materials

Article
Influence of Oxygen Content in the Protective Gas on Pitting
Corrosion Resistance of a 316L Stainless Steel Weld Joint
Mohammad Maroufkhani 1, * , Soroosh Hakimian 1 , Alireza Khodabandeh 1 , Iulian Radu 2 , Lucas A. Hof 1
and Mohammad Jahazi 1, *

1 Department of Mechanical Engineering, École de Technologie Supérieure, Montreal, QC H3C 1K3, Canada;
[email protected] (S.H.); [email protected] (A.K.); [email protected] (L.A.H.)
2 PCL Industrial Constructors Inc., Edmonton, AB T6E 3P4, Canada; [email protected]
* Correspondence: [email protected] (M.M.); [email protected] (M.J.)

Abstract: Gas tungsten arc welding (GTAW) is commonly used for joining pipelines; however, it
often leads to discoloration in the heat-affected zone (HAZ). In this study, 316L pipes were welded
with different concentrations of oxygen present in the argon purge gas during welding. The objective
of this study was to investigate the effect of oxygen concentration in the protective gas on the pitting
corrosion resistance of welded pipes. The experimental results showed that the thickness of the oxide
layer formed in the HAZ depends on the concentration of oxygen in the protective gas. Increasing
the oxygen concentration in the protective gas resulted in an increase in pitting corrosion resistance
until a critical value, beyond which the resistance decreased. The results showed that the thickness
of the oxide layer formed in the HAZ depends on the concentration of oxygen in the protective gas.
Increasing the oxygen concentration in the protective gas increased the pitting corrosion resistance
until a critical value, beyond which the resistance decreased due to the formation of iron oxide. This
study provides valuable insights for improving the corrosion resistance of welded pipes in the oil
and gas industry.

Keywords: 316L SS; pitting corrosion; oil and gas pipeline; discoloration; oxygen concentration

Citation: Maroufkhani, M.;

Hakimian, S.; Khodabandeh, A.;
Radu, I.; Hof, L.A.; Jahazi, M. 1. Introduction
Influence of Oxygen Content in the
Although stainless steels (SS) are known for their good corrosion resistance, some
Protective Gas on Pitting Corrosion
industrial welded joints made of SS are prone to pitting corrosion. The presence of chloride
Resistance of a 316L Stainless Steel
ions can break down the passive film, leading to the formation of pits in both welded and
Weld Joint. Materials 2023, 16, 5968.
non-welded SS. Pitting corrosion is considered one of the most dangerous types of corrosion
https://doi.org/10.3390/ma16175968
due to its ability to cause localized damage and lead to sudden failures in equipment or
Academic Editor: Chih-Chun Hsieh structures [1,2].
Received: 2 August 2023
The welding process is a complex and intricate process that requires careful attention
Revised: 28 August 2023
to various parameters. These parameters have a significant impact on the quality and
Accepted: 29 August 2023 performance of welded joints. Among the critical parameters that need to be monitored
Published: 31 August 2023 and controlled are voltage, current, welding speed, shielding, and backing gas, as well
as pre-and post-weld heat treatments [3,4]. Therefore, these parameters must be carefully
monitored and adjusted to ensure the production of high-quality welded joints that meet
the required performance standards [5,6]. To this end, during welding, particularly for the
Copyright: © 2023 by the authors. root pass, it is necessary to remove air from the fusion zone by employing an inert purging
Licensee MDPI, Basel, Switzerland. gas on the back side of the weld (i.e., from the inside of the tube). It has been reported that
This article is an open access article the presence of oxygen at atmospheric concentration levels, i.e., about 210,000 ppm, can lead
distributed under the terms and to the formation of an oxide layer on the surface of the weld root [7]. Some authors have
conditions of the Creative Commons reported that this oxide layer can lead to contamination that results in reduced mechanical
Attribution (CC BY) license (https://
properties and lower corrosion resistance in the weldments. Specifically, the occurrence
creativecommons.org/licenses/by/
of pitting corrosion in the HAZ of welded austenitic SS joints has been reported by many
4.0/).

Materials 2023, 16, 5968. https://doi.org/10.3390/ma16175968 https://www.mdpi.com/journal/materials

Materials 2023, 16, 5968 2 of 18

authors [7–9]. To reduce and alleviate the susceptibility to pitting corrosion of welded
joints, Argon (Ar) has been used as a purging gas in the GTAW of various austenitic SS.
Although pure Ar would be the ideal option, its rising cost provides a strong incentive to
switch to a lower-purity option [7,10].
The typically studied corrosion parameters include the passive current density (ipassive ),
passive potential (Epassive ), pitting current density (ipit ), and pitting potential (Epit ). The
passive current density is defined as the current density required to maintain a passive film
on the metal surface, the Epassive value indicates the passive potential for a given metal,
which is the potential at which the metal surface becomes passive, causing a protective ox-
ide film to form, and the ipit and Epit values indicate the current density required to initiate
localized corrosion in the form of pits on the metal surface and the potential at which the
metal surface begins to undergo localized corrosion in the form of pits, respectively [11].
Ling et al. studied the oxidation behavior of 304L during GTAW and found that,
without inert gas shielding, the surface near the weld oxidized at a higher rate [12]. Specifi-
cally, they observed that the extent of oxidation increased as the distance to the fusion line
decreased, leading to the growth of an oxide film and cracks on the oxide film. The HAZ
region mainly consisted of the Fe-Cr oxides and the oxidation extent and severity increased
with the welding temperature [12]. Huang et al. found an outer layer of Cr2 O3 and Fe2 O3
and an inner layer of FeCr2 O4 in samples exposed at 400 ◦ C, while at 600 ◦ C, an oxide film
with a Fe-rich outer layer and a Cr-rich inner layer in 316L SS was observed [13]. According
to Xie et al., Fe and Cr oxide were detected in the oxide scale of 316L SS, while the high
vacuum-generated initial oxide layer consisted of a Fe3 O4 outer layer and an inner layer
enriched with both Fe and Cr [14].
In the discoloration area of the weld zone, a range of colors from straw yellow to blue,
purple, or even black is observed. The extent of colors is influenced by a combination
of factors, including oxidation and temperature changes during welding [15]. One of
the primary factors contributing to the colors observed in the HAZ is the formation of
oxide layers on the metal surface [15]. The thickness and composition of the oxide layer
can vary depending on factors such as the welding process, the atmosphere, the heat
input, the exposure time, and the specific metal being welded [15]. These oxide layers
interact with light, resulting in the selective absorption, reflection, and scattering of different
wavelengths, and leading to the observed colors [16]. Furthermore, the thickness of the
oxide layer also influences the observed color. For example, it has been reported that thicker
oxide layers can result in a broader range of colors, including blues, purples, and even
black, while thinner oxide layers appear as straw yellow or light shades [17].
Equation (1) illustrates the Ray optic concept, where an incident light ray (I) interacts
with the surface of an oxide layer. The incident ray splits into a reflected ray (R1) and
a transmitted ray (T1) that passes through the oxide layer. Upon reaching the oxide
layer/bulk material interface, the transmitted ray is reflected (RT1) and travels back through
the oxide layer before exiting into the air. The color that is observed depends on the optical
path difference (OPD) between the reflected ray (R1) and the transmitted-reflected ray (T
RT1). This path difference is determined by the refraction angle, which follows Snell’s law
(θ2), the thickness of the oxide layer (t), the refractive index of the oxide layer (n2), and the
wavelength of light (λ). The refractive index (n2) is wavelength-dependent, influencing the
interference pattern [16–18].

