deposition levels starting from 4 µg/cm2 and lower, with the central hole
steel (Figure 2-22). They described the appearance as “multiple droplets with
anodic and cathodic domains”. These authors attributed the discontinuity of the
film to the shrinkage of the droplet while it is reaching equilibrium with the
compared to that obtained in the present study may be related to the fact that in
192
their study, relatively larger droplets of MgCl2 of 5.5 to 6.5 mm in diameter
that the easy oxygen access at the thinner parts of the droplets, i.e. at the edge,
may change the interfacial energy between the passive film and the solution
and allow a reduced contact angle to occur at very low CDDs. Products with low
CO2), may also form around the edge of the droplet due to the change in pH,
It was also observed that at lower deposition levels (<10 µg/cm2) e.g. 5 µg/cm2
(Figure 6-9), ≤1 µg/cm2 (Figure 6-12), and 0.5 µg/cm2 (Figure 6-14) pits were
found to be associated with small clusters of very small pits. Figure 6-15(b)
lowest CDD is about the same as the inclusions, consistent with inclusion
dissolution, and at slightly higher CDDs the pits are slightly larger, so the pits
have grown a bit. This was attributed to the fact that once a pit initiated and start
to grow an acidic region will develop around it [88, 97, 99]. It is possible that the
transport of ionic species, can play an important role in pitting corrosion [4, 5,
18-20, 22, 214]. Tomashov [214] described the effect of water film thickness on
193
the atmospheric corrosion rate, and reported that this can have a maximum
For atmospheric pitting corrosion to occur, it has been reported that there is a
lower threshold of chloride deposition density of 10-25 µg/cm2 [19, 20, 163, 215]
. However, in the present work pits were observed at lower CDDs (Table 6-2)
and the CDD threshold was seen to be affected by the initial deposition area
(droplet diameter) and exposure time. This might be because, in the present
attack, on the order of 10 µm and below. The main difference between CDD
thresholds identified in the current work compared with previous work might
come from what is defined “a pit”. For example, Albores-Silva et al. [19]
steel. However, no images of pits are shown and the minimum pit size recorded
is not stated. Later, based on the same results of Albores-Silva, Padovani et al.
[20] reported that, pit depths were in the range of 10-20 µm for lower chloride
deposition densities, while for higher deposition densities they reported that the
pit depth was in the range of 20-70 µm. However, Padovani reported the
maximum pit diameter to be ~90 µm for the pits observed at 10 and 25 µg/cm2.
In the present work, looking at Figure 6-10 and Figure 6-20, we get a threshold
194
6.4.4 Secondary spreading and micro-droplet formation
The formation of micro-droplets and their growth which led to the formation of
observed, where the preferential pitting initiated close to droplet edge on both
304L and 316L stainless steels under droplets of MgCl2 at different fixed RH
values at 30 °C. Researchers who carried out experiments under the same
conditions presented in this work (MgCl2 salt solution, 45% RH and ~30 °C) on
304L stainless steel plate [7] and 304L stainless steel foil [18, 22] have not seen
microdroplets; this could be due to the lower magnification images used in their
studies or the pit location within the droplet since in the present work micro-
droplets were only observed when the pit initiated at or close to the droplet
edge.
range of salts (e.g. NaCl, KCl, and MgCl2,...etc.) on different materials such as
carbon steel, stainless steel, aluminium, copper and zinc. However, secondary
spreading was not observed on stainless steel under any of the examined salt
solutions including MgCl2 [136, 151, 156, 165, 166]. Tsuru et al. [136] and
Zhang et al. [151] attributed the prevention of formation and expansion of the
periphery of the main droplet: this diminishes the concentration of OH- and
increases the ionic strength at the edge of the main droplet (Figure 6-26).
195
Bian et al. [166] reported that no obvious micro-droplet formation was seen on
the stainless steel surface under KCl droplets. They attributed this to the very
around pits initiated at or close to the droplet edge (≤ 250 µm from the droplet
adge). However, not all the pits formed in this range showed the formation of
micro-droplet. It was almost but not always. The reason for why some pits in
this range did not show micro-droplet formation is still unknown. Some previous
researchers also reported that secondary micro-droplets formed around the pit
initiated close to the droplet edge on carbon steel [166] and aluminium [167,
168].
In this study it is also observed that the micro-droplets region grew outwards in
a semi-circle or oval shape; this agrees with the shape previously reported by
196
some researchers [165, 166]. Bian et al. [166] attributed the oval shape to the
superficial grinding marks on carbon steel. Similarly, Zhang et al. [165] reported
that the boundary of the elliptic area covered with micro-droplets spread faster
This effect is confirmed by the time-lapse carried out in the present work (Figure
6-22).
Figure 6-22 shows that the size of the micro-droplet region and the size of
individual droplets increase with increasing pit size over time. This indicates that
(e.g. Mg+2) in the primary droplet towards the spreading region. This finding is
consistent with the findings of previous authors who investigated the secondary
previously reported to form under NaCl and other salts on carbon steel [136,
156, 166], aluminium [164, 166, 167], zinc [216], magnesium alloys [217], and
Figure 6-23 shows the formation of many micro-pits under micro-droplet region
which formed close to the pit initiated near to the main droplet edge. This
stainless steel under MgCl2 droplets or other salts. For the study of the
aluminium alloy AA7075‐T6 under NaCl droplets, Morton and Frankel [167]
droplets (Figure 6-27). They proposed that the anode (preferential pitting site)
197
close to the main droplet edge can form an acidic zone that develops further pit
initiation and growth. The growth of the secondary droplet is driven by the
generation of ions from the anodic region at the edge of the droplet, which can
absorb water from the atmosphere. Morton and Frankel also reported that the
generating many small pits throughout the wetted surface area. The mechanism
proposed by Morton and Frankel is similar to the mechanism considered for the
for Al–Mg–Si alloy under droplets of MgCl2 solution. Thus, in the present work,
an acidic zone that develops further pit initiation and growth. Such micro-pits
were not observed under micro-droplets at high RH values (>56% RH) but they
were observed at 56% RH and below for both 304L and 316L stainless steels.
This might be due to the less concentrated (more dilute) conditions at high RH
198
Figure 6-27 A Schematic representation of secondary droplet phenomenon of
AA7075-T6 under NaCl droplet salt solution. Region 1 represents the pit
initiation site (anode) within the main droplet associated with many small pits
where the secondary spreading takes place. Region 3 represents the main
supporting cathodic area surrounding the anode. Region 2 is covered with
corrosion products. The dotted arrows represent the ion transport between
these two regions. Region 4 represents the main part of the droplet [167].
droplets and secondary spreading area by Tsuru et al. [136] for MgCl2 and NaCl
(Figure 6-26), Zhang [156] for NaCl (Figure 2-28) , and Morton and Frankel
[167] for a pit at the edge of the droplet under NaCl salt solution on aluminium
(Figure 6-27), a mechanism can be proposed for the formation of such features
under MgCl2 salt solution for a pit initiated at or close to the edge of the original
It is assumed that a very thin water layer will form around the original droplet
because of the presence of highly humid environment. Once a pit initiates next
to or at the edge of the droplet, an acidic zone will develop and act as a local
199
anode. When preferential sites act as cathodes on the monolayer, this will form
reactions that occur on the water clusters would create a charge imbalance,
driving cations (e.g. Mg+2) towards them and forming microdroplets. Over time,
demonstration of this mechanism for the case of pit initiated at/close to the main
spreading area under MgCl2 on stainless steel is consistent with the previous
spreading area in terms of cations being drawn towards OH- at the edge of
200
migration of cations is unlikely, as this mechanism appeared to work on when
The explanation given by Morton and Frankel [167] for the formation of multiple
pits under the secondary spreading region in Al alloy may be the correct
explanation for the formation of multiple pits under the secondary spreading
region in austenitic stainless steels investigated in this study. They states that
the local acidic region that is resulted from the cation hydrolysis promotes
further pit initiation and growth (Figure 2-29 and Figure 6-27). It has been
reported that separate corrosion cells can form in single micro-droplet(s) and
cause corrosion [19]. Albores-Silva et al. [19], reported that when the continuity
of the salt film is interrupted, this would lead to the formation of multiple droplets
with anodic and cathodic domains and this can transfer the corrosion under a
on the surface outside the border of the main droplet, According to both Zhang
et al. [151] (Figure 2-28) and Tsuru et al. [136] (Figure 6-26). This circuit will
have a very high IR drop due to the thinness of the ionic pathway. As the
circuit with anodes that develop locally rather than crossing the monolayer
towards the main droplet. As such, dissolution can occur under microdroplets.
