Materials Today Communications 29 (2021) 102948

Contents lists available at ScienceDirect

Materials Today Communications

journal homepage: www.elsevier.com/locate/mtcomm

Microstructure and corrosion characterization of weld metal in stainless

steel and low carbon steel joint under different heat input
Yanqin Huang a, Jankang Huang a, *, Jianxiao Zhang b, c, Xiaoquan Yu a, Qi Li b, Zhen Wang b,
Ding Fan b, *
a
State Key Laboratory of Advanced Processing and Recycling of Non-ferrous Metal, Lanzhou University of Technology, Lanzhou 730050, China
b
School of Materials Science and Engineering, Lanzhou University of Technology, Lanzhou 730050, China
c
Lanzhou LS Heavy Equipment Co., LTD., Lanzhou 730050, China

A R T I C L E I N F O A B S T R A C T

Keywords: In this study, a novel method of the arc assisted tungsten insert gas (AA-TIG) backing welding and the submerged
Stainless steel/low carbon steel arc welding (SAW) to fill the groove was used to join 12Cr2Mo1R low carbon steel and S30408 stainless steel.
Dissimilar steel welding The weld metal microstructure characteristics of different heat input and its microhardness and corrosion
Microstructure
resistance were investigated in detail. The optical microscope (OM) and scanning electron microscope (SEM) was
Corrosion resistance
used to analyze the microstructure and composition distribution of weld zone. The cyclic potentiodynamic po­
larization and electrochemical impedance spectroscopy were employed to determine corrosion resistance of weld
zone under different heat input. The results show that the combination welding process of AA-TIG backing
welding and SAW can effectively join thick plate with high efficiency, and the weld shape is good without
incomplete fusion and penetration. A higher ratio between ferrite and austenite under higher heat input,
resulting in finer microstructure with increased hardness. Further, discovered by discussing the corrosion
mechanism is due to the high content of Ni and C elements in the weld are likely to form Ni-C phase, which
become the inducing point of pitting corrosion. It is recommended to use a higher heat input to obtain the best
corrosion resistance of the welded area.

1. Introduction become a challenge. Therefore, the realization of efficient welding be­

tween low carbon steel and stainless steel in petrochemical and natural
Cr-Mo steel, especially 12Cr2Mo1R low carbon steel, has a high gas industries is significant. However, these industries have a variety of
impact strength which often used to make shell of hydrogenation reactor environmental conditions, and the installed steel parts and facilities will
[1,2]. However, when Cr-Mo steel works in the temperature range of fail rapidly under the influence of these environmental conditions. For
325–575 ℃ for a long time, the toughness of the material will decrease the sack of meeting the needs of actual production, the research on the
and temper brittleness will occur. In order to avoid temper embrittle­ welding problems of stainless steel and low carbon steel has been paid
ment of shell material due to hydrogen corrosion, a layer of more and more attention in engineering application [8].
corrosion-resistant stainless steel is welded on the inner wall to solve the However, the failure occurrence of dissimilar welded joint of low
problem. Austenitic stainless steel has strong corrosion resistance, good carbon steel and stainless steel in application of hydrogenation reactor is
ductility, weldability and other excellent mechanical properties, which the main reason to reduce its service life. The heterogeneity across the
has become the stainless steel grade with the most applications [3–7]. It weldments and the appearance of pore and crack are common factors to
is employed to manufacture the inner wall of hydrogenation reactor. It is reduce the mechanical performances in welded joints [9–11], which
indicated that the material of hydrogenation reactor is welded structural have a closed relationship with the formation of δ ferrite in HAZ, un­
parts of low carbon steel and stainless steel. The demand for dissimilar mixed zone formation, carbon diffusion, composition difference. Among
metal welded structural parts of austenitic stainless steel and low carbon them, carbon diffusion plays a significant role in the failure of dissimilar
steel in petrochemical and natural gas industries is also growing. How to steel joint. The carbon migrates from low carbon steel to weld zone in
realize efficient production of low carbon steel and stainless steel has the heating process, which leads to the formation of C-depleted soft area