OPD = 2·(n2(λ)·t·cos(θ2)) (1)

In the case of oxide films, which could be composed by different layers of materials
with different refractive indices, interference effects can occur, as the thickness of the thin
film influences the optical path difference [16,17]. The optical path difference determines
the phase relationship between the interfering waves, which, in turn, affects the resulting
observed color. Depending on the film’s thickness and the wavelength of the incident light,
different colors will be determined [17].
Materials 2023, 16, 5968 3 of 18

The current study aims to fill existing gaps in the literature by investigating the
impact of oxygen content in the purging gas during GTAW on the discoloration levels of
the root joint area and the pitting corrosion resistance of AISI 316L SS. Specifically, this
research focuses on evaluating the pitting corrosion resistance of steel without removing the
discoloration. The present study aims to investigate the critical value of oxygen impurity in
argon backing gas, with a specific focus on its influence on the pitting corrosion resistance.
By identifying the optimum oxygen content for enhanced corrosion resistance, this research
contributes to the development of practical solutions for industries that are seeking both
effective welding outcomes and cost efficiency. The principal objective of this research is to
quantify the impact of variations in oxygen content on the pitting corrosion resistance of
the welded joint. Therefore, four different oxygen contents in backing gas were selected
for the analysis. A combination of microscopic and chemical analysis methods, as well
as thermodynamic modeling, were used to characterize the discoloration, then pitting
corrosion tests were carried out and the oxide layers’ nature and compositions were
analyzed. Then, the operating mechanisms are discussed. Finally, recommendations on
operational conditions are proposed.

2. Materials and Methods

The specimens used in this work were made of 316L SS with a chemical composition
(wt.%) as indicated in Table 1. The chemical composition was determined through the
utilization of spectrometry, a technique well-regarded for its precision in analyzing material
composition. The analysis process involved taking multiple measurements to ensure
accuracy and repeatability. In fact, the measurements were carried out five times to
establish the reliability of the results. To perform the chemical composition determination,
a Thermo Scientific ARLTM 4460 Optical Emission Spectrometer (Thermo Fisher Scientific
Inc., Waltham, MA, USA) was employed. In this study, 316L SS specimens were welded
using GTAW with varying concentrations of oxygen in the purging gas. The effect of oxygen
content on the discoloration levels and pitting corrosion resistance of the root welded
joints was investigated. The interior of the pipe was purged with argon gas containing
specific oxygen concentrations, ranging from 50 to 5000 ppm. The thermodynamic software
FactSage was used to predict the formation of possible phases based on oxygen content
and temperature. Potentiodynamic anodic polarization tests were conducted to determine
the pitting potential. Discoloration images were acquired using a 3D scanner, and the oxide
layer was analyzed using scanning electron microscopy (SEM) with the help of mounting
techniques and ion milling methods. The detailed methodology and analysis procedures
are described in the subsequent paragraphs.

Table 1. Chemical composition of 316L SS (wt %).

Element C Mn Si P S Mo Cr
Amount 0.02 0.97 0.43 0.28 0.001 2.05 16.47
Element Ni Al Co Cu W N Other
Amount 11.37 0.0162 0.22 0.25 0.073 0.093 0.13

The pipe diameter was 154.4 mm (6 in) and the thickness of the base metal was
7.11 mm (0.285 in). GTAW was used for welding the 316L SS pipe. A total of four (4) passes
were applied for welding to be as close as possible to the actual industrial conditions.
The joint geometry is shown in Figure 1. The welding parameters of each of the four
welding passes shown in Figure 1 are presented in Table 2. Before welding, the interior of
the pipe (i.e., the back of the weld) was purged with Ar gas containing a specific oxygen
content. Four different oxygen concentrations, namely 50, 200, 500, and 5000 ppm, were
used for the purging, as indicated in Table 3. The oxygen gas used for welding was
obtained from commercially available sources offering varying degrees of argon purity.
 Element C Mn Si P S Mo Cr
Amount 0.02 0.97 0.43 0.28 0.001 2.05 16.47
Element Ni Al Co Cu W N Other
Amount 11.37 0.0162 0.22 0.25 0.073 0.093 0.13
Materials 2023, 16, 5968 4 of 18
Table 2. Welding condition parameters of 316L SS samples.

Process GTAW GTAW GTAW GTAW

To ensure its suitability, the gas underwent meticulous analysis conducted by Airliquid, a
Pass distinguished company1 2
specializing in precise 3
gas assessments. Throughout the 4 welding
Shielding gas Ar Ar Ar
procedure, continuous monitoring of oxygen concentration was undertaken using the Ar
Purge gas flow rate (CFH) 40
PurgEye 300 instrument. 40 maintenance of40
The initiation and 40
gas purging was meticulously
Current (A) 125–202
executed until the 166–215
desired and predetermined oxygen200–223 208–243
level was successfully attained.
Voltage (V) 10.6–14 conducive11.8–15.7
To create an environment 12.8–16.4 the pipe was
to accurate measurements, 13.8–16.4
effectively
sealed from both ends.
Tube to work distance (mm) 9.52 One end of the pipe9.52 was subjected 9.52
to inert gas introduction,
9.52 while,
Heat In Put (kJ.mm−1)at the opposite end,
1.31the oxygen concentration
1.17 was carefully
1.28 gauged. The formation
1.44 of
Filer material possible phases as a function
ER316L (Exocor) of the oxygen content and temperature was predicted
ER316L (Exocor) ER316L (Exocor) ER316L (Exocor) using
the thermodynamic software FactSage [19].

Figure
Figure 1.
1. Weld
Weldzone
zonegeometry
geometry (dimensions
(dimensions in
in mm).
mm).

Table
Table 3.
2. Purging
Weldinggas concentrations.
condition parameters of 316L SS samples.
Oxygen Content (PPM) 50 200 500 5000
Process GTAW GTAW GTAW GTAW
Component Ar + 0.005% O Ar + 0.02% O Ar + 0.05% O Ar + 0.5% O
Pass 1 2 3 4
Shielding gas Ar
Potentiodynamic Ar
anodic polarization (PAP) was used Ar for determining theAr pitting po-
Purge gas flow rate (CFH) tential of the40HAZ. The tests were40 performed based on the 40 ASTM G5-13 standard 40 [20]. The
Current (A) size of the test specimens was 15
125–202 × 15 mm., with an area
166–215 of 1 cm² that was exposed
200–223 208–243 to the
solution and examined. Each sample was placed in a flask containing the test solution, 0.5
Voltage (V) 10.6–14 11.8–15.7 12.8–16.4 13.8–16.4
M NaCl and 0.5 M H2SO4, for the pitting corrosion tests. The open circuit potential (OCP)
Tube to work distance (mm) values, also9.52 9.52
referred to as the corrosion 9.52experiment was measured
potential, for each 9.52 with
− 1
Heat In Put (kJ.mm ) a delay of 300
1.31s.The polarization curve
1.17 was plotted from1.2850 mV below OCP to 20001.44 mV (vs.
Filer material ER316L (Exocor) ER316L (Exocor) ER316L (Exocor) ER316L (Exocor)

Table 3. Purging gas concentrations.