In this work, pitting under microdroplets is not seen in high RH conditions, such
as observed at 65% RH. This could be because at this RH there are more
201
and the anode in the main droplet. At 35% RH, however, the ionic pathway can
ferrite can effect on pit initiation and/or propagation. Pits with layered attack
morphology were observed on 304L stainless steel (e.g. Figure 6-2 and Figure
6-25). This was attributed to the preferential attack (propagation) along ferrite
bands. Pits with rough interior were observed on 316L with no layered attack.
On the other hand, it is well agreed that inclusions such as MnS [1, 51, 88-90]
and mixed oxide inclusions [50, 91-93] can play a vital role in pitting corrosion in
stainless steels. For 304L stainless steel (LT plane), in the present study, pits
were often found to be associated with small clusters of very small pits
especially at lower deposition levels (<10 µg/cm2) e.g. 5 µg/cm2 (Figure 6-9),
≤1 µg/cm2 (Figure 6-12), and 0.5 µg/cm2 (Figure 6-14)).The typical inclusions
CDD, the smallest pits are about the same as the inclusions. At slightly higher
CDDs the pits were slightly larger, so the pits have grown a bit (consistent with
inclusion dissolution (see Figure 2-10 [88], Figure 2-12 [97], and Figure 2-15
[4])). It is widely accepted that once a pit initiated and start to grow an acidic
region will develop around it [88, 97, 99]. It is well agreed that the acidic region
will dissolve nearby inclusions. Therefore, the formation of small pits was
202
Among all the droplets investigated in this section (328 droplets), only one
droplet had an elongated pit that seemed to be growing along the rolling
alloy surface.
6.5 Conclusions
1. For atmospheric pitting corrosion of both 304L and 316L stainless steels
consistent with the view that this is under diffusion control. For the same
2. For both 304L and 316L stainless steels, for a fixed droplet area at 45%
RH, the active pit diameter (as distinct from the initial shallow dish area)
for a 1-day corrosion test shows a relatively small increase with CDD
203
3. At a CDD of 10 µg/cm2 and below, the active pit diameter is far smaller
dissolved, but the corrosion does not spread. For 304L, this “threshold”
is lower for longer times and larger droplets, and the threshold is higher
MgCl2 salt droplets when preferential pitting occurred, almost but not
always, at or close to the droplet edge (≤ 250 µm). As the pit mouth grew
area (micro-pits were observed at 56% RH and below but not observed
204
7 General Discussion
corrosion of stainless steel will take place preferentially along delta ferrite bands
parallel to the surface of the plate. This is probably due to the difference in the
chemical composition of the ferrite phase with respect to the austenite phase
duplex stainless steels. The alloy plate used in the present work is typical of that
are very narrow (typically <1 µm), so if stress were to be applied perpendicular
to the fissures, then cracking could be a significant risk owing to the high stress
intensity. However, this risk is low because in such containers the residual
stresses will be parallel to the plate surface [26]. Furthermore, when pitting
corrosion develops on the plate surface, the pits spread laterally parallel to the
The development of intermetallic phases e.g. MnS and mixed oxides inclusions
which are also aligned along the rolling direction can leads to the formation of
elongated pits. The use of a clean grade of Type 304 stainless steel alloys,
which are processed with the intent of having fewer and smaller inclusions [50],
is recommended.
205
Solution annealing of samples has shown that residual delta ferrite can be
eliminated or reduced from the alloy. As a result the pit morphology has
changed from irregular to equiaxed etched pits. Knowledge of the ferrite content
and to better develop new containers that avoid the existence of these phases,
for example by better processing (e.g. more thorough solution treatment). Such
extra processing steps, however, may raise the cost of the containers, and so
It has been observed that plate thickness might have a role in influencing the pit
the same plane e.g. Figure 5-35 no layered attack was observed and instead
the pit interior was rough and correlated with the grain size of the alloy. It is
speculated that this is due to the presence of less centreline segregation at the
edge of the 316L stainless steel plate compared to 304L. This could be an
to prevent the formation of pits that could develop into AISCC over long time-
In the present work, pits have been observed at CDDs as low as 0.1 µg/cm2.
However, these are small, formed in clusters and were similar in size to
206
inclusions, likely to represent dissolution of inclusions. It is unlikely that such
It has been possible to say that in general the pit diameter increases with
increasing chloride deposition density (Figure 6-10 and Figure 6-20). Note that
Padovani et al. [20] reported that, pit depths were in the range of 10-20 µm on
316L after 12 weeks for “lower” chloride deposition densities (unspecified, but
densities they reported that the pit depth was in the range of 20-70 µm. These
results match with those found here after only 1 day exposure. Although the
~43% RH for the tests discussed here), suggesting that pit depth does not
reported the maximum pit diameter to be ~90 µm for the pits observed at 10
suggests that the pits are still developing laterally, if not increasing in depth.
Padovani et al. [20] suggested that the reasons for the existence of thresholds
for the chloride deposition density required to assist pitting corrosion are linked
observed at lower CDD values (Figure 6-8 and Figure 6-17). Padovani and
Albores-Silva, also show this ring like formation at low CDDs. These
207
observations are consistent with the idea that the 10 µg/cm 2 threshold is the
electrolyte cannot provide enough cathode areas to support pit propagation that
nuclear waste is one of the objectives of the current study. Although in the
present study the threshold of chloride deposition density was even lower
compared to that obtained in literature 10-25 µg/cm2 [19, 20, 163, 215], the
concentrations are present, controlled by the use of air filters in storage areas.
This work has confirmed that the pit morphology is a function of RH. At lower
RH values, satellite and earing pits (both had shallow dish region) were
observed while at higher RH values only circular pits were observed (with no
shallow dish region). The formation of shallow dish has been attributed to the
passive current dissolution. It has been demonstrated that the diameter of the
shallow dishes is a function of the distance from the edge of the droplet and it is
208
the shallow dish is most strongly influenced by cathodic area. It has also been
observed that a greater number of earing pits formed toward the centre of the
droplet while satellite pits formed toward the droplet edge. This morphology
difference has been attributed to the changes in the IR drop, and differences in
solution diffusivity and conductivity, which are linked with RH. It has been
corrosion is in the range of 65%–70% RH for 304L stainless steel while that for
It has been possible to say that in general the pit diameter increases with
increasing chloride deposition density. This might be related with a lower ohmic
drop between the pit and the droplet layer. It has also been observed that
localised pitting corrosion can take place on stainless steels with extremely low
30 °C. Definite of CDD threshold has been determined for both 304L and 316L
stainless steels. It has been shown that this threshold is higher for 316L
compared with 304L. And for 304L the threshold is lower for longer times and
conditions of intermediate level nuclear waste and keep the chloride deposition
levels limited.
By investigating large arrays of droplets it has been observed that the pit
exposure of both 304L and 316L stainless steels beneath MgCl2 droplets at
209
45% RH and 30 °C. This has been attributed to an increase in the effective
cathode area, which can cause an increase in potential. However, it has been
observed that there is no significant change in pit depth, which indicates that
this is under diffusion control. It has also been observed that slightly wider pits
was seen. These effects have been observed under full immersion
Although previous work has made comments regarding the position of pits
within a droplet [6, 23, 82, 199, 220], this work has measured the effect of the
pit location within a droplet on its morphology. It has been demonstrated that
pits that form close and/or at the droplet edge have a bigger size. This has been
attributed to three effects: (1) the effect of a three phase boundary, where there
is particularly easy access of oxygen for the cathodic reaction, (2) the formation
of an acidic zone that develop more pit initiation and growth, and (3) the
formation of secondary spread zone which increases the covered wetted area
stainless steel samples have shown that secondary spreading processes take
place when pits form close and/or at the droplet edge, a process which has
previously been unreported under MgCl2 droplets. This occurs very rapidly and
beneath MgCl2 droplets. Consequently this increases the size of the potential
cathodic region.
210
7.6 Deposition technique
conditions. Salt droplets can be deposited in arrays with controlled volumes and
concentrations. In this work large arrays of droplets were used (up to 75) in
211
8 Future work
In the current work, the results showed that the faceted pits (flat walled pits) can
possible that pit bounding planes are belong to {111} crystal planes of austenite
which have the lowest dissolution rate. However, preferential attacked planes
plays an important role in localized corrosion. Thus this effect needs further
study.
corrosion pits
Containers have vertical sides and upward facing and downward facing
surface orientation under full immersion conditions [75, 222]. The effect of
sample position (Figure 8-1) on the shape and the kinetics of pit growth in
reported that pits grew faster into the metal in the case of covered pits facing
intermediate rate were reported in the case of pits growing on a vertical mode
212
(type 3). Figure 8-1 shows the assumed distribution of salt layer (dotted areas)
in pits growing in three different modes. Donohoe et al. (Figure 8-2(a)) showed
used in a nuclear waste cooling water system. When looking at the cross-
section of this pit (Figure 8-2(b)) it was observed that the pit propagated
vertically up the tank wall without penetrating the other side of the metal. This
was attributed to variation in alloy composition in the centre of the steel plate.