* Corresponding authors.
E-mail addresses: [email protected] (J. Huang), [email protected] (D. Fan).

https://doi.org/10.1016/j.mtcomm.2021.102948
Received 12 May 2021; Received in revised form 26 October 2021; Accepted 26 October 2021
Available online 30 October 2021
2352-4928/© 2021 Elsevier Ltd. All rights reserved.
Y. Huang et al. Materials Today Communications 29 (2021) 102948

Table 1
Chemical constitutions of base metal and welding wire(wt%).
C Si Mn P S Cr Ni Mo Cu Ti Fe

ENiCrFe-3 0.02 0.28 3.22 0.004 0.003 19.6 72.2 — — — 1.75

ER308L 0.028 060 0.36 0.004 0.003 20.15 12.83 — — — Bal
S30408 ≤0.08 ≤1.0 1.2 ≤2.0 ≤0.03 18 8.0 — — — Bal
12Cr2Mo1R 0.14 0.04 0.5 0.005 0.002 2.44 0.11 1.02 0.2 0.01 Bal

Fig. 1. Schematic of V-shape groove.

near the low carbon steel side and the C- enriched region near weld zone uniformity of dissimilar steel welded joint, and the overall enhancement
[12]. To inhibit the carbon migration in dissimilar steel joint, S.Sirohi et. of mechanical properties for welded joint.
al [11] employed the Ni-based wire as filler material to join SS304H and The weld joint of Cr-Mo steel and stainless steel is the major part of
P91 by GTAW. They found ERNiCrMo-3 filler improved microstructure hydrogenation reactor, which works in strong corrosive conditions. The

Fig. 2. Schematic of experimental setup.

2
Y. Huang et al. Materials Today Communications 29 (2021) 102948

Fig. 3. The welded plates prepared for different heat inputs, (a) 95 J/cm, (b) 80 J/cm, (c) 66 J/cm.

corrosion resistance is also a significantly factor affect failure in dis­

Table 2
similar steel welded joint. The carbon diffusion in the weld might cause
Parameters of SAW welding.
corrosion damage and even failure of welded joints, and they have
serious corrosion in salt and acid environment [13]. The research on the Weld speed Weld current Weld voltage gas flow (L/ Heat input
(cm/min) (A) (V) min) (J/cm)
corrosion of dissimilar welded joints will provide a deep understanding
to realize a high-quality weld joint. Zhang et al. [14] conducted a 35 145 23 11 95
42 145 23 11 80
corrosion study of dissimilar welded joints between high-strength steel
50 145 23 11 66
and austenitic stainless steel. According to reports, the martensite in the
weld zone has weaker corrosion resistance than the austenite in the
mixed zone. Vermad. J et al. [15] studied the corrosion failure of welded
joint of chromium series stainless steel and 316 L dissimilar steel in a
deteriorating environment. They propounded that there are alloying
elements with low synergy in the welding seam, which results in the
welding seam has lower corrosion resistance than base metal. Okonkwo
Bo et al. [16] studied the galvanic corrosion damage mechanism of A508
low alloy steel and 309 L/308 L surfacing layer dissimilar joints in
aqueous solution containing sulfate impurities. It was reported that the
pitting corrosion and galvanic corrosion caused by MnS occurred
simultaneously in the weld.
Although there are many studies on the corrosion performance of
dissimilar steel joints, they are limited to the study of thin plate welding.
For thick plate welding, there are few researches on the effect of heat
input on microstructure and corrosion resistance of dissimilar welded
joints. The goal in this work focuses on the effect of heat input on
microstructure and corrosion characterization of dissimilar joint of steel
of S30408 stainless steel and 12Cr2Mo1R low carbon steel joined by Arc
assisted active argon tungsten arc welding (AA-TIG) and submerged arc
welding (SAW) processes.