Oxygen Content (PPM) 50 200 500 5000

Component Ar + 0.005% O Ar + 0.02% O Ar + 0.05% O Ar + 0.5% O

FactSage is a comprehensive software package widely used in the field of computa-

tional thermochemistry. Originally designed for pyrometallurgical processing, FactSage has
evolved to encompass a wide range of applications. The Equilib module within FactSage
performs Gibbs energy minimization calculations, enabling the determination of chemical
species concentrations in systems reaching a state of chemical equilibrium. Additionally,
the phase diagram module offers the capability to calculate various types of phase diagram
sections, providing insights into phase transformations, stability ranges, and boundaries of
different phases. In this study, the Equilib and Phase Diagram modules of FactSage were
utilized to analyze the thermodynamic behavior of the steel [19].
Potentiodynamic anodic polarization (PAP) was used for determining the pitting
potential of the HAZ. The tests were performed based on the ASTM G5-13 standard [20].
The size of the test specimens was 15 × 15 mm., with an area of 1 cm2 that was exposed
to the solution and examined. Each sample was placed in a flask containing the test
solution, 0.5 M NaCl and 0.5 M H2 SO4 , for the pitting corrosion tests. The open circuit
Materials 2023, 16, 5968 5 of 18
Materials 2023, 16, x FOR PEER REVIEW 5 of 19

potential (OCP) values, also referred to as the corrosion potential, for each experiment
was measured
Ag/AgCl) with a delay
at a potential rateof
of300 s.The
0.0016 Vspolarization
−1, as suggested curveby was plotted from
the literature [21].50Potentiody-
mV below
OCP to 2000 mV (vs. Ag/AgCl) at a potential rate of 0.0016 Vs −1 , as suggested by the
namic polarization tests were repeated three times to determine the average pitting poten-
literature
tial (ipit) for[21].
eachPotentiodynamic
condition. The tests polarization tests were
were performed at repeated
23 ± 2 °Cthree times to
according to determine
the ASTM
the average pitting potential
G5-13 standard test method. pit (i ) for each condition. The tests were performed at 23 ± 2 ◦ C
accordingFor the to electrochemical
the ASTM G5-13polarization
standard test method.
tests, an electrochemical cell consisting of three
For the electrochemical polarization
electrodes was deployed. The 316L sample (working tests, an electrochemical
electrode (WE)) cell consisting of three
and the platinum
electrodes was deployed. The 316L sample (working electrode
counter electrode (CE) were immersed in the electrolyte and connected to a saturated (WE)) and the platinum
counter
Ag/AgClelectrodereference(CE) were(RE)
electrode immersed
by a saltinbridge.
the electrolyte
The sample andholder
connected to aholding
used for saturated
the
Ag/AgCl reference electrode (RE) by a salt bridge. The sample holder used for holding
316L sample (WE) was provided by Redox Company, (Norrköping, Sweden) and the elec-
the 316L sample (WE) was provided by Redox Company, (Norrköping, Sweden) and the
trochemical cell, CE, and RE were obtained from Pine Research Inc. The electrodes were
electrochemical cell, CE, and RE were obtained from Pine Research Inc. The electrodes were
connected to an Autolab potential/galvanostat PGSTAT302N potentiostat (Metrohm, FL,
connected to an Autolab potential/galvanostat PGSTAT302N potentiostat (Metrohm, FL,
USA), which was controlled by a computer using ANOVA software version 2.1.6
USA), which was controlled by a computer using ANOVA software version 2.1.6 (Metrohm
(Metrohm Autolab B.V, Utrecht, The Netherlands).
Autolab B.V, Utrecht, The Netherlands).
In order to determine the pitting parameters (Epit, ipit, ipassive, Epassive) in this study, the
In order to determine the pitting parameters (Epit , ipit , ipassive , Epassive ) in this study,
inflection point method was employed, as suggested by the literature [22–24]. The method
the inflection point method was employed, as suggested by the literature [22–24]. The
suggested by researchers involves identifying the point on the polarization curve where a
method suggested by researchers involves identifying the point on the polarization curve
change in slope occurs. It is characterized by the intersection of two tangent lines that rep-
where a change in slope occurs. It is characterized by the intersection of two tangent lines
resent the transition from activity to passivity (Epassive, ipassive) and from passivity to pitting
that represent the transition from activity to passivity (Epassive , ipassive ) and from passivity
(E
to pit , ipit). (E , i ).
pitting pit pit
Discoloration images
Discoloration images were
were acquired
acquired using
using aa Keyence
Keyence VR2500
VR2500 3D 3D scanner (KEYENCE
scanner (KEYENCE
CANADA INC,
CANADA INC, Mississauga,
Mississauga, ON ON L5N
L5N 0A4,
0A4, Canada)
Canada) to to examine
examine the the oxide
oxide layer.
layer. Subse-
Subse-
quently, the oxide layer was carefully analyzed using
quently, the oxide layer was carefully analyzed using scanning electron microscopy scanning electron microscopy
(SEM).
(SEM).
Two Two high-resolution
high-resolution SEM instruments,
SEM instruments, the Hitachi the Hitachi
SEM TM3000 SEM TM3000
and the and the SU8230
Hitachi Hitachi
SU8230 FE-SEM
FE-SEM (Hitachi (Hitachi
High-Tech High-Tech
Canada, Canada,
Toronto, Toronto,
ON, Canada),ON, Canada),
were employedwere employed for
for the anal-
the analysis.
ysis. To ensure Tothe
ensure the preservation
preservation of theoxide
of the delicate delicate oxide
layers, layers, a combination
a combination of mounting of
mounting techniques
techniques and cross-section
and cross-section and face ion and face ion
milling millingwere
methods methods were employed.
employed. Fig-
Figure 2 illus-
ure 2 illustrates
trates the specific the specific technique
technique that was developed
that was developed and employedand employed in this investiga-
in this investigation. This
tion. This developed method involved the use of hot mounting
developed method involved the use of hot mounting resin and cross-section and surface resin and cross-section and
surface ion milling, effectively protecting the integrity of the oxide
ion milling, effectively protecting the integrity of the oxide layers. For the cross-section layers. For the cross-
section
part, ionpart,
millingionwas
milling was conducted
conducted for 2 h at for
6 kV2 with
h at 6a kV
speedwith
of a60speed of 60 reciprocations
reciprocations per minute
per minute
and a swingand of ±a15swing of ±15
degrees. Thisdegrees. This was
was followed by followed
1 h of ionby 1 h ofation
milling milling
5 kV with aatspeed
5 kV
with
of 30 areciprocations
speed of 30 reciprocations
per minute and peraminute
swing of and±30a swing
degrees.of ±30
Fordegrees.
the surfaceForionthe milling,
surface
aion milling,
3 min a 3 min
process wasprocess
carriedwasoutcarried
at 5 kVout withat 515kV with 15 reciprocations
reciprocations per minute perandminute and
a swing
a swing of ±60 degrees. In order to prepare the sample surface
of ±60 degrees. In order to prepare the sample surface for analysis, the IM4000 Plus ion for analysis, the IM4000
Plus ionHitachi
milling millingmachine
Hitachiwas machine wasallowing
utilized, utilized,for allowing for precise
precise surface surface and
preparation preparation
optimal
and optimal
imaging imaging
quality. Figurequality. Figure
2 provides 2 providesrepresentation
a schematic a schematic representation
of a preparedofsample a prepared
after
sample
the after theprocess.
ion milling ion millingTheseprocess. Thesewere
techniques techniques
crucial were crucial in
in obtaining obtaining
accurate andaccurate
reliable
results
and reliablefor the analysis
results for of
thethe oxide layers.
analysis of the oxide layers.