However, under atmospheric conditions such effect has not been investigated.
to study this effect. It is expected that dissolution will follow structural features
(microstructural orientation).
Figure 8-1 distribution of salt layer (dotted areas) in pits growing in three
different modes; downward mode (Type 1); upward mode (Type 2); and vertical
mode (type 3)[222].
213
Figure 8-2 Localized corrosion of 18Cr‑8Ni‑1Ti stainless steel tank used in a
nuclear waste cooling water system; (a) a pit on a section of the internal surface
; (b) a cross-sectional view of pit shown in (a). The symbol “E” in both figures
represents the approximate position of the pit mouth (''exit hole''). The plate
thickness was nominally ~4.8 mm. This figure adapted from [75].
214
8.3 Effect of deposit chemistry on atmospheric pitting
Ca2+, Cl-, Mg2+, SO42-, K+, PO43-, NH4+, NO3-, Na+ [223]. Surveys within nuclear
waste stores have shown sulphate (SO42−) and nitrate (NO3−) [163]. These are
MgCl2 [4, 5, 15, 21, 139, 146], pure NaCl [15, 23, 129, 146, 147], CaCl2 [16] and
artificial sea water [4, 5], whereas, only few of them investigate the effect of
other mixed salts [7]. In the present work, all the experiments have been
performed under MgCl2 salt solution droplets. Atmospheric pitting corrosion test
could be planned on the same materials under mixed salt solution droplets at
procedures.
the current work were performed under constant relative humidity while in real
realistic.
Although some previous studies have investigated the effect of wet-dry cycling
for different materials [7-14], there is need for more detailed work to study this
215
effect under atmospheric corrosion conditions on stainless steels in particular
216
9 Appendices
Figure 9-1 and Figure 9-2 give the mill certificates provided by the
manufacturers of the stainless steel plates used in the work presented in this
thesis.
217
9.2 Appendix 2, data logger diagrams
220
Figure 9-4 Monitored RH and temperature of saturate salts, given in table 1-3,
for 1 week of exposure.
221
10 References
222
15. Prosek, T., A. Iversen, C. Taxén, and D. Thierry, Low-Temperature
Stress Corrosion Cracking of Stainless Steels in the Atmosphere in the
Presence of Chloride Deposits. Corrosion, 2009. 65(2): p. 105-117.
16. Prosek, T., A. Le Gac, D. Thierry, S. Le Manchet, C. Lojewski, A. Fanica,
E. Johansson, C. Canderyd, F. Dupoiron, T. Snauwaert, F. Maas, and B.
Droesbeke, Low-Temperature Stress Corrosion Cracking of Austenitic
and Duplex Stainless Steels Under Chloride Deposits. Corrosion, 2014.
70(10): p. 1052-1063.
17. Prosek T., Material classification for climbing anchors - Phase 2:
Development of an anchor classification and related test procedure for
the evaluation of stress corrosion cracking (SCC) and corrosion
resistances of climbing anchors. 2014, Institut de la Corrosion (French
Corrosion Institute): Brest (France). p. 1-27.
18. Mi N., Synchrotron X-ray Studies of Atmospheric Pitting Corrosion of
Stainless Steel, [PhD thesis]. Birmingham (UK):University of
Birmingham; 2013.
19. Albores-Silva, O.E., E.A. Charles, and C. Padovani, Effect of chloride
deposition on stress corrosion cracking of 316L stainless steel used for
intermediate level radioactive waste containers. Corrosion Engineering,
Science and Technology, 2011. 46(2): p. 124-128.
20. Padovani, C., O.E. Albores-Silva, and E.A. Charles, Corrosion Control of
Stainless Steels in Indoor Atmospheres—Laboratory Measurements
Under MgCl2 Deposits at Constant Relative Humidity (Part 1). Corrosion,
2015. 71(3): p. 292-304.
21. Maier, B. and G.S. Frankel, Pitting corrosion of bare stainless steel 304
under chloride solution droplets. Journal of the Electrochemical Society,
2010. 157(10): p. C302-C312.
22. Mi, N., M. Ghahari, T. Rayment, and A.J. Davenport, Use of inkjet
printing to deposit magnesium chloride salt patterns for investigation of
atmospheric corrosion of 304 stainless steel. Corrosion Science, 2011.
53(10): p. 3114-3121.
23. Wang, Y., W. Wang, Y. Liu, L. Zhong, and J. Wang, Study of localized
corrosion of 304 stainless steel under chloride solution droplets using the
wire beam electrode. Corrosion Science, 2011. 53(9): p. 2963-2968.
24. Örnek C., A. H. Ahmed, and D. L. Engelberg, Effect of Microstructure on
Atmospheric Induced Corrosion of Heat treated Grade 2205 and 2507
Duplex Stainless Steels, in Eurocorr 2012. 2012: Istanbul -Turkey.
25. Örnek C; D. L. Engelberg, S.B.L., T. L. Ladwein, Effect of “475°c
embrittlement” on the corrosion behaviour of grade 2205 duplex stainless
steel investigated using local prob ing techniques. Corrosion
Management 2013: p. 9-11.
26. Davenport, A.J., L. Guo, N. Mi, H. Mohammed-Ali, M. Ghahari, S.R.
Street, N.J. Laycock, T. Rayment, C. Reinhard, C. Padovani, and D.
Krouse, Mechanistic studies of atmospheric pitting corrosion of stainless
steel for ILW containers. Corrosion Engineering, Science and
Technology, 2014. 49(6): p. 514-520.
27. Örnek, C. and D.L. Engelberg, SKPFM measured Volta potential
correlated with strain localisation in microstructure to understand
223
corrosion susceptibility of cold-rolled grade 2205 duplex stainless steel.
Corrosion Science, 2015. 99: p. 164-171.
28. Örnek, C. and D.L. Engelberg, Correlative EBSD and SKPFM
characterisation of microstructure development to assist determination of
corrosion propensity in grade 2205 duplex stainless steel. Journal of
Materials Science, 2015: p. 1-18.
29. Örnek, C., X. Zhong, and D.L. Engelberg, Low-Temperature
Environmentally Assisted Cracking of Grade 2205 Duplex Stainless Steel
Beneath a MgCl2:FeCl3 Salt Droplet. Corrosion, 2016. 72(3): p. 384-399.
30. Laycock, N.J. and R.C. Newman, Temperature dependence of pitting
potentials for austenitic stainless steels above their critical pitting
temperature. Corrosion Science, 1998. 40(6): p. 887-902.
31. Leckie, H.P. and H.H. Uhlig, ENVIRONMENTAL FACTORS AFFECTING
CRITICAL POTENTIAL FOR PITTING IN 18-8 STAINLESS STEEL.
Journal of the Electrochemical Society, 1966. 113(12): p. 1262-&.
32. Asami, K. and K. Hashimoto, Importance of initial surface film in the
degradation of stainless steels by atmospheric exposure. Corrosion
Science, 2003. 45(10): p. 2263-2283.
33. Sugimoto, K. and Y. Sawada, The role of molybdenum additions to
austenitic stainless steels in the inhibition of pitting in acid chloride
solutions. Corrosion Science, 1977. 17(5): p. 425-445.
34. Schweitzer P.A., Fundamentals of Metallic Corrosion: Atmospheric and
Media Corrosion of Metals. Second ed. 2007, USA: CRC Press.
35. Kyriakongonas A. P., 3D Numerical Modeling of Austenitic Stainless
Steel 316L Multi-pass Butt Welding and Comparison with Experimental
Results [Master of Scince]. National Technical University of Athens;
2009.
36. Akgün, O.V., M. Ürgen, and A.F. Çakir, The effect of heat treatment on
corrosion behavior of laser surface melted 304L stainless steel. Materials
Science and Engineering: A, 1995. 203(1–2): p. 324-331.
37. Fu J. W., Y. S. Yang, J. J. Guo, J. C. Ma, and W. H. Tong, Microstructure
evolution in AISI 304 stainless steel during near rapid directional
solidification. Materials Science and Technology, 2009. 25(8): p. 1013-
1016.
38. Fu, J.W., Y.S. Yang, J.J. Guo, J.C. Ma, and W.H. Tong, Formation of a
two-phase microstructure in Fe–Cr–Ni alloy during directional
solidification. Journal of Crystal Growth, 2008. 311(1): p. 132-136.
39. Lippold J.C. and Savage W.F., Solidification of austenitic stainless steel
weldments, part 1 - a proposed mechanism. Welding Journal, 1979.
58(12): p. 362s-374s.
40. Cieslak, M.J., Ritter, A. M., Savage, W. F., Solidification cracking and
analytical electron microscopy of austenitic stainless steel weld metals.