2. Experimental details Fig. 4. Macro-scale morphology weld joint cross section.

The heat-treated 12Cr2Mo1R low carbon steel and S30408 stainless

welding speeds is:
steel with a dimension was 450 mm × 150 mm× 45 mm as base metals.
Before welding, nickel base welding rod (ENiCrFe-3, diameter is 2 mm) I × U × 60
Q= (1)
was used to build-up welding a transition layer on 12Cr2Mo1R low v
carbon steel side by shielded metal arc welding (SMAW) (v:155 mm/
min; I:155 A; U:23 V; Q≤14.4 kJ/cm). The thickness of nickel base alloy Where, I is welding current; U is welding voltage; Q is heat input;v is
transition layer was 10 mm, and then annealed at 690 ℃ for 32 h to welding speed. The welding energy of different welding speed are
eliminate residual stress. After heat treatment, the dissimilar steel was summarized in Table 2:
welded by AA-TIG (v: 60 mm/min; I: I(Main arc)= 165 A; I(Auxiliary arc)= 60 After welding, the dimension of 65 mm × 45 mm× 10 mm samples
A; U: 60 V; Q≤ 22.5 kJ/cm)backing and submerged arc welding. ER308L were cut along the weld direction by wire electrical discharge cutting
welding wire with 5 mm diameter is applied as filler material for SAW. machine, then ground and polished. The microstructure was corroded
The chemical composition of base metal, ENiCrFe-3 filler rod, and by regia. Scanning electronic microscope (SEM) and optical microscopy
ER308L welding wire were itemized in Table 1. The V-shape groove was (OM) were used to study microstructure, and Energy dispersive spec­
machined on each base metal before the welding process to increase the trometer (EDS) was applied to obtain element distribution. DHv-1000
penetration and reduce the dilution rate of the filler metal, as shown Vickers hardness tester was used for hardness test. The test load was
Fig. 1. The groove angle, root gap and root height of 65◦ , 10 mm, 8 mm, 200 g and the holding time was 15 s. The CHI760E workstation with
respectively. Acetone was used to clean all impurities on base metal conventional three electrode electrolyzer was applied to test the corro­
surface before welding. Fig. 2 displays the schematic of experimental sion behaviour of the welded joint. The electrochemical corrosion
setup. Fig. 3 shows the welded plates prepared for different heat inputs. sample was 25 mm × 25 mm× 2 mm extracted from the weld zone, and
It can be observed that all welded plates with different heat input shows the schematic obtained from weld zone as displayed in Fig. 1. Then
continuous formation without obvious surface defects. The mathemat­ polished with sandpaper and then packed with epoxy resin. The elec­
ical calculation formula for the heat input of SWA welding at different trochemical corrosion solution was 3.5 wt% NaCl solution, which tested

3
Y. Huang et al. Materials Today Communications 29 (2021) 102948

Fig. 5. Base metal microstructure: (a) S30408 stainless steel, (b) 12Cr2Mo1R steel.

Fig. 6. Microstructure of weld zone; (a) 95 J/cm, (b) 80 J/cm, (c) 66 J/cm.