Figure 2. Schematic showing the ion mill regions of the samples for accurate determination of the
thickness of the oxide layers and their composition.
Figure 2. Schematic showing the ion mill regions of the samples for accurate determination of the
thickness of the oxide layers and their composition.
Materials 2023, 16, 5968 6 of 18

3. Results
3.1. Influence of Oxygen Content on Discoloration
Figure 3 provides illustrative examples of the extent of discoloration that was observed
in the HAZ of the tested samples with different oxygen concentrations. Additionally, it
must be noted that the calculated widths reported in Table 4 are average values obtained
Materials 2023, 16, x FOR PEER REVIEW 6 of 19
from three samples taken from different locations along the circumference of the pipe. The
results indicate that the highest discoloration width occurs for 5000 ppm oxygen, whereas
the differences in width diminish as the oxygen content rises. Elevated oxygen levels can
3. Results
lead to higher oxidation rates, particularly at elevated temperatures. This means that the
3.1. at
rate Influence
which oftheOxygen
metalContent
reacts withon Discoloration
oxygen to form oxides increases. Consequently, the
Figure 3width
discoloration provides illustrative
expands as moreexamples
metal reacts,of the extent
forming anof discoloration
oxide layer [25]. that
This was
findingob-
served inthat
suggests the the
HAZ of the tested
presence samples
of oxygen with
plays different
a crucial oxygen
role in theconcentrations.
formation and Addition-
extent of
ally, it must be
discoloration noted
in the thatThese
HAZ. the calculated widthsindicate
findings, which reportedvariations
in Table 4inare theaverage values
discoloration
obtained
width from
based onthree samples
different takencontents,
oxygen from different
will be locations
utilizedalong
in thethesubsequent
circumference of the
sections
pipe.
of thisThe
paperresults indicate
to deepen our that the highest discoloration
understanding of the relationship width occurs
between foroxygen
5000 ppm oxy-
content
gen,discoloration,
and whereas the differences
and its impactin width diminish
on pitting as the oxygen
corrosion. content rises.
As discussed Elevated
in Section 1, oxy-
the
colors of the
gen levels canoxide
leadlayers correspond
to higher oxidationtorates,
different oxide thicknesses,
particularly at elevated and each welding
temperatures. This
condition
means that exhibits a specific
the rate at whichdiscoloration pattern.
the metal reacts withTherefore,
oxygen toitformis crucial
oxides toincreases.
determineCon- the
width of discoloration
sequently, for each
the discoloration welding
width condition
expands as more to gain
metal comprehensive
reacts, forming insights
an oxideintolayer
the
corrosion behavior.
[25]. This finding suggests that the presence of oxygen plays a crucial role in the formation
and extent of discoloration in the HAZ. These findings, which indicate variations in the dis-
3.2. Thermodynamic
coloration width basedCalculation
on different oxygen contents, will be utilized in the subsequent sec-
tionsFigure
of this 4paper
presents the oxide
to deepen scale formation
our understanding at different
of the temperatures
relationship between oxygen in 316L SS,
content
calculated using FactSage,
and discoloration, and represents
and its impact on pittingthe differentAs
corrosion. zones of the in
discussed HAZ, ranging
Section 1, thefrom
colorsthe
of
weld pool to the base metal. The type and structure at various temperatures
the oxide layers correspond to different oxide thicknesses, and each welding condition exhib- and different
partial pressures
its a specific of oxygenpattern.
discoloration indicateTherefore,
the amount it isofcrucial
oxygen to available
determinefrom the airof
the width todiscolora-
the base
material.
tion for each welding condition to gain comprehensive insights into the corrosion behavior.

(a)

(b)
Figure 3. Cont.
 Materials 2023,16,
Materials2023, 16,5968
x FOR PEER REVIEW 77 of
of 18
19

Materials 2023, 16, x FOR PEER REVIEW 7 of 19

(c)

(d)
Figure 3. Discoloration zone with different oxygen concentrations in purging gas: (a) 50 ppm, (b)
200 ppm, (c) 500 ppm, and (d) 5000 ppm.

3.2. Thermodynamic Calculation

Figure 4 presents the oxide scale formation at different temperatures in 316L SS, cal-
culated using FactSage, and represents (d)the different zones of the HAZ, ranging from the
weld pool to the base metal. The type and structure at various temperatures and different
Figurepressures
Figure
partial 3. Discoloration
3. Discoloration zone
zone
of oxygen with
withdifferent
indicate oxygen
different
the amount concentrations
oxygen
of in purging
concentrations
oxygen fromgas:
in purging
available (a) 50
thegas:
air ppm,
to(a)the (b)
50base
ppm,
200200
(b) ppm, (c) (c)
ppm, 500500
ppm, and
ppm, (d)(d)
and 5000 ppm.
5000 ppm.
material.
3.2. Thermodynamic Calculation
Figure 4 presents the oxide scale formation at different temperatures in 316L SS, cal-
culated using FactSage, and represents the different zones of the HAZ, ranging from the
weld pool to the base metal. The type and structure at various temperatures and different
partial pressures of oxygen indicate the amount of oxygen available from the air to the base
material.

Figure 4. Oxide scale formation at different temperatures in 316L SS. Yellow zones: Oxide structure
Figure 4. Oxide scale formation at different temperatures in 316L SS. Yellow zones: Oxide struc-
without Liquid oxide phase, Pink zones: Oxide structure with Oxide liquid phase, Green zones:
ture without Liquid oxide phase, Pink zones: Oxide structure with Oxide liquid phase, Green
Oxide structure with liquid phase and Oxide liquid phase.
zones: Oxide structure with liquid phase and Oxide liquid phase.

The above graph indicates that the oxide layer contains Spinel, Olivine, and Corun-
dum with different concentrations until around 1300 °C, when delta ferrite begins to form
(yellow zone). At around 1300 °C, the formation of oxide liquid begins to increase with the
higher partial pressure of oxygen, as shown in the pink zones. Between 1423 °C and 1445
°C, the material experiences a mushy zone consisting of delta ferrite, austenite, and liquid
metal, after which the base metal starts to melt at temperatures above 1445 °C.
Materials 2023, 16, 5968 8 of 18

Table 4. Components in different zones.

Zone Phase
1 Liquid
2 Liquid + MeS
3 Liquid + alfa Ca2 SiO4
4 Liquid + Oxide Liquid
5 Liquid + Oxide Liquid + Spinel
6 Liquid + Oxide Liquid + Corundum + SiO2 + TSpinel
7 Delta ferrite + Austenite + MeS
8 Austenite + MeS
9 Oxide Liquid + Austenite + Spinel + TSpinel
10 Austenite + MeS + Corundum + SiO2 + TSpinel + Ca3 Cr2Si3 O12
11 Austenite + MeS + Wollastonite
12 Austenite + MeS
13 Austenite + MeS + Corundum + SiO2 + TSpinel + Wollastonite
14 Spinel + Austenite + Olivine + Beta Ni3 S2
15 Spinel + Olivine + Rhodonite + Pyrrhotite
16 Oxide Liquid + Spinel
17 Spinel + Corundum + SiO2 + MnSO4 + CaSO4
18 Austenite + MeS + Spinel + Rhodonite + TSpinel + Wollastonite
19 Spinel + Olivine + Rhodonite + MnSO4 + CaSO4
20 Spinel + Corundum + Olivine + CaSO4 + MnSO4
21 Spinel + Corundum + SiO2 + MnSO4 + CaSO4 + MnO2 + CrO3
22 Spinel + Olivine
23 Delta ferrite + Austenite + MeS + Liquid
24 Delta ferrite + MeS + Liquid

The above graph indicates that the oxide layer contains Spinel, Olivine, and Corundum
with different concentrations until around 1300 ◦ C, when delta ferrite begins to form (yellow
zone). At around 1300 ◦ C, the formation of oxide liquid begins to increase with the higher
partial pressure of oxygen, as shown in the pink zones. Between 1423 ◦ C and 1445 ◦ C, the
material experiences a mushy zone consisting of delta ferrite, austenite, and liquid metal,
after which the base metal starts to melt at temperatures above 1445 ◦ C.
In Figure 4, the mass fractions of different phases at 800 ◦ C under different partial
pressures of oxygen are shown using the Equilib module of FactSage software version
8.2 (Thermfact Ltd., Mount Royal, QC, Canada). The figure indicates the presence of the
Corundum phase for two different ranges of partial pressures of oxygen. The first layer
of Corundum, which contains chromium, nickel, and oxygen, is present between 10−31
and 10−27 atm, while the second layer, which is the outer layer of the oxide and begins
at 10−9 atm, contains Cr, Fe, Ni, and oxygen (Figure 5c). The other phase observed after
10−25 atm is the Spinel phase consisting of Ni, Fe, Cr, and oxygen. In the Spinel phase,
chromium is found to be more stable until around a 10−12 atm partial pressure of oxygen,
followed by iron becoming more stable. Nickel is formed at 10−15 atm and becomes more
stable than chromium after 10−9 atm (Figure 5b).
 Materials
als 2023, 16, 16, 5968
2023, REVIEW
x FOR PEER 9 of 19 9 of 18