Welding Journal, 1982. 61(1): p. 1-s to 8-s.
41. Elmer, J.W., S.M. Allen, and T.W. Eagar, Microstructural development
during solidification of stainless steel alloys. Metallurgical Transactions A,
1989. 20(10): p. 2117-2131.
224
42. Rho, B.S., H.U. Hong, and S.W. Nam, The effect of δ-ferrite on fatigue
cracks in 304L steels. International Journal of Fatigue, 2000. 22(8): p.
683-690.
43. Manning, P.E., C.E. Lyman, and D.J. Duquette, A STEM Examination of
the Localized Corrosion Behavior of a Duplex Stainless Steel. Corrosion,
1980. 36(5): p. 246-251.
44. Mangonon, P. and G. Thomas, The martensite phases in 304 stainless
steel. Metallurgical Transactions, 1970. 1(6): p. 1577-1586.
45. Tavares, S.S.M., D. Fruchart, and S. Miraglia, A magnetic study of the
reversion of martensite α′ in a 304 stainless steel. Journal of Alloys and
Compounds, 2000. 307(1–2): p. 311-317.
46. Seetharaman, V. and R. Krishnan, Influence of the martensitic
transformation on the deformation behaviour of an AISI 316 stainless
steel at low temperatures. Journal of Materials Science, 1981. 16(2): p.
523-530.
47. Childress, J., S.H. Liou, and C.L. Chien, Ferromagnetism in metastable
304 stainless steel with bcc structure. Journal of Applied Physics, 1988.
64(10): p. 6059-6061.
48. Shteinberg, M.M., Y.N. Goikhenberg, M.A. Smirnov, and D.A. Mirzaev,
Strengthening of stainless steels by deformation, the martensitic
transformation, and aging. Metal Science and Heat Treatment, 1972.
14(7): p. 586-588.
49. Altenberger, I., B. Scholtes, U. Martin, and H. Oettel, Cyclic deformation
and near surface microstructures of shot peened or deep rolled austenitic
stainless steel AISI 304. Materials Science and Engineering: A, 1999.
264(1–2): p. 1-16.
50. Jun, J., K. Holguin, and G.S. Frankel, Pitting Corrosion of Very Clean
Type 304 Stainless Steel. Corrosion, 2014. 70(2): p. 146-155.
51. Stewart, J. and D.E. Williams, The initiation of pitting corrosion on
austenitic stainless steel: on the role and importance of sulphide
inclusions. Corrosion Science, 1992. 33(3): p. 457-474.
52. Llewellyn, D.T. and R.C. Hudd, Steels - Metallurgy and Applications (3rd
Edition). 1998, Elsevier: Butterworth-Heinemann: Oxford.
53. DENG;, X., M. JIANG;, and X. WANG;, Mechanisms of inclusion
evolution and intra-granular acicular ferrite formation in steels containing
rare earth elements. Acta Metall Sin(English), 2012. 25(3): p. 241-248.
54. Holappa, L., M. Hämäläinen, M. Liukkonen, and M. Lind, Thermodynamic
examination of inclusion modification and precipitation from calcium
treatment to solidified steel. Ironmaking & Steelmaking, 2003. 30(2): p.
111-115.
55. Lind, M., MECHANISM AND KINETICS OF TRANSFORMATION OF
ALUMINA INCLUSIONS IN STEEL BY CALCIUM TREATMENT,
[Doctoral Thesis]. Helsinki University of Technology; 2006.
56. Yang, J., X.-h. Wang, M. Jiang, and W.-j. Wang, Effect of Calcium
Treatment on Non-Metallic Inclusions in Ultra-Low Oxygen Steel Refined
by High Basicity High Al2O3 Slag. Journal of Iron and Steel Research,
International, 2011. 18(7): p. 8-14.
225
57. Oikawa, K., H. Ohtani, K. Ishida, and T. Nishizawa, The Control of the
Morphology of MnS Inclusions in Steel during Solidification. ISIJ
International, 1995. 35(4): p. 402-408.
58. Schaeffler, A.L., Constitution diagram for stainless steel weld metal.
Metal Progress, 1949. 56(11): p. 680-680B.
59. DeLong W.T., G.A. Ostrom, and E.R. Szumachowski, Measurement and
calculation of ferrite in stainless-steel weld metal. Welding Journal, 1956.
35(11): p. 526.
60. McCowan C.N., T.A. Siewert, and D.L. Olson, Stainless steel weld metal:
prediction of ferrite content. WRC Bulletin, 1989. 342.
61. Lake, F.B., Effect of copper on stainless steel weld metal ferrite content.
Paper presented at the 1990 AWS Annual Meeting Anaheim CA.
62. Dong-Cherng Wen, Influence of Hot Rolling and Post-Tempering on the
Mechanical Properties of Duplex Stainless Steel Containing Martensite
and Ferrite. MATERIALS TRANSACTIONS, 2006. 47(11): p. 2779-2785
63. Lake F., A new constitution diagram for predicting ferrite content of
stainless steel weld metals. Materials & Design, 1993. 14(6): p. 345-348.
64. Elmer, J.W. and T.W. Eagar, Measuring the residual ferrite content of
rapidly solidified stainless steel alloys. Welding Journal, 1990: p. 141-s-
150-s.
65. Cullity, B.D., Elements of X-ray Diffraction. 2nd ed. 1978, USA: Addison–
Wesley Publishing Company.
66. Bermejo, M.A.V., Predictive and measurement methods for delta ferrite
determination in stainless steels. Welding journal, 2012. 91: p. 113.
67. Kim, S.H., H.K. Moon, T. Kang, and C.S. Lee, Dissolution kinetics of
delta ferrite in AISI 304 stainless steel produced by strip casting process.
Materials Science and Engineering: A, 2003. 356(1–2): p. 390-398.
68. Kožuh, S., M. Goic, and L. Kosec, The effect of annealing on properties
of AISI 316L base and weld metals. RMZ-Materials and
Geoenvironment, 2007. 54(3): p. 331-344.
69. Toor, I.-u.-H., Effect of Mn Content and Solution Annealing Temperature
on the Corrosion Resistance of Stainless Steel Alloys. Journal of
Chemistry, 2014. 2014: p. 8.
70. Kim S. Y., H.S. Kwon, and H. Kim, Effect of Delta ferrite on corrosion
resistance of type 316L stainless steel in acidic chloride solution by
micro-droplet cell. Solid State Phenomena, 2007. 124-126: p. 1533-1536.
71. Hull, F.C., WELDING JOURNAL, 1967. 46: p. 399.
72. Matsuda, F., H. Nakagawa, T. Uehara, S. Katayama, and Y. Arata, A
New Explanation for Role of Delta-Ferrite Improving Weld Solidification
Crack Susceptibility in Austenitic Stainless Steel(Materials, Metallurgy,
Weldability). Transactions of JWRI, 1979. 8(1): p. 105-112.
73. BROOKS J. A., A. W. THOMPSON, and J.C. WILLIAMS, A Fundamental
Study of the Beneficial Effects of Delta Ferrite in Reducing Weld
Cracking. WELDING JOURNAL, 1984: p. 71s-83s.
74. Manning, P.E., D.J. Duquette, and W.F. Savage, , Technical Note: The
Effect of Retained Ferrite on Localized Corrosion in Duplex 304L
Stainless Steel. 1980, Welding Research Supplement. p. 260-s-262-s.
226
75. Donohoe, C.J., G.O.H. Whillock, and P.J. Apps, Localized Corrosion of
Stainless Steel in a Nuclear Waste Cooling Water System—Part 2: Plant
Inspection Findings. Corrosion, 2012. 68(9): p. 844-852.
76. Eghbali, F., M. Moayed, A. Davoodi, and N. Ebrahimi, Critical pitting
temperature (CPT) assessment of 2205 duplex stainless steel in 0.1 M
NaCl at various molybdate concentrations. Corrosion Science, 2011.
53(1): p. 513-522.
77. Tomashov, N.D., G.P. Chernova, and O.N. Marcova, Effect of
Supplementary Alloying Elements On Pitting Corrosion Susceptibility Of
18Cr-14Ni Stainless Steel. Corrosion, 1964. 20(5): p. 166t-173t.
78. Ebrahimi, N., M.H. Moayed, and A. Davoodi, Critical pitting temperature
dependence of 2205 duplex stainless steel on dichromate ion
concentration in chloride medium. Corrosion Science, 2011. 53(4): p.
1278-1287.
79. Deng, B., Y. Jiang, J. Gong, C. Zhong, J. Gao, and J. Li, Critical pitting
and repassivation temperatures for duplex stainless steel in chloride
solutions. Electrochimica Acta, 2008. 53(16): p. 5220-5225.