corrosion resistance at 30 ℃ and scanning speed of 1.667 mA/s. Fig. 6 depicts the OM iamgs of microstructure from weld metal zone
Fig. 4 displays the cross section of weld joint after welding process. for different samples. The weld metal microstructure is composed by
There is an obvious fusion line between the welding area and the base dendritic and cellular. Due to the weld metal is Ni-based alloy, which
material. This indicates available heat input is enough to completely leads to the segregation of Ni. It is confirmed by the EDS mapping re­
melt the base metal and electrode. sults. The weld zone microstructure includes austenite and ferrite
structure. The dendrite is austenite, while the ferrite phase is formed at
3. Results dendritic boundary, as shown in Fig. 6(a~c). The weld microstructure is
mainly skeletal shape ferrite and austenite structure composing of
3.1. Microstructure analysis columnar dendrites and cellular at the heat input is 95 J/cm. Some
wormlike ferrite particles can be observed in Fig. 6(b, c). As presented in
Fig. 5 illustrates the microstructure of S30408 steel and 12Cr2Mo1R Fig. 6, the lower heat input, the higher austenite content. As report of
steel, respectively. The microstructure of S30408 austenitic stainless Waynereitz [17], Ni element, as an austenite promotive element, has
steel is composed of austenite grain and twin crystal structure. There is more content at low heat input, which causes more ferrite transform into
almost no ferrite distribution on the grain boundary and no trace of austenite, resulting in the change of ferrite austenite ratio. On the other
carbide (see Fig. 5(a)). The 12Cr2Mo1R shows a typical low carbon steel hand, all ferrite are distributed at the austenite boundary. This micro­
microstructure, which includes ferrite phase and bainite phase, as shown structure could be attributed the transfer of Cr element between outside
in Fig. 4(b). and inside of dendrite. The central of ferrite dendrite is rich in Cr content

4
Y. Huang et al. Materials Today Communications 29 (2021) 102948

Fig. 7. SEM image of selected location and EDS spectrum of map analysis.

resulting in element segregation and enrichment. The Fe element con­

Table 3
tent of 80 J/cm heat input is lower than other heat input, which might
EDS surface analysis quantitative test results (wt%).
be caused by the dissolution of FeC phase.
Heat input (J/cm) C Cr Mn Fe Ni Fig. 8(a~c) shows the optical micrographs of austenite steel/weld
95 2.80 20.58 3.47 4.52 68.62 zone interface of three heat input. The HAZ region of S30408 steel
80 4.57 20.94 3.26 2.40 68.82 composed of the coarse polygonised austenite grains and ferrite struc­
66 3.45 20.24 3.41 3.34 69.56
ture and a small of twins. Fig. 8 reveals the structure of ferrite and
austenite grows perpendicular to vertical direction of the fusion line,
at the commence of solidification, while the outer part has lower Cr and the ferrite grows along the austenite grain boundary in a lath form.
content due to the lower temperature of outer. Then, the outside of the The phenomenon of epitaxial growth will generate during the remelting
dendrite with less chromium transforms into austenite during cooling, process when the workpiece is a multiphase material [19]. During the
leaving a Cr rich δ-ferrite skeleton in the dendrite core. Austenite grows solidification process, when the grains are perpendicular to the bound­
epitaxially at the boundary of fusion, while the nuclear of δ-ferrite ary of the molten pool, it is easier to grow in the < 100 > direction [19].
rapidly form at the solidification front. The competitive growth mechanism of the grains will result in the large
EDS map scanning analyses were used to identify the elements dis­ grain structure appearing in the weld. However, the austenite grains did
tribution in weld metal. Fig. 7 shows EDS analysis selected location and not grow excessively at different heat input. The reason is that the
the corresponding element distribution for the heat input of 95 J/cm. austenite microstructure is composed of layered ferrite precipitation
The chemical compositions of WZ at three kinds of heat input are listed (Fig. 8), and ferrite is formed along the grain boundary of heat affected
in Table 3. As shown in Fig. 7, the chemical enrichment elements include zone. Because the coefficient of thermal expansion of austenite is higher
Fe, Cr, Ni, Mn and C in WZ. The white particle is analyzed by EDS point than ferrite [20], the growth of austenite grain is restricted by ferrite
scanning. It can be observed that the white particles have a higher C growing along austenite grain boundary. Therefore, it is not found
content, especially spectrum 1. The EDS analysis suggested the segre­ obvious phase transformation in the heat affected zone of S30408
gation C and Ni in small white particles, as shown in Fig. 7, which austenitic stainless steel.
indicate the white particles are carbide particles. Additionally, the more The optical micrographs of the interface between transition layer and
content of Fe, Ni and Cr might result in the formation of hard FeNi/ welding zone are shown in Fig. 9(a-c). There is an unmixed zone (UZ)
CrNiFe phase [18]. For EDS map analysis of different samples, it can be next to the weld fusion line, which solidification pattern composes of
observed that different heat input shows different elements content. The beaches, peninsula, and islands, as shown Fig. 9(b). It can be attributed
alloy elements in different heat input lead to solute redistribution, to the different of melting range and chemical composition between