(a)

(b)

(c)
Figure 5. (a) Mass fractions
Figure of different
5. (a) Mass phases
fractions at 800 °C
of different as a function
phases at 800 ◦ Cofas
partial pressure
a function of oxy-
of partial pressure of oxygen,
gen, (b) Mole fraction of elements in the Spinel phase, (c) Mole fraction of elements in the Corun-
(b) Mole fraction of elements in the Spinel phase, (c) Mole fraction of elements in the Corundum phase.
dum phase.
Materials 2023, 16, x FOR PEER REVIEW 10 of 19

Materials 2023, 16, 5968 10 of 18

Oxide formation is initiated at a higher partial pressure of oxygen, specifically at a

−17
level of 10 atm, when the temperature reaches 1400 °C (Figure 6). This indicates that the
Oxide formation is initiated at a higher partial pressure of oxygen, specifically at a
oxide formation process is more favorable at higher temperatures compared to 800 °C.
level of 10−17 atm, when the temperature reaches 1400 ◦ C (Figure 6). This indicates
−17
that
The amount
the oxide of the Corundum
formation phase
process is more becomesatless
favorable andtemperatures
higher appears between 10 to
compared atm ◦ C.
800and
−13
The amount of the Corundum phaseafter
becomes lesspressure
and appears between −
−16 17
10 atm atm
10 atm. Oxide liquid starts to form a partial of oxygen of 10 in and
the
10 −13 atm. Oxide liquid starts to form after a partial pressure of oxygen of 10−16 atm in the
outer layer of oxide, but it is less stable than the Spinel phase.
outer layer of oxide, but it is less stable than the Spinel phase.

Figure 6. The structure of oxide formation at 1400 ◦ C in 316L SS for various partial pressures
Figure 6. The structure of oxide formation at 1400 °C in 316L SS for various partial pressures of
of oxygen.
oxygen.

In Figure 7a, the phases present at 1440 ◦ C and different partial pressures of oxygen
In Figure 7a, the phases present at 1440 °C and different partial pressures of oxygen
are displayed. Compared to 1400 ◦ C, the outer layer contains a higher amount of oxide
are displayed. Compared to 1400 °C, the outer layer contains a higher amount of oxide
liquid. Figure 5b provides information on the phases present in the oxide liquid at different
liquid. Figure 5b provides information on the phases present in−the oxide liquid at differ-
partial pressures of oxygen. The most stable phase between 10 21 atm −21
to 10−9 atm−9
is FeO.
ent
Thepartial
inner andpressures of oxygen.
outer layers of theThe
oxidemost stable
liquid phase
phase between
contain SiO210 atm phase
, a stable to 10 atatm is
these
FeO. The inner
pressures. and outer
In addition, layers
there of theamounts
are small oxide liquid
of Fe2phase
O3 andcontain SiO2, ainstable
NiO present phase
the outer at
layer
these pressures.
and CrO In addition,
in the inner there are
layer. These small of
analyses amounts of Fe
the oxide 2O3 and NiO
composition andpresent
phaseintransitions
the outer
layer andvaluable
provide CrO in the innerinto
insights layer.
theThese
behavioranalyses
of theofmaterial
the oxide composition
under and phase tran-
different temperature and
sitions provide valuable insights into the behavior of the material under different
oxygen partial pressure conditions. Indeed, the presence of Spinel, Olivine, and Corundum tempera-
ture and with
phases, oxygen partialproportions,
varying pressure conditions.
indicates Indeed, the presence
the complex nature ofofthe
Spinel,
oxideOlivine,
layer that andis
Corundum
formed at the phases,
surfacewith varying
of the proportions,
material. indicates
The identification ofthe complex
specific phases nature of thestability
and their oxide
layer
underthat is formed
different at the surface
conditions allows offorthe material.
a better The identification
understanding of specificofphases
of the structure the phasesand
their stabilitydiscolored
in different under different
zones.conditions allowssections
The following for a better understanding
delve of theofstructure
into the impacts specific
of the phases
phases on theincorrosion
differentresistance
discoloredofzones. The following sections delve into the impacts
the material.
of specific phases on the corrosion resistance of the material.
3.3. Oxide Geometrical Characteristics
By increasing the oxygen content in the backing gas, the growth of the oxide layer was
observed. Figure 8 illustrates such a variation in the oxide layer thickness with the oxygen
content for a given use case. The image was taken at a distance of 1.5 ± 0.3 mm from the
fusion line, where the thickness of the oxide layer increases from 700 ± 300 nm for 50 ppm
oxygen to 2.7 ± 0.3 µm, 2.8 ± 0.3 µm, and 3 ± 0.3 µm for 200, 500, and 5000 ppm oxygen,
respectively. It must be noted that other images were taken at other distances from the
weld line, and they all showed the same trend.
2023, 16, x Materials
FOR PEER 2023, 16, 5968
REVIEW 11 of 19 11 of 18

(a)

(b)
Figure 7. (a) MassFigure
fraction7. of
(a)different phasesofatdifferent
Mass fraction 1440 °C phases at 1440of◦ C
as a function partial pressureofofpartial
as a function oxy- pressure of oxygen.
gen. (b) Component(b) distribution
Component in the oxide liquid
distribution in the phase.
oxide liquid phase.

3.3. Oxide GeometricalIn order to comprehensively analyze the kinetics of oxide growth in the weld zone of
Characteristics
316L SS, it is essential
By increasing the oxygen contenttoin
consider the combined
the backing gas, the effects
growthofof oxygen content
the oxide in the atmosphere
layer
and the temperature. To incorporate the impact of the oxygen
was observed. Figure 8 illustrates such a variation in the oxide layer thickness with the content, we can introduce
a term which represents the partial pressure of oxygen
oxygen content for a given use case. The image was taken at a distance of O (P ).
1.5 Assuming
2 ± 0.3 mm
a power-law
from the fusion line, where the thickness of the oxide layer increases from 700 ± 300 nm the equation
relationship between the rate constant and the partial pressure of oxygen,
becomes
for 50 ppm oxygen to 2.7 [26]:
± 0.3 µm, 2.8 ± 0.3 µm, and 3 ± 0.3 µm for 200,
n 500, and 5000 ppm
oxygen, respectively. It must be noted that other images × POtaken
K = Awere 2 at other distances (2)
from the weld line, and they
In this all showed
equation, A isthe same trend.
a constant and n denotes the power-law exponent that describes
In order to comprehensively
the dependence ofanalyze
the ratethe kinetics
constant onofthe
oxide growth
oxygen in the
partial weld zone of
pressure.
316L SS, it is essential to consider the combined effects of oxygen content
Next, we can consider the Arrhenius equation, which relates in the atmos-
the rate constant to
phere and the temperature.
temperatureTo incorporate the impact of the oxygen content, we can intro-
[27]:
Ea
duce a term which represents the partial pressure of oxygen
K = A0(𝑃 ×𝑂e2).− Assuming
RT a power-law (3)
relationship between the rate constant and the partial pressure of oxygen, the equation
Using the two equations, we get:
becomes [26]:
Ea
𝐾 𝐴 𝑃 Kα PO2 n , e− RT