80. Garfias-Mesias, L.F., J.M. Sykes, and C.D.S. Tuck, The effect of phase
compositions on the pitting corrosion of 25 Cr duplex stainless steel in
chloride solutions. Corrosion Science, 1996. 38(8): p. 1319-1330.
81. Luppo, M.i., A. Hazarabedian, and J. Ovejero-García, Effects of delta
ferrite on hydrogen embrittlement of austenitic stainless steel welds.
Corrosion Science, 1999. 41(1): p. 87-103.
82. Örnek C., Xiangli Zhong, and D. Engelberg, Low-temperature
Environment Assisted Cracking of Grade 2205 Duplex Stainless Steel
beneath a MgCl2:FeCl3 Salt Droplet. Corrosion, 2015. in press.
83. Lo, I.H., Y. Fu, C.-J. Lin, and W.-T. Tsai, Effect of electrolyte composition
on the active-to-passive transition behavior of 2205 duplex stainless steel
in H2SO4/HCl solutions. Corrosion Science, 2006. 48(3): p. 696-708.
84. Yau, Y.H. and M.A. Streicher, The Effect of Chromium Content (0 to
35%) in Fe-Cr Alloys on Corrosion Rates and Mechanisms in 1.0 N
Sulfuric Acid. Corrosion, 1991. 47(5): p. 352-359.
85. Llewellyn, D.T., Steels: Metallurgy and Applications. First ed. (1992), UK:
Butterworth-Heinemann.
86. Trethewey K.R. and J. Chamberlain, Corrosion for science and
engineering. 2nd ed. 1995: Longman Scientific & Technical.
87. Sharland, S.M., A review of the theoretical modelling of crevice and
pitting corrosion. Corrosion Science, 1987. 27(3): p. 289-323.
88. Wranglen, G., Pitting and sulphide inclusions in steel. Corrosion Science,
1974. 14(5): p. 331-349.
89. Eklund, G.S., Initiation of Pitting at Sulfide Inclusions in Stainless Steel.
Journal of the Electrochemical Society, 1974. 121(4): p. 467-473.
90. Baker, M.A. and J.E. Castle, The initiation of pitting corrosion at MnS
inclusions. Corrosion Science, 1993. 34(4): p. 667-682.
91. Ke, R. and R. Alkire, Surface Analysis of Corrosion Pits Initiated at MnS
Inclusions in 304 Stainless Steel. Journal of the Electrochemical Society,
1992. 139(6): p. 1573-1580.
227
92. Ke, R. and R. Alkire, Initiation of Corrosion Pits at Inclusions on 304
Stainless Steel. Journal of The Electrochemical Society, 1995. 142(12):
p. 4056-4062.
93. Wijesinghe, T.L.S.L. and D.J. Blackwood, Real time pit initiation studies
on stainless steels: The effect of sulphide inclusions. Corrosion Science,
2007. 49(4): p. 1755-1764.
94. Smialowski, M., Z. Szklarska-Smialowska, M. Rychcik, and A. Szummer,
Effect of sulphide inclusions in a commercial stainless steel on the
nucleation of corrosion pits. Corrosion Science, 1969. 9(2): p. 123-125.
95. Zheng, S.J., Y.J. Wang, B. Zhang, Y.L. Zhu, C. Liu, P. Hu, and X.L. Ma,
Identification of MnCr2O4 nano-octahedron in catalysing pitting corrosion
of austenitic stainless steels. Acta Materialia, 2010. 58(15): p. 5070-
5085.
96. Sedriks, A.J., Plenary Lecture—1986: Effects of Alloy Composition and
Microstructure on the Passivity of Stainless Steels. Corrosion, 1986.
42(7): p. 376-389.
97. Castle, J.E. and R. Ke, Studies by auger spectroscopy of pit initiation at
the site of inclusions in stainless steel. Corrosion Science, 1990. 30(4–5):
p. 409-428.
98. Lott, S.E. and R.C. Alkire, The Role of Inclusions on Initiation of Crevice
Corrosion of Stainless Steel: I . Experimental Studies. Journal of The
Electrochemical Society, 1989. 136(4): p. 973-979.
99. Webb, E.G. and R.C. Alkire, Pit Initiation at Single Sulfide Inclusions in
Stainless Steel: I. Electrochemical Microcell Measurements. Journal of
the Electrochemical Society, 2002. 149(6): p. B272-B279.
100. Webb, E.G. and R.C. Alkire, Pit Initiation at Single Sulfide Inclusions in
Stainless Steel: II. Detection of Local pH, Sulfide, and Thiosulfate.
Journal of the Electrochemical Society, 2002. 149(6): p. B280-B285.
101. Webb, E.G. and R.C. Alkire, Pit Initiation at Single Sulfide Inclusions in
Stainless Steel: III. Mathematical Model. Journal of the Electrochemical
Society, 2002. 149(6): p. B286-B295.
102. Laycock, N.J., Effects of Temperature and Thiosulfate on Chloride Pitting
of Austenitic Stainless Steels. Corrosion, 1999. 55(6): p. 590-595.
103. Newman, R.C., W.P. Wong, H. Ezuber, and A. Garner, Pitting of
Stainless Steels by Thiosulfate Ions. Corrosion, 1989. 45(4): p. 282-287.
104. Marcus, P., A. Teissier, and J. Oudar, The influence of sulphur on the
dissolution and the passivation of a nickel-iron alloy—I. electrochemical
and radiotracer measurements. Corrosion Science, 1984. 24(4): p. 259-
268.
105. Marco, J.F., J.R. Gancedo, W. Meisel, P. Griesbach, and P. Gütlich,
Technical Note: Study of the Cl−-Induced Breakdown of the Passive
Layer on Steel. Corrosion, 1991. 47(7): p. 498-500.
106. Webb, E.G., T. Suter, and R.C. Alkire, Microelectrochemical
Measurements of the Dissolution of Single MnS Inclusions, and the
Prediction of the Critical Conditions for Pit Initiation on Stainless Steel.
Journal of the Electrochemical Society, 2001. 148(5): p. B186-B195.
228
107. Ghahari S.M., In Situ Synchrotron X-Ray Characterisation and Modelling
of Pitting Corrosion of Stainless Steel, [PhD thesis]. University of
Birmingham; 2012.
108. Manning, P.E., D.J. Duquette, and W.F. Savage, The Role of Sulfide
Inclusion Morphology in Pit Initiation of Several Type 300 Series
Stainless Steels. Corrosion, 1980. 36(6): p. 313-319.
109. Suter, T. and H. Böhni, A new microelectrochemical method to study pit
initiation on stainless steels. Electrochimica Acta, 1997. 42(20–22): p.
3275-3280.
110. Szklarska-Smialowska Z., Pitting Corrosion of Metals. 1986, Houston,
Texas: National Association of Corrosion Engineers.
111. Noh, J.S., N.J. Laycock, W. Gao, and D.B. Wells, Effects of nitric acid
passivation on the pitting resistance of 316 stainless steel. Corrosion
Science, 2000. 42(12): p. 2069-2084.
112. Muto, I., S. Kurokawa, and N. Hara, Microelectrochemistry on CrS and
MnS Inclusions and Its Relation with Pitting Potentials of Stainless
Steels. ECS Transactions, 2009. 16(52): p. 269-279.
113. Muto, I., D. Ito, and N. Hara, Microelectrochemical Investigation on Pit
Initiation at Sulfide and Oxide Inclusions in Type 304 Stainless Steel.
Journal of The Electrochemical Society, 2009. 156(2): p. C55-C61.
114. Burstein, G.T., P.C. Pistorius, and S.P. Mattin, The nucleation and
growth of corrosion pits on stainless steel. Corrosion Science, 1993.
35(1–4): p. 57-62.
115. Pistorius, P.C. and G.T. Burstein, Metastable Pitting Corrosion of
Stainless Steel and the Transition to Stability. Philosophical
Transactions: Physical Sciences and Engineering, 1992. 341(1662): p.
531-559.
116. Newman, R.C. and E.M. Franz, Growth and Repassivation of Single
Corrosion Pits in Stainless Steel. Corrosion, 1984. 40(7): p. 325-330.
117. Suleiman, M.I. and R.C. Newman, The use of very weak galvanostatic
polarization to study localized corrosion stability in stainless steel.
Corrosion Science, 1994. 36(9): p. 1657-1665.
118. Galvele, J.R., Transport Processes and the Mechanism of Pitting of
Metals. Journal of the Electrochemical Society, 1976. 123(4): p. 464-474.
119. Davenport, A.J., Second year corrosion lecture notes. 2015, University of
Birmingham: Birmingham.
120. Novakovski, V.M. and A.N. Sorokina, Model study of chloride pitting in
18-8 stainless steel. Corrosion Science, 1966. 6(5): p. 227-233.
121. Isaacs, H.S., The Behavior of Resistive Layers in the Localized
Corrosion of Stainless Steel. Journal of The Electrochemical Society,
1973. 120(11): p. 1456-1462.