5
Y. Huang et al. Materials Today Communications 29 (2021) 102948

Fig. 8. Microstructure of interface S30408 steel and weld joint, (a) 95 J/cm, (b) 80 J/cm, (c) 66 J/cm.

Fig. 9. Microstructure of interface transition layer and weld joint; (a) 95 J/cm, (b) 80 J/cm, (c) 66 J/cm.

filler wire and base metal, which lead to form a viscous layer at the solidifying point process [21]. On the other hand, a UZ formation is
interface. A narrow band without mixing the liquid base metal was associated with the improper heat input of the appropriate welding pool
formed next to the molten pool boundary during the weld pool [9]. The microstructure of transition layer consists of a few sums of

6
Y. Huang et al. Materials Today Communications 29 (2021) 102948

Fig. 10. EDS line scan results of interface, (a) S30408/weld zone, (b) transition layer/weld zone.

Fig. 11. The microhardness distribution of weld zone. Fig. 12. Strong polarization curves for specimens with weld speed.

rough bulk ferrite and a large number of scaly granular bainite. The Table 4
dimension of the original sub grain and austenite grain in the transition Ecorr and Icorr of at three welding speed.
layer increase significantly, and with the decrease of heat input, the
Heat input (J/cm) 95 80 66
original austenite grain size increases.
Fig. 10 displays the EDS line scanning result of interface region for （Ecorr/V） -0.339 -0.773 -0.412
(Icorr/10− 6A•cm− 2) 5.71 13.18 11.48
heat input of 95 J/cm. The element distribution between the S30408
HAZ and the weld zone interface is shown in Fig. 10(a). The Fe content
of S30408 HAZ is significantly higher than weld zone, while the Ni higher heat input, and Cr-rich phase formed in the weld zone due to the
content shows an opposite trend. The Cr content of S30408 HAZ is heat effect, which lead to the weld zone have higher microhardness at
similar to the weld zone. Due to the unmixed zone is narrow, the ele­ higher heat input. As report by Rahmani et al.[21], the difference of
ments distributions dramatically change from S30408 HAZ to weld interstitial alloying elements in ferrite and austenite leads to differences
zone, and this variation is confirmed by the line mapping. However, the hardness of austenite and ferrite in the WZ. The higher the strength and
Cr content of weld zone is significantly higher than transition layer on hardness are reflected by the more the ferrite contents [21,22].
the side of 12Cr2Mo1R steel, as shown Fig. 10(b). A wider UZ is pre­
sented between transition layer and weld, which leads to the elements
3.3. Electrochemical test
gradually change from transition to the weld zone. The EDS spectra
suggested the weld zone has more concentration of Ni and Cr elements
Fig. 12 displays the cyclic polarization curves of welded joint at three
compared to the unmixed zone. More Fe element enrich in the unmixed
different heat input, all polarization diagrams follow Tafel-style
zone compared to the region near the UZ. Therefore, it is confirmed that
behavior. The self-corrosion potential is employed as a parameter to
the unsuitable mixing between the filler material and base metal.
characterize the electrochemical reaction process. The lower the self-
corrosion potential, the greater the potential difference between the
3.2. Hardness cathode and the anode, that is, the greater the thermodynamic trend of
the reaction. Siow et al. suggested that the sample has worse corrosion
As shown in Fig. 11, the microhardness of different weldments was resistance at larger corrosion current density (Icorr) and smaller corro­
studied. It was observed that the microhardness of the welding zone sion potential (Ecorr), which is owing to the nonuniform passivated film.
changed significantly with different heat input. The microhardness of Table 4 presents Icorr and Ecorr of different heat input. Different heat
weld joints increase gradually with the increase of heat input. The input of corrosion test samples show different corrosion density and
reason is that the more content of austenite transform to ferrite under corrosion potential. It indicates that the difference stability of the oxide

7
Y. Huang et al. Materials Today Communications 29 (2021) 102948

Fig. 13. EIS spectra diagrams of the sample with welding speed.

film formed on the superficies of the samples in the corrosive solution.