(2) (4)
 Materials 2023, 16, 5968 12 of 18

ials 2023, 16, x FOR PEER REVIEW 12 of 19

In this equation, assuming a linear relationship between the oxide thickness and time
(Wagner equation), X = Kt:
In this equation, A is a constant and n denotes theXα power-law
n − Ea
PO2 , eexponent
RT
that describes (5)
the dependence of the rate constant on the oxygen partial pressure.
Wherethe
Next, we can consider t isArrhenius
the time and X is the
equation, oxide
which thickness.
relates the rateThis equation
constant shows that the oxide
to tem-
perature [27]: thickness also increases with the oxygen content and temperature.
Figure 9 shows the elemental map of the oxide layer at a distance of 1.7 mm from
the fusion zone in theK sample A′ ewith 5000 ppm oxygen content. The (3)image reveals that
the inner
Using the two layerwe
equations, contains
get: iron, Cr, and Mn, which are Spinel and Corundum phases based
on Figures 5–7. The outer layer consists of iron oxide, which is a mixture of the outer
Spinel with an iron-rich Kα area,
P , e the second layer of Corundum.(4)
and Moreover, there is a
porous structure that appears in the 5000 ppm sample near the fusion line, possibly an
In this equation, assuming a linear relationship between the oxide thickness and time
oxide liquid with an iron oxide structure [28,29]. These porous structures are shown in
(Wagner equation), X = Kt:
Figure 10. The findings presented in this section highlight the influence of oxygen content
on the oxide layer’s thickness,
Xα P ,composition,
e and morphology. This (5) knowledge is crucial
to understanding pitting corrosion behavior. The quantitative data and visual evidence
Where t isprovided
the time and
in theX isfigures
the oxide thickness.
above serve asThis equation shows
a foundation that the
for further oxide and discussions in
analysis
thickness also increases with the oxygen content
the subsequent section of the paper. and temperature.

(a)

(b)
Figure 8. Cont.
als 2023, 16,Materials
x FOR PEER2023,REVIEW
16, 5968 13 of 19 13 of 18

(c)

(d)
Figure
Figure 8. Oxide layer at 8. Oxide layer
a distance of 1.5atmm
a distance
from theof 1.5 mm
fusion line:from
(a) 50the fusion
ppm, line:
(b) 200 (a) 50 ppm, (b) 200 ppm,
ppm,
(c) 500 ppm, and(c)
(d) 5000
500 ppm, ppm.
and (d) 5000 ppm.

Figure 9 shows the elemental

3.4. Pitting Corrosionmap of the oxide layer at a distance of 1.7 mm from the
Tests
fusion zone in the sample with 5000 ppm oxygen content. The image reveals that the inner
The
layer contains iron, Cr, andpolarization
Mn, which are curves
Spineland
andcorrosion
Corundumparameters
phases based foronthe samples with different
Figures
5–7. The outer layer consists of iron oxide, which is a mixture of the outer Spinel with in
oxygen contents are presented in Table 5 and Figure 11. As shown anTable 5, the passive
current densities for all samples initially increased with the
iron-rich area, and the second layer of Corundum. Moreover, there is a porous structure increase in oxygen content up
that appears in the 5000 ppm sample near the fusion line, possibly an oxide liquid withcan be attributed to
to 500 ppm, followed by a decrease in the sample with 5000 ppm. This
the porous
an iron oxide structure structure
[28,29]. observed
These porous in the oxide
structures layerinnear
are shown the10.
Figure fusion line of the sample with
The find-
5000 ppm, which may have led to the formation of more active
ings presented in this section highlight the influence of oxygen content on the oxide layer’s sites for corrosion. For the
thickness, composition, and morphology. This knowledge is crucial to understanding pitting to be 463 ± 10 µA.
sample with 50 ppm of oxygen, the passive current was determined
This The
corrosion behavior. value indicates the
quantitative datarate of visual
and spontaneous
evidenceoxidation
providedprocesses occurring on the surface,
in the figures
above serve as asuggesting
foundation aformoderate level ofand
further analysis corrosion resistance.
discussions As the oxygen
in the subsequent content increased to
section
of the paper. 200 ppm, the passive current decreased to 210 ± 10 µA. This reduction in passive current
density suggests an improvement in the corrosion resistance of the material, as a more
protective passive film formed on the surface. Further increasing the oxygen content to
500 ppm led to a slightly higher passive current of 230 ± 20 µA. Although the passive
current density increased slightly, it still indicates a relatively good corrosion resistance
compared to the sample with 50 ppm. However, in the sample with 5000 ppm of oxygen,
the passive current exhibited a further increase to 290 ± 30 µA. This unexpected rise in the
Materials 2023, 16, 5968 14 of 18

passive current density suggests a slight decrease in the corrosion resistance compared to
the samples with lower oxygen contents. The presence of a porous oxide structure near the
Materials 2023, 16, x FOR PEER REVIEW fusion line, as observed in the composition analysis, contributed to this14 phenomenon
of 19 by
creating more active sites for corrosion [30].

(a)

(b) (c)

(d) (e)

(f)
Figure 9. (a)
Figure 9. Oxide scalescale
(a) Oxide at 1.7at
mm1.7distance from fusion
mm distance from line of 5000
fusion ppm
line of oxygen
5000 ppm content
oxygen content and
and elemental map: (b) O, (c) Fe, (d) Cr, (e) Ni, (f) Mn.
elemental map: (b) O, (c) Fe, (d) Cr, (e) Ni, (f) Mn.
Materials 2023,16,
Materials 2023, 16,5968
x FOR PEER REVIEW 15 of 18
15 19

Figure 10.
Figure 10. SEM
SEM image
imageofofthe
thesurface
surfaceofofthe
thesample with
sample 5000
with ppm
5000 oxygen
ppm content,
oxygen showing
content, the the
showing
presence of porous structures near the fusion line.
presence of porous structures near the fusion line.