122. Tester, J.W. and H.S. Isaacs, Diffusional Effects in Simulated Localized
Corrosion. Journal of The Electrochemical Society, 1975. 122(11): p.
1438-1445.
123. Ghahari, M., D. Krouse, N. Laycock, T. Rayment, C. Padovani, M.
Stampanoni, F. Marone, R. Mokso, and A.J. Davenport, Synchrotron X-
ray radiography studies of pitting corrosion of stainless steel: Extraction
of pit propagation parameters. Corrosion Science, 2015. 100: p. 23-35.
229
124. ASTM G46-94, Standard Guide for Examination and Evaluation of Pitting
Corrosion. 2005, ASTM International: West Conshohocken, PA, United
States.
125. Schwenk W., Theory Of Stainless Steel Pitting. Corrosion, 1964. 20(4):
p. 129t-137t.
126. Janik-Czachor M. and Z. Szklarska-Smialowska, Pitting corrosion of
single crystals of the Fe-16 Cr alloy in solutions containing Cl− ions.
Corrosion Science, 1968. 8(4): p. 215-220.
127. Pickering, H.W. and R.P. Frankenthal, On the Mechanism of Localized
Corrosion of Iron and Stainless Steel: I . Electrochemical Studies.
Journal of The Electrochemical Society, 1972. 119(10): p. 1297-1304.
128. Frankenthal, R.P. and H.W. Pickering, On the Mechanism of Localized
Corrosion of Iron and Stainless Steel: II . Morphological Studies. Journal
of The Electrochemical Society, 1972. 119(10): p. 1304-1310.
129. Li, S.X. and L.H. Hihara, Atmospheric corrosion initiation on steel from
predeposited NaCl salt particles in high humidity atmospheres. Corrosion
Engineering, Science and Technology, 2010. 45(1): p. 49-56.
130. Albores-Silva, O.E., E.A. Charles, and C. Padovani, Effect of chloride
deposition on stress corrosion cracking of 316L stainless steel used for
intermediate level radioactive waste containers. Corrosion Engineering
Science and Technology, 2011. 46(2): p. 124-128.
131. Cole, I.S., N.S. Azmat, A. Kanta, and M. Venkatraman, What really
controls the atmospheric corrosion of zinc? Effect of marine aerosols on
atmospheric corrosion of zinc. International Materials Reviews, 2009.
54(3): p. 117-133.
132. Cole, I.S., W.D. Ganther, D.A. Paterson, G.A. King, S.A. Furman, and D.
Lau, Holistic model for atmospheric corrosion: Part 2 - Experimental
measurement of deposition of marine salts in a number of long range
studies. Corrosion Engineering, Science and Technology, 2003. 38(4): p.
259-266.
133. Mendoza, A.R. and F. Corvo, Outdoor and indoor atmospheric corrosion
of carbon steel. Corrosion Science, 1999. 41(1): p. 75-86.
134. Schindelholz, E., B.E. Risteen, and R.G. Kelly, Effect of Relative
Humidity on Corrosion of Steel under Sea Salt Aerosol Proxies: I. NaCl.
Journal of The Electrochemical Society, 2014. 161(10): p. C450-C459.
135. Schindelholz, E., B.E. Risteen, and R.G. Kelly, Effect of Relative
Humidity on Corrosion of Steel under Sea Salt Aerosol Proxies: II.
MgCl2, Artificial Seawater. Journal of The Electrochemical Society, 2014.
161(10): p. C460-C470.
136. Tsuru, T., K.-I. Tamiya, and A. Nishikata, Formation and growth of micro-
droplets during the initial stage of atmospheric corrosion. Electrochimica
Acta, 2004. 49(17–18): p. 2709-2715.
137. Dubuisson, E., P. Lavie, F. Dalard, J.-P. Caire, and S. Szunerits, Study
of the atmospheric corrosion of galvanised steel in a micrometric
electrolytic droplet. Electrochemistry Communications, 2006. 8(6): p.
911-915.
230
138. Dubuisson, E., P. Lavie, F. Dalard, J.-P. Caire, and S. Szunerits,
Corrosion of galvanised steel under an electrolytic drop. Corrosion
Science, 2007. 49(2): p. 910-919.
139. Padovani, C., O. Albores-Silva, and A. Charles, Corrosion control of
stainless steels in indoor atmospheres – laboratory measurements under
MgCl2 deposits at constant relative humidity (Part 1). Corrosion, 2014.
140. Schindelholz E. and R.G. Kelly, Application of Inkjet Printing for
Depositing Salt Prior to Atmospheric Corrosion Testing. Electrochemical
and Solid State Letters 2010. 13(10): p. C29-C31.
141. Tsutsumi, Y., A. Nishikata, and T. Tsuru, Initial Stage of Pitting Corrosion
of Type 304 Stainless Steel under Thin Electrolyte Layers Containing
Chloride Ions. Journal of The Electrochemical Society, 2005. 152(9): p.
B358-B363.
142. Nishikata A, A.H. Nakamura, T.V. Nam, and Tada. E., Relative Humidity
for Onsets of Pitting Corrosion and Repassivation of Stainless Steels
under Wet-dry Cyclic Conditions Containing Chloride, in EUROCORR
2014. 2014: Pisa, Italy.
143. Matsumi M., A. Nishikata, and T. Tsuru., in Preceedings of the Japan
Conference on Material and Environments 46. Fukuoka, 1999: p. 13.
144. Tsutsumi, Y., A. Nishikata, and T. Tsuru, Monitoring of Rusting of
Stainless Steels in Marine Atmospheres Using Electrochemical
Impedance Technique. Journal of The Electrochemical Society, 2006.
153(7): p. B278-B282.
145. Cook, A.B., S.B. Lyon, N.P.C. Stevens, M. Gunther, G. McFiggans, R.C.
Newman, and D.L. Engelberg, Assessing the risk of under-deposit
chloride-induced stress corrosion cracking in austenitic stainless steel
nuclear waste containers. Corrosion Engineering, Science and
Technology, 2014. 49(6): p. 529-534.
146. Tada, E. and G.S. Frankel, Effects of Particulate Silica Coatings on
Localized Corrosion Behavior of AISI 304SS under Atmospheric
Corrosion Conditions. Journal of the Electrochemical Society, 2007.
154(6): p. C318-C325.
147. Chen, Z.Y. and R.G. Kelly, Computational Modeling of Bounding
Conditions for Pit Size on Stainless Steel in Atmospheric Environments.
Journal of the Electrochemical Society, 2010. 157(2): p. C69-C78.
148. Neufeld, A.K., I.S. Cole, A.M. Bond, and S.A. Furman, The initiation
mechanism of corrosion of zinc by sodium chloride particle deposition.
Corrosion Science, 2002. 44(3): p. 555-572.
149. Cole I.S. Lau D.; Paterson D.A., Holistic model for atmospheric
corrosion Part 6 – From wet aerosol to salt deposit. Corrosion
Engineering, Science and Technology, 2004. 39(3): p. 209-218.
150. Chen, Z.Y., D. Persson, F. Samie, S. Zakipour, and C. Leygraf, Effect of
Carbon Dioxide on Sodium Chloride-Induced Atmospheric Corrosion of
Copper. Journal of The Electrochemical Society, 2005. 152(12): p. B502-
B511.
151. Zhang, J., J. Wang, and Y. Wang, Electrochemical investigations of
micro-droplets formed on metals during the deliquescence of salt
231
particles in atmosphere. Electrochemistry Communications, 2005. 7(4):
p. 443-448.
152. Lobnig, R., J.D. Sinclair, M. Unger, and M. Stratmann, Mechanism of
Atmospheric Corrosion of Copper in the Presence of Ammonium Sulfate
Particles: Effect of Surface Particle Concentration. Journal of The
Electrochemical Society, 2003. 150(6): p. A835-A849.
153. Katayama, H., K. Noda, H. Masuda, M. Nagasawa, M. Itagaki, and K.
Watanabe, Corrosion simulation of carbon steels in atmospheric
environment. Corrosion Science, 2005. 47(10): p. 2599-2606.
154. Chen, Z.Y., S. Zakipour, D. Persson, and C. Leygraf, Effect of Sodium
Chloride Particles on the Atmospheric Corrosion of Pure Copper.
Corrosion, 2004. 60(5): p. 479-491.
155. Maier, B. and G.S. Frankel, Pitting Corrosion of Silica-Coated Type 304
Stainless Steel Under Thin Electrolyte Layers. Corrosion, 2011. 67(3): p.
035004-1-035004-10.
156. Zhang, J., J. Wang, and Y. Wang, Micro-Droplets Formation during the
Deliquescence of Salt Particles in Atmosphere. Corrosion, 2005. 61(12):
p. 1167-1172.