The sample at the heat input of 95 J/cm has the highest Ecorr and lowest
current density. Among the welding samples, the sample with the heat
input of 95 J/cm has the highest Ecorr, and its corrosion resistance is the
strongest.
The electrochemical impedance spectroscopy of weld metal was
tested by electrochemical workstation. Fig. 13 shows the Nyquist and
Bode diagrams of welded joints in 3.5 wt% NaCl solution for different
heat input. Fig. 13(a) reveals the capacitive impedance arc radius of
three samples, the larger the impedance radius indicates the stronger the
corrosion resistance. Comparing to heat input of 80 J/cm and 66 J/cm, Fig. 14. equivalent circuit used for impedance analysis.
the weld metal has a larger capacitive arc radius at the heat input is
95 J/cm. Therefore, the weld metal with heat input of 95 J/cm has the film formed on the weld metal in 3.5 wt% NaCl solution at the heat input
highest impedance modulus and provides the best corrosion resistance. of 95 J/cm is corrosion-resistant. When the heat input is 95 J/cm, the
According to the total impedance diagram of the welded joins under weld metal has the best corrosion resistance since the passivation film
different heat input, as shown in Fig. 13 (b). There is a flat part in the formed on the surface of the weld is thicker. The sample of the heat input
area of high frequency (104~105 Hz) of the three groups of samples, and of 80 J/cm is the worst. This is also consistent with the test results of
the slope is about 0. In the low and medium frequency range (10− 1~103 polarization curve.
Hz), the impedance spectrum reveals the impedance presents a linear According to the EIS test results of the weld metal at three heat input,
decrease as the frequency logarithm increases. Additionally, the low- the equivalent circuit diagram is displayed in Fig. 14. The EIS results
frequency impedance modulus reaches the order of 105 Ω⋅cm2. Fig. 13 show that the passivation layer formed on the sample superficies is not
(c) illustrates the phase angles of all samples first increase and then uniform and contains pores and impurities. Therefore, the equivalent
decrease with the increase of frequency. The phase angle increases with circuit consists of two capacitor loops. (R1Q1) and (R2Q2) are related to
the increase of frequency, indicating the contribution of the oxide film of the passive film and defects and impurities, respectively. In the first, Rs
the weld metal to the impedance. The phase angle decreases rapidly and Rf indicates the solution resistance passive film resistance, respec­
with the increase of frequency in the high frequency zone, explaining tively [24]. In the second circuit, Rct is the electrolyte resistance in pores.
that the solution resistance determines the impedance in this region. CPEf and CPEdI are the constant phase element, which is related the
Meanwhile, the maximum phase angle is about 80◦ and a small range of passivation film and the imperfection, respectively. The definition of
flat parts appeared in the intermediate frequency of 10 ~ 103 Hz for CPE in impedance can be expressed by following equation:
three groups of samples. Therefore, from the point of view of EIS, the