3.4. Pitting
Table Corrosion
5. Corrosion Tests of 316L SS with varying oxygen content.
parameters
The polarization curves and corrosion parameters for the samples with different ox-
Oxygen Content (PPM) 50 200 500 5000 316L SS Base
ygen contents are presented in Table 5 and Figure 11. As shown in Table 5, the passive
currentipassive
densities 463 ± 10 initially
(µA) for all samples 210 ± 10
increased 230
with± 20 290 ±in
the increase 30oxygen198 ± 30 up
content
to 500 ppm,i (µA) 463 ± 10 200 ± 10 210 ± 10
pit followed by a decrease in the sample with 5000 ppm. This can be attributed 260 ± 20 297 ± 30 to
the porous
Epassivestructure
(V) observed
0 ± 0.01in the0.05oxide
± 0.03layer near
0.07 ±the0.03 fusion0.09line of the −
± 0.01 sample with
0.04 ± 0.01
5000 ppm, which
Epit (V) may have led
0.42 ± 0.06to the formation
0.53 ± 0.01 of more
0.54 ± 0.01 active sites for
0.49 ± 0.02 corrosion.
0.96 ± 0.01 the
For
sample with 50 ppm of oxygen, the passive current was determined to be 463 ± 10 µA. This
valueInindicates
terms ofthe therate
pitofcurrent densities,
spontaneous as shown
oxidation in Figure
processes 11, the
occurring on pit current sug-
the surface, was
measured to be 463 ±
gesting a moderate 10 µA.
level This can resistance.
of corrosion be attributed As tothethe limited
oxygen availability
content of oxygen,
increased to 200
which
ppm, theimpedespassive thecurrent
formation of a fullytoprotective
decreased 210 ± 10 passive
µA. This film. The lower
reduction oxygen content
in passive current
restricts the extentanofimprovement
density suggests passivation, rendering the material
in the corrosion resistance moreofsusceptible
the material, to as
localized
a more
corrosion. As the oxygen
protective passive film formed content increased
on the surface.toFurther
200 ppm, the pit the
increasing current
oxygendecreased
content to to
200 ± 10 µA. Further increasing the oxygen content to
500 ppm led to a slightly higher passive current of 230 ± 20 µA. Although the passive500 ppm resulted in a slightly
higher
currentpit current
density of 210 ±slightly,
increased 30 µA. This canindicates
it still be attributed to the formation
a relatively good corrosionof a more stable
resistance
and protective
compared to thepassive
sample film on 50
with theppm.
material’s
However,surface. Thesample
in the presence withof 5000
oxygenppm promotes
of oxy-
the
gen,development
the passive current of this exhibited
film, enhancing
a further theincrease
material’s resistance
to 290 ± 30 µA. toThis
localized corrosion.
unexpected rise
Interestingly,
in the passivefor samples
current with 5000
density ppm aofslight
suggests oxygen, the pitin
decrease current density did
the corrosion not decrease
resistance com-
as anticipated
pared to the samples but remained
with lower slightly
oxygen elevated.
contents.ThisThe behavior
presencecan of abe attributed
porous oxideto the
struc-
presence of a porous oxide structure near the fusion line. The porous
ture near the fusion line, as observed in the composition analysis, contributed to this phe- nature of the oxide
layer
nomenoncreates by preferential
creating more sites for localized
active corrosion [30].
sites for corrosion initiation, contributing to the higher
pit current density.
In terms of the pit current densities, as shown in Figure 11, the pit current was
The passive
measured to be 463 potential
± 10 µA. values
This canare bereported
attributedin Table
to the5,limited
showing zero for the
availability sample
of oxygen,
with 50 ppm oxygen, indicating the absence of a stable passive
which impedes the formation of a fully protective passive film. The lower oxygen content film. In contrast, the
passive potential value increases to 0.05 V for the sample with 200
restricts the extent of passivation, rendering the material more susceptible to localized cor-ppm oxygen and then to
0.07 V and 0.09 V for the two other samples with 500 and 5000
rosion. As the oxygen content increased to 200 ppm, the pit current decreased to 200 ± 10 ppm oxygen, respectively.
The
µA. obtained results indicate
Further increasing that, by
the oxygen increasing
content to 500theppm oxygen
resultedcontent, the formation
in a slightly higher pitof
the passive film is accelerated. In the base metal, where there are no oxides present, the
current of 210 ± 30 µA. This can be attributed to the formation of a more stable and pro-
passive potential is lower when compared to the samples with discoloration. At 50 ppm
tective passive film on the material’s surface. The presence of oxygen promotes the devel-
of oxygen, the pitting potential was found to be 0.42 V. This relatively higher pit potential
opment of this film, enhancing the material’s resistance to localized corrosion. Interest-
suggests a greater susceptibility to localized corrosion. The pitting potential of the base
ingly, for samples with 5000 ppm of oxygen, the pit current density did not decrease as
metal was significantly higher compared to the sample with discoloration, which measures
anticipated but remained slightly elevated. This behavior can be attributed to the presence
at 0.96 V. This disparity can be attributed to the heat and temperature experienced during
of a porous oxide structure near the fusion line. The porous nature of the oxide layer cre-
welding in the heat-affected zone. The limited availability of oxygen at this concentration
ates preferential sites for localized corrosion initiation, contributing to the higher pit cur-
may hinder the formation of a robust passive film, making the material more vulnerable
rent density.
 Materials 2023, 16, 5968 16 of 18

to localized attack. As the oxygen concentration increased to 200 and 500 ppm, the pit
potential increased to 0.53 and 0.54 V, respectively. This indicates a slight improvement in
the resistance to localized corrosion. The higher pit potentials suggest the formation of a
more stable and protective passive film, which provides better defense against localized
corrosion initiation and propagation. In samples with 5000 ppm of oxygen, the pitting
potential decreased to 0.49 V. This decrease in pit potential can be attributed to the presence
of a porous iron oxide structure near the fusion line, as observed in previous analyses.
s 2023, 16, x FOR PEER REVIEW The porous nature of the oxide layer promotes the formation of localized16 of 19 corrosion sites,
resulting in a reduction in the pit potential. This implies a higher susceptibility to localized
corrosion, potentially compromising the integrity of the material.

(a)

(b)
Figure 11. (a) Polarization
Figure 11.curve of 316L SS curve
(a) Polarization with varying
of 316L oxygen
SS withcontent,
varying(b) high content,
oxygen magnification of magnification of
(b) high
the red area in the plot (a).
the red area in the plot (a).

4. parameters
Table 5. Corrosion Conclusionsof 316L SS with varying oxygen content.
This research study presents an investigation into the corrosion behavior and the
Oxygen Con-
50 of oxygen 200
influence 500
content on the properties of a 5000 316L
welded 316L SS SS Base in the root joint
pipeline
tent (PPM)
ipassive (µA) 463 ± 10 210 ± 10 230 ± 20 290 ± 30 198 ± 30
ipit (µA) 463 ± 10 200 ± 10 210 ± 10 260 ± 20 297 ± 30
Epassive (V) 0 ± 0.01 0.05 ± 0.03 0.07 ± 0.03 0.09 ± 0.01 −0.04 ± 0.01
Materials 2023, 16, 5968 17 of 18

area. By exploring the microstructural characteristics, polarization curves, and corrosion

parameters of the studied material, valuable insights are gained into the material’s corrosion
resistance. The investigation presented in this work yields the following results:
1. The width of discoloration increased with increasing oxygen content, and notable differ-
ences in color were observed between the samples with 5000 ppm and 500 ppm/200 ppm
of oxygen. In particular, the 5000 ppm sample exhibited an opaque zone located
behind the welding line, indicating distinct characteristics in the oxide formation
process at high oxygen concentrations.
2. The results indicate that the oxide thickness increases with an increase in the oxygen
content of the purging gas, when it ranges from 50 to 5000 ppm. This suggests that a
higher concentration of oxygen promotes the formation of a thicker oxide layer on
the surface of the material. The observed trend in oxide thickness provides valuable
insights into the influence of oxygen content on the oxide formation process and its
subsequent impact on the corrosion resistance of the material.
3. The oxide structure was found to be a mixture of Spinel and Corundum phases, with
the presence of an oxide liquid phase at high temperatures. The oxide liquid phase
exhibited a higher concentration of FeO, which resulted in a porous structure.
4. The pitting corrosion resistance showed an increasing trend up to a 500 ppm oxygen
content, followed by a decrease at 5000 ppm. The sample with 500 ppm exhibited
the highest pit potential, suggesting the formation of a more stable passive film that
is less susceptible to localized corrosion. The presence of a porous oxide structure
near the fusion line in the 5000 ppm sample may have contributed to its lower pitting
corrosion resistance.

Author Contributions: Conceptualization, M.M., I.R., L.A.H., M.J. and S.H.; methodology, M.M.,
S.H., A.K. and M.J.; software, M.M.; formal analysis, M.J.; investigation, M.M., S.H. and L.A.H.;
resources, M.M. and I.R.; data curation, S.H.; writing—original draft preparation, M.M.; writing—
review and editing, M.M., A.K., L.A.H. and M.J.; visualization, M.M.; supervision, A.K., I.R. and M.J.
All authors have read and agreed to the published version of the manuscript.
Funding: The research project presented in this article was conducted in collaboration with PCL
Construction in Edmonton, who provided the materials for the study, and was funded by The Natural
Sciences and Engineering Research Council of Canada (NSERC-Grant number RDC 542299-19).
Institutional Review Board Statement: Not applicable.
Informed Consent Statement: Not applicable.
Acknowledgments: The author would like to express gratitude to PCL Construction for their support
and contribution to this project. Special thanks to Mohammad Saadati for his invaluable assistance
for electron microscopy examinations.
Conflicts of Interest: The authors declare no conflict of interest.