157. Stratmann, M. and H. Streckel, On the atmospheric corrosion of metals
which are covered with thin electrolyte layers—I. Verification of the
experimental technique. Corrosion Science, 1990. 30(6–7): p. 681-696.
158. Stratmann, M. and H. Streckel, On the atmospheric corrosion of metals
which are covered with thin electrolyte layers—II. Experimental results.
Corrosion Science, 1990. 30(6–7): p. 697-714.
159. Stratmann, M., H. Streckel, K.T. Kim, and S. Crockett, On the
atmospheric corrosion of metals which are covered with thin electrolyte
layers-iii. the measurement of polarisation curves on metal surfaces
which are covered by thin electrolyte layers. Corrosion Science, 1990.
30(6–7): p. 715-734.
160. Nishikata, A., Y. Ichihara, and T. Tsuru, An application of electrochemical
impedance spectroscopy to atmospheric corrosion study. Corrosion
Science, 1995. 37(6): p. 897-911.
161. Nishikata, A., Y. Ichihara, and T. Tsuru, Electrochemical impedance
spectroscopy of metals covered with a thin electrolyte layer.
Electrochimica Acta, 1996. 41(7–8): p. 1057-1062.
162. Winsley R.J., S.N.R., Reddy B., Rance A.P., Fennell PAH. 4 Meter Box
Monitoring Programme - Final Report for the Period 2007-2010. Serco
Report SERCO/TCS/006028.01/Issue 1. 2011 report to RWMD and
SERCO/TCS/006028.01/Issue001/FinalReport.
163. Padovani, C., R.J. Winsley, N.R. Smart, P.A.H. Fennell, C. Harris, and K.
Christie, Corrosion Control of Stainless Steels in Indoor Atmospheres—
Practical Experience (Part 2). Corrosion, 2015. 71(5): p. 646-666.
164. du Plessis, A., Studies on Atmospheric Corrosion Processes in AA2024,
[Ph.D. thesis]. Birmingham, UK:University of Birmingham; 2015.
165. ZHANG J.-b.; WANG J.; WANG Yan-hua, Phenomenon of Micro-droplets
Formation on Metals during the Deliquescence of Salt Particles in
Atmosphere. Acta Phys. Chim. Sin., 2005. 21(09): p. 993-996.
232
166. Bian, L., Y. Weng, and X. Li, Observation of micro-droplets on metal
surface in early atmospheric corrosion. Electrochemistry
Communications, 2005. 7(10): p. 1033-1038.
167. Morton, S.C. and G.S. Frankel, Atmospheric pitting corrosion of AA7075-
T6 under evaporating droplets with and without inhibitors. Materials and
Corrosion, 2014. 65(4): p. 351-361.
168. Li, J.F., B. Maier, and G.S. Frankel, Corrosion of an Al–Mg–Si alloy
under MgCl2 solution droplets. Corrosion Science, 2011. 53(6): p. 2142-
2151.
169. Wang, J., L. Liang, and J. Jiang, The role of electrochemical polarization
in micro-droplets formation. Electrochemistry Communications, 2008.
10(11): p. 1788-1791.
170. DiamondLightSource. How Diamond Works [cited 2015/ 12/018].
Available from: http://www.diamond.ac.uk/Home/About/How-Diamond-
Works.html.
171. Salvo, L., M. Suéry, A. Marmottant, N. Limodin, and D. Bernard, 3D
imaging in material science: Application of X-ray tomography. Comptes
Rendus Physique, 2010. 11(9–10): p. 641-649.
172. Stock, S.R., X-ray microtomography of materials. International Materials
Reviews, 1999. 44(4): p. 141-164.
173. Banhart J., Advanced Tomographic Methods in Materials Research and
Engineering. 2008, New York: Oxford University Press Inc.
174. http://www.chemicalghosts.org/research/synchrotron-science/. 2015
[cited 2015 01/12].
175. Ghahari, S.M., A.J. Davenport, T. Rayment, T. Suter, J.-P. Tinnes, C.
Padovani, J.A. Hammons, M. Stampanoni, F. Marone, and R. Mokso, In
situ synchrotron X-ray micro-tomography study of pitting corrosion in
stainless steel. Corrosion Science, 2011. 53(9): p. 2684-2687.
176. Marrow, T.J., L. Babout, A.P. Jivkov, P. Wood, D. Engelberg, N. Stevens,
P.J. Withers, and R.C. Newman, Three dimensional observations and
modelling of intergranular stress corrosion cracking in austenitic stainless
steel. Journal of Nuclear Materials, 2006. 352(1–3): p. 62-74.
177. Burnett, T.L., S.A. McDonald, A. Gholinia, R. Geurts, M. Janus, T. Slater,
S.J. Haigh, C. Ornek, F. Almuaili, D.L. Engelberg, G.E. Thompson, and
P.J. Withers, Correlative Tomography. Sci. Rep., 2014. 4.
178. Connolly, B.J., D.A. Horner, S.J. Fox, A.J. Davenport, C. Padovani, S.
Zhou, A. Turnbull, M. Preuss, N.P. Stevens, T.J. Marrow, J.Y. Buffiere,
E. Boller, A. Groso, and M. Stampanoni, X-ray microtomography studies
of localised corrosion and transitions to stress corrosion cracking.
Materials Science and Technology, 2006. 22(9): p. 1076-1085.
179. Knight, S.P., M. Salagaras, A.M. Wythe, F. De Carlo, A.J. Davenport,
and A.R. Trueman, In situ X-ray tomography of intergranular corrosion of
2024 and 7050 aluminium alloys. Corrosion Science, 2010. 52(12): p.
3855-3860.
180. Knight, S.P., M. Salagaras, and A.R. Trueman, The study of intergranular
corrosion in aircraft aluminium alloys using X-ray tomography. Corrosion
Science, 2011. 53(2): p. 727-734.
233
181. Davenport, A.J., C. Padovani, B.J. Connolly, N.P.C. Stevens, T.A.W.
Beale, A. Groso, and M. Stampanoni, Synchrotron X-Ray
Microtomography Study of the Role of Y in Corrosion of Magnesium Alloy
WE43. Electrochemical and Solid-State Letters, 2007. 10(2): p. C5-C8.
182. Xu, W., Synchrotron x-ray and electrochemical studies of pitting
corrosion of iron, [Ph.D. ]. University of Birmingham, UK: 2014.
183. Nagy, Z. and H. You, RADIOLYTIC EFFECTS ON THE IN-SITU
INVESTIGATION OF BURIED INTERFACES WITH SYNCHROTRON X-
RAY TECHNIQUES. Journal of Electroanalytical Chemistry, 1995. 381(1-
2): p. 275-279.
184. MIG WELD GmbH International. Schaeffler diagramm [cited 2016 07 23];
Available from: http://www.migweld.de/english/service/welding-stainless-
steels/schaeffler-diagram-without-filler-metal/.
185. Schindelin J, A.-C.I., Frise E, Kaynig V, Longair M, Pietzsch T, et al. Fiji:
an open-source platform for biological-image analysis. Nature Methods.
2012;9(7):676-82.
186. ASTM E104-02, Standard Practice for Maintaining Constant Relative
Humidity by Means of Aqueous Solutions. 2007, ASTM International:
West Conshohocken, PA, United States.
187. Vacher H.C. and C.J. Bechtoldt, Delta Ferrite-Austenite Reactions and
the Formation of Carbide, Sigma, and Chi Phases in 18 Chromium-8
Nickel-3.5 Molybdenum Steels. Journal of Research of the National
Bureau of Standards, 1954. 35(2): p. 67-76.
188. Ahmed, I.I., J.Q. da Fonseca, and A.H. Sherry, Effect of strain paths and
residual delta ferrite on the failure of cold rolled austenitic stainless
steels, type 304L. The Journal of Strain Analysis for Engineering Design,
2013. 48(7): p. 410-419.
189. Mesu J.G., B., A. M., De Groot, F. M. F.,Weckhuysen, B. M. , Observing
the influence of X-rays on aqueous copper solutions by in situ combined
video/XAFS/UV-Vis spectroscopy. 13th International Conference on X-
Ray Absorption Fine Structure (XAFS13); 2007; Stanford, CA: American
Institute of Physics.
190. Chiba, A., I. Muto, Y. Sugawara, and N. Hara, A Microelectrochemical
System for In Situ High-Resolution Optical Microscopy: Morphological
Characteristics of Pitting at MnS Inclusion in Stainless Steel. Journal of
The Electrochemical Society, 2012. 159(8): p. C341-C350.
191. Saluja R. and K.M. Moeed, METALLURGICAL BEHAVIOUR OF AISI
304 STEEL BUTT WELDS UNDER SEGREGATION. INTERNATIONAL
JOURNAL OF MECHANICAL ENGINEERING AND TECHNOLOGY,
2014. 5(2): p. 36-43.