8
Y. Huang et al. Materials Today Communications 29 (2021) 102948

Table 5 Fig. 16 shows EDS line scanning of corrosion pitting, the content of
EIS parameters of equivalent circuit for weld zone. Cr decreasing at the pit location which indicates a Cr poor region is
Heat input (J/cm) 95 80 66 formed in the phase boundary between ferrite and austenite. The effect
2 of Cr element poor region on the corrosion behavior will be discussed in
RS(Ω∙cm ) 5.799 6.384 4.518
CPE1(F∙cm2) 5.93E-5 2.50E-5 3.11E-5 the discussion part.
n1 0.88 0.86 0.91
Rf(Ω∙cm2) 19,633 9610 13,172 4. Discussions
CPE2 (F∙cm2) 4.28E-4 4.64E-4 2.49E-4
n2 1.26 1.27 1.13
Rct(Ω∙cm2) 3920 1606 2584
As described in the experiment results, one interesting finding is the
different ratio of austenite to ferrite for different heat input and another
important question is correlation of elements distribution and corrosion.
ZCPE = Y0 − 1 (jω)− n
(2) Therefore, the phase transform and corrosion mechanism were dis­
cussed in this part.
Where n represents its exponent, w represents the angular frequency, As demonstrated by Kangazian J et. al’s [26], Ni element is the stable
and Y0 is a constant. When n = 0, CPE presents resistance; when n = 1,
CPE presents pure capacitance [25]. Table 5 demonstrates the results for
the equivalent circuit of EIS impedance spectra. The n value represents
the corrosion process of the weld metal face. The n parameters for
different heat input show different values, it indicates that the constant
phase elements. These results show CPEf and CPEdI display almost all the
capacitance functions of the samples in each circuit. According to the
larger Rf, the greater the resistance of the oxide film. Thereby, the results
show that the corrosion resistance of the weld under three different heat
input is as follows: 95 J/cm > 66 J/cm > 80 J/cm.
Fig. 15 is the OM images of pit morphology on weld metal after
electrochemical corrosion. The mixed microstructure includes austenite
and ferrite appeared in the weld zone. However, metastable and stable
pits exist in austenite phase in all weld microstructures. The corrosion
pits on the weld metal surface are smaller and fewer when the heat input
is 95 J/cm. However, the corrosion pits on the weld metal surface are
dense and uniform when the heat input is 80 J/cm and 66 J/cm. From
the corrosion pits formed on the weld metal surface, it is concluded that
the weld joint has the best resistance corrosion when the heat input is
95 J/cm. This result is consistent with the polarization curve and EIS test
results. Fig. 16. Cr content of corrosion pit analyzed by EDS line scanning.

Fig. 15. Morphology of the corrosion surface of weld zone; (a) 95 J/cm, (b) 80 J/cm, (c) 66 J/cm.

9
Y. Huang et al. Materials Today Communications 29 (2021) 102948

Fig. 17. Corrosion process of sample in 3.5 wt% NaCl solution.