References
1. Revie, R.W. Uhlig’s Corrosion Handbook; John Wiley & Sons: Hoboken, NJ, USA, 2011; Volume 51.
2. Dai, P.; Wang, Y.; Li, S.; Lu, S.; Feng, G.; Deng, D. FEM analysis of residual stress induced by repair welding in SUS304 stainless
steel pipe butt-welded joint. J. Manuf. Process. 2020, 58, 975–983. [CrossRef]
3. Paulraj, P.; Garg, R. Effect of welding parameters on pitting behavior of GTAW of DSS and super DSS weldments. Eng. Sci.
Technol. Int. J. 2016, 19, 1076–1083. [CrossRef]
4. Yang, Y.; Wang, Z.; Tan, H.; Hong, J.; Jiang, Y.; Jiang, L.; Li, J. Effect of a brief post-weld heat treatment on the microstructure
evolution and pitting corrosion of laser beam welded UNS S31803 duplex stainless steel. Corros. Sci. 2012, 65, 472–480. [CrossRef]
5. Chuaiphan, W.; Srijaroenpramong, L. Effect of welding speed on microstructures, mechanical properties and corrosion behavior
of GTA-welded AISI 201 stainless steel sheets. J. Mater. Process. Technol. 2014, 214, 402–408. [CrossRef]
6. Mohammed, G.R.; Ishak, M.; Aqida, S.N.; Abdulhadi, H.A. Effects of heat input on microstructure, corrosion and mechanical
characteristics of welded austenitic and duplex stainless steels: A review. Metals 2017, 7, 39. [CrossRef]
7. Bergquist, E.-L.; Huhtala, T.; Karlsson, L. The effect of purging gas on 308L TIG root pass ferrite content. Weld. World 2011, 55,
57–64. [CrossRef]
8. Taban, E.; Kaluc, E.; Aykan, T. Effect of the purging gas on properties of 304H GTA welds. Weld. J. 2014, 93, 124s–130s.
Materials 2023, 16, 5968 18 of 18

9. Shirwaikar, C.; Reddy, G. Purging with nitrogen in the welding of austenitic stainless steels. Weld. J. 1975, 12s–15s. Available
online: http://files.aws.org/wj/supplement/WJ_1975_01_s12.pdf (accessed on 2 August 2023).
10. Trigwell, S.; Selvaduray, G. Effects of welding on the passive oxide film of electropolished 316L stainless steel. J. Mater. Process.
Technol. 2005, 166, 30–43. [CrossRef]
11. Perez, N. Corrosivity and passivity. In Electrochemistry and Corrosion Science; Springer: Cham, Switzerland, 2016; pp. 199–264.
[CrossRef]
12. Ling, L.; Liu, T.; Lu, Y.; Guo, P. Investigation of the oxides film on 304L base metal produced during welding process without
inert gas shielding. Appl. Surf. Sci. 2019, 465, 780–786. [CrossRef]
13. Huang, X.; Xiao, K.; Fang, X.; Xiong, Z.; Wei, L.; Zhu, P.; Li, X. Oxidation behavior of 316L austenitic stainless steel in high
temperature air with long-term exposure. Mater. Res. Express 2020, 7, 066517. [CrossRef]
14. Xie, B.; Sun, M.; Xu, B.; Wang, C.; Jiang, H.; Li, D.; Li, Y. Oxidation of stainless steel in vacuum and evolution of surface oxide
scales during hot-compression bonding. Corros. Sci. 2019, 147, 41–52. [CrossRef]
15. Kimbrel, K. Determining Acceptable Levels of Weld Discoloration on Mechanically Polished and Electropolished Stainless Steel
Surface. Pharm. Eng. 2011, 31, 1–7.
16. Macleod, H.A. Thin-Film Optical Filters; CRC Press: Boca Raton, FL, USA, 2010.
17. Pedrotti, L.S. Basic physical optics. Fundam. Photonics 2008, 1, 152–154. [CrossRef]
18. Khafaji, N.Y.; Demir, A.G.; Vitali, L.; Fustinoni, D.; Niro, A.; Previtali, B.; Taha, Z.A. Optical characterization of laser coloured
titanium under different processing atmospheres. Surf. Coat. Technol. 2017, 321, 156–163. [CrossRef]
19. Bale, C.; Bélisle, E.; Chartrand, P.; Decterov, S.; Eriksson, G.; Hack, K.; Jung, I.-H.; Kang, Y.-B.; Melançon, J.; Pelton, A.; et al.
FactSage thermochemical software and databases. Calphad 2002, 26, 189–228. [CrossRef]
20. ASTM G5-14; Standard Reference Test Method for Making Potentiodynamic Anodic Polarization Measurements. ASTM: West
Conshohocken, PA, USA, 2014.
21. Kumar, S.; Shahi, A.S.; Sharma, V.; Malhotra, D. Effect of welding heat input and post-weld thermal aging on the sensitization
and pitting corrosion behavior of AISI 304L stainless steel butt welds. J. Mater. Eng. Perform. 2021, 30, 1619–1640. [CrossRef]
22. Bellezze, T.; Giuliani, G.; Roventi, G. Study of stainless steels corrosion in a strong acid mixture. Part 1: Cyclic potentiodynamic
polarization curves examined by means of an analytical method. Corros. Sci. 2018, 130, 113–125. [CrossRef]
23. Comotti, I.M.; Trueba, M.; Trasatti, S.P. The pit transition potential in the repassivation of aluminium alloys. Surf. Interface Anal.
2013, 45, 1575–1584. [CrossRef]
24. Finšgar, M.; Milošev, I. Corrosion behaviour of stainless steels in aqueous solutions of methanesulfonic acid. Corros. Sci. 2010, 52,
2430–2438. [CrossRef]
25. AWS D18.2:2020; Guide to Weld Discoloration Levels on Inside of Austenitic Stainless Steel Tube. American Welding Society:
Doral, FL, USA, 2020.
26. Geffroy, P.-M.; Guironnet, L.; Bouwmeester, H.; Chartier, T.; Grenier, J.-C.; Bassat, J.-M. Influence of oxygen partial pressure on the
oxygen diffusion and surface exchange coefficients in mixed conductors. J. Eur. Ceram. Soc. 2019, 39, 59–65. [CrossRef]
27. Laidler, K.J. The development of the Arrhenius equation. J. Chem. Educ. 1984, 61, 494. [CrossRef]
28. Li, J.; Li, H.; Zhao, G.; Zhou, C.; Ma, L.; Liu, H. Insights into the High-Temperature Oxidation Behavior of Austenitic Stainless
Steel and Influence of Copper on It. J. Mater. Eng. Perform. 2020, 29, 3661–3669. [CrossRef]
29. Habib, K.A.; Damra, M.S.; Saura, J.J.; Cervera, I.; Bellés, J. Breakdown and evolution of the protective oxide scales of AISI 304 and
AISI 316 stainless steels under high-temperature oxidation. Int. J. Corros. 2011, 2011, 824676. [CrossRef]
30. Prado, L.H.; Anastasiou, E.; Virtanen, S. Corrosion Behavior of Anodic Self-Ordered Porous Oxide Layers on Stainless Steel. J.
Electrochem. Soc. 2021, 168, 021507. [CrossRef]

Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual
author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to
people or property resulting from any ideas, methods, instructions or products referred to in the content.