192. Mumtaz, K., S. Takahashi, J. Echigoya, Y. Kamada, L.F. Zhang, H.
Kikuchi, K. Ara, and M. Sato, Magnetic measurements of martensitic
transformation in austenitic stainless steel after room temperature rolling.
Journal of Materials Science, 2004. 39(1): p. 85-97.
193. Nakada, N., H. Ito, Y. Matsuoka, T. Tsuchiyama, and S. Takaki,
Deformation-induced martensitic transformation behavior in cold-rolled
and cold-drawn type 316 stainless steels. Acta Materialia, 2010. 58(3): p.
895-903.
234
194. Naraghi, R., Martensitic Transformation in Austenitic Stainless Steels,
[Masters thesis]. Sweden:Royal Institute of Technology; Masters
Degree, 2009.
195. Zhang, M.X., P.M. Kelly, L.K. Bekessy, and J.D. Gates, Determination of
retained austenite using an X-ray texture goniometer. Materials
Characterization, 2000. 45(1): p. 39-49.
196. Lothongkum, G., P. Wongpanya, S. Morito, T. Furuhara, and T. Maki,
Effect of nitrogen on corrosion behavior of 28Cr–7Ni duplex and
microduplex stainless steels in air-saturated 3.5 wt% NaCl solution.
Corrosion Science, 2006. 48(1): p. 137-153.
197. Batista S.R.F. and S.E. Kuri;, Aspect of selective and pitting corrosion in
cast duplex stainless steel. Anti-Corrosion Methods and Materials, 2004.
51(3): p. 205–208.
198. Azmat, N.S., K.D. Ralston, T.H. Muster, B.C. Muddle, and I.S. Cole, A
High-Throughput Test Methodology for Atmospheric Corrosion Studies.
Electrochemical and Solid-State Letters, 2011. 14(6): p. C9-C11.
199. Wang, Y.H., Y.Y. Liu, W. Wang, L. Zhong, and J. Wang, Influences of the
three-phase boundary on the electrochemical corrosion characteristics of
carbon steel under droplets. Materials and Corrosion, 2013. 64(4): p.
309-313.
200. Jiang, J., J. Wang, Y.-h. Lu, and J.-z. Hu, Effect of length of
gas/liquid/solid three-phase boundary zone on cathodic and corrosion
behavior of metals. Electrochimica Acta, 2009. 54(5): p. 1426-1435.
201. Nishikata, A., Y. Ichihara, Y. Hayashi, and T. Tsuru, Influence of
Electrolyte Layer Thickness and pH on the Initial Stage of the
Atmospheric Corrosion of Iron. Journal of the Electrochemical Society,
1997. 144(4): p. 1244-1252.
202. Horton D. J., A. W. Zhu, J.R. Scully, and M. Neurock, Crystallographic
controlled dissolution and surface faceting in disordered face-centered
cubic FePd. MRS Communications 2014. 4: p. 113-119.
203. Krishnan S., J. Dumbre, S. Bhatt, Esther T. Akinlabi, and R.
Ramalingam, Effect of Crystallographic Orientation on the Pitting
Corrosion Resistance of Laser Surface Melted AISI 304L Austenitic
Stainless Steel. International Journal of Mechanical, Aerospace,
Industrial and Mechatronics Engineering, 2013. 7(4): p. 239-242.
204. Ravi Kumar, B., R. Singh, B. Mahato, P.K. De, N.R. Bandyopadhyay,
and D.K. Bhattacharya, Effect of texture on corrosion behavior of AISI
304L stainless steel. Materials Characterization, 2005. 54(2): p. 141-147.
205. Shahryari, A., J.A. Szpunar, and S. Omanovic, The influence of
crystallographic orientation distribution on 316LVM stainless steel pitting
behavior. Corrosion Science, 2009. 51(3): p. 677-682.
206. Kumar, B.R., A.K. Singh, S. Das, and D.K. Bhattacharya, Cold rolling
texture in AISI 304 stainless steel. Materials Science and Engineering: A,
2004. 364(1–2): p. 132-139.
207. Buck, W.R. and H. Leidheiser, The Corrosion of Single Crystals and
Recrystallized Single Crystals of Iron and Steel in Citric Acid. Journal of
The Electrochemical Society, 1957. 104(8): p. 474-481.
235
208. Cui, F.S., F.J. Presuel-Moreno, and R.G. Kelly, Computational modeling
of cathodic limitations on localized corrosion of wetted SS 316L at room
temperature. Corrosion Science, 2005. 47(12): p. 2987-3005.
209. Chen, Z.Y. and R.G. Kelly, An analytical modeling method for calculating
the maximum cathode current deliverable by a circular cathode under
atmospheric exposure, in Simulation of Electrochemical Processes II,
V.G. DeGiorgi, C.A. Brebbia, and R.A. Adey, Editors. 2007. p. 33-41.
210. Ernst, P. and R.C. Newman, Pit growth studies in stainless steel foils. I.
Introduction and pit growth kinetics. Corrosion Science, 2002. 44(5): p.
927-941.
211. Hunkeler, F., A. Krolikowski, and H. Böhni, A study of the solid salt film
on nickel and stainless steel. Electrochimica Acta, 1987. 32(4): p. 615-
620.
212. Newman, R.C., The dissolution and passivation kinetics of stainless
alloys containing molybdenum—1. Coulometric studies of Fe Cr and Fe
Cr Mo alloys. Corrosion Science, 1985. 25(5): p. 331-339.
213. Street, S.R., Personal Communication. 2015.
214. TOMASHOV N. D., Development of the Electrochemical Theory of
Metallic Corrosion. Corrosion, 1964. 20(1): p. 7t-14t.
215. Padovani, C., Overview of UK research on the durability of container
materials for radioactive wastes. Corrosion Engineering, Science and
Technology, 2014. 49(6): p. 402-409.
216. Muster, T.H., A. Bradbury, A. Trinchi, I.S. Cole, T. Markley, D. Lau, S.
Dligatch, A. Bendavid, and P. Martin, The atmospheric corrosion of zinc:
The effects of salt concentration, droplet size and droplet shape.
Electrochimica Acta, 2011. 56(4): p. 1866-1873.
217. Chen, J., J. Wang, E. Han, and W. Ke, In situ observation of formation
and spreading of micro-droplets on magnesium and its alloy under cyclic
wet–dry conditions. Corrosion Science, 2007. 49(3): p. 1625-1634.
218. Chen, Z.Y., D. Persson, A. Nazarov, S. Zakipour, D. Thierry, and C.
Leygraf In Situ Studies of the Effect of CO2 on the Initial NaCl-Induced
Atmospheric Corrosion of Copper. Journal of The Electrochemical
Society, 2005. 152(9): p. B342-B351.
219. Jung, R.-H., H. Tsuchiya, and S. Fujimoto, XPS characterization of
passive films formed on Type 304 stainless steel in humid atmosphere.
Corrosion Science, 2012. 58(0): p. 62-68.
220. Frankel, G., G. Thornton, S. Street, T. Rayment, D. Williams, A. Cook, A.
Davenport, S. Gibbon, D. Engelberg, C. Ornek, A. Mol, P. Marcus, D.
Shoesmith, C. Wren, K. Yliniemi, G. Williams, S. Lyon, R. Lindsay, T.
Hughes, J. Lutzenkirchen, S.T. Cheng, J. Scully, S.F. Lee, R. Newman,
C. Taylor, R. Springell, J. Mauzeroll, S. Virtanen, S. Heurtault, and J.
Sullivan, Localised corrosion: general discussion. Faraday Discussions,
2015. 180: p. 381-414.
221. Smart, N.R., R.J. Winsley, P.A.H. Fennell, B. Reddy, A.P. Rance, and C.
Padovani, Waste container durability: monitoring of a stainless steel 4
metre box over 12 years. Mineralogical Magazine, 2012. 76(8): p. 2891-
2899.
236
222. Mankowski, J. and Z. Szklarska-Smialowska, The effect of specimen
position on the shape of corrosion pits in an austenitic stainless steel.
Corrosion Science, 1977. 17(9): p. 725-735.
223. Harris, C., A Survey of Current Environmental Conditions Within ILW
Interim Stores and Potential GDF Analogues. 2015, Amec,
17391/TR/0007.
224. Brossia, C.S. and R.G. Kelly, Influence of Alloy Sulfur Content and Bulk
Electrolyte Composition on Crevice Corrosion Initiation of Austenitic
Stainless Steel. Corrosion, 1998. 54(2): p. 145-154.
225. UHLIG, H.H. and J.R. GILMAN, Pitting of 18-8 Stainless Steel in Ferric
Chloride Inhibited by Nitrates. Corrosion, 1964. 20(9): p. 289t-292t.
237