element of austenite, while Cr element is the stable element of ferrite. 1. The combination welding process of AA-TIG backing welding and
The changes of austenite phase and ferrite for different heat input were SAW can effectively join thick plate with high efficiency, and the weld
analyzed by Creq/Nieq equivalent, which can be calculated according to shape is good without incomplete fusion and penetration. The Ni based
the Molian P A’s method [27]. As shown in EDS results [Table 3], Si, Mo surfacing transition layer on 12Cr2Mo1R side can inhibit the formation
and Nb were not detected in the quantitative analysis of elements of martensite phase.
because of their low content. When the heat input is 66 J/cm, 80 J/cm 2. The stable elements (Ni and Cr) in austenite phase and ferrite
and 95 J/cm, Creq/Nieq is as follow: 95 J/cm > 66 J/cm > 80 J/cm. phase of weld metal under different heat input are significantly
Owing to the content of ferrite decreases with the decrease of Creq/Nieq different. The lower heat input, the lower ferrite content, and the
value [28], the weld zone at the heat input of 80 J/cm is the minimum, microhardness also decrease with the decrease of ferrite phase.
which has higher content of ferrite. Additionally, the map analysis il­ 3. The analysis of pitting formation demonstrates that the chromium
lustrates a large amount of Ni element exist the weld zone. Therefore, poor area is the most vulnerable area to form pitting corrosion by
during the cooling process of liquid metal, more ferrite phase transforms chloride ion. The high content of Ni and the C element in the weld are
into austenite, and the content of austenite increases. While the heat likely to form carbon and nickel, and become the inducing point of
input decrease, and the liquid metal has a low cooling rate during so­ pitting corrosion. The heat input is 95 J/cm has stronger corrosion
lidification, which causes the ferrite phase content to decrease [29]. resistance.
In general, the pitting index of steel can be evaluated by the pitting
resistance equivalent number (PREN) and estimated by the empirical CRediT authorship contribution statement
relationship PREN= %Cr+ 20% N + 3.3%Mo [30]. Therefore, balanced
austenite phase and ferrite phase and high content of ferrite forming Yanqin Huang: Data curation, Writing – original draft. Jankang
elements such as Cr, Mo elements can improve the pitting resistance Huang: Supervision. Jianxiao Zhang: Visualization, Investigation.
[31]. In this study, the EDS analysis shows that these corrosion products Xiaoquan Yu: Writing – review & editing. Qi Li: Investigation. Zhen
are produced in the chromium poor region of the passivation layer, as Wang: Methodology. Ding Fan: Conceptualization.
shown in Fig. 16, so the passivation tendency may be caused by the Cr
content. However, the phase boundary between ferrite and austenite is
Declaration of Competing Interest
likely to form a Cr poor zone, which provides a place for pitting nucle­
ation [32]. The cathodic influence of the chromium rich region on the
The authors declare that they have no known competing financial
embedded chromium poor compound results in the formation of an
interests or personal relationships that could have appeared to influence
imperfective passivation oxide layer. The resistance of the film is
the work reported in this paper.
reduced by this layer to corrosive environment. Moreover, the presence
of chloride ions can accelerate the passivation breakdown [33,34]. In
Acknowledgment
addition, the high content of Ni and the C element present in there are
likely to form Ni-C phase at the interface of ferrite and austenite, as
This study thanks to the support of the National Natural Science
displayed in Fig. 17(a). In the second stage, as shown in Fig. 17(b), so­
Foundation of China (51775256).
dium chloride solution Chloride ions can accelerate the breakdown of
the passivation layer, and undergo redox reactions with nickel carbide
existing in the weld. Nickel gradually loses electrons and forms Ni2Cl References
with Cl- in the solution, causing more passivation film to be destroyed
[1] B. Mathew, Benkim, J. Babu, Multiple process parameter optimization of WEDM on
and pits are formed, as shown in Fig. 17(c). AISI304 using taguchi grey relational analysis, Proc. Mater. Sci. 5 (2014)
1613–1622.
5. Conclusions [2] Eszter KALÁCSKA, Kornél MÁJLINGER, Enikö R.éka FÁBIÁN,
Russo PASQUALE Spena, MIG-welding of dissimilar advanced high strength steel
sheets, Mater. Sci. Forum 885 (2017) 80–85.
The effect of heat input on microstructure and corrosion behavior of [3] Balázs Varbai, Kornél Májlinger, Optimal etching sequence for austenite to ferrite
S30408 austenitic stainless steel and 12Cr2Mo1R dissimilar steel welded ratio evaluation of two lean duplex stainless steel weldments, Measurement 147
(2019), 106832.
joint was studied in detail. According to the current research, the [4] P. Reccagni, L.H. Guilherme, Q. Lu, M.F. Gittos, D.L. Engelberg, Reduction of
following important conclusions can be drawn：. austenite-ferrite galvanic activity in the heat-affected zone of a Gleeble-simulated
grade 2205 duplex stainless steel weld, Corros. Sci. 161 (2019) 1–6.

10
Y. Huang et al. Materials Today Communications 29 (2021) 102948

[5] X. Chen, M. Luo, R. Hu, R. Li, L. Liang, S. Pang, Thermo-electromagnetic effect on [19] E. Taban, E. Kaluc, Welding behaviour of duplex and superduplex stainless steels
weld microstructure in magnetically assisted laser welding of austenite steel, using laser and plasma ARC welding processes. Welding in the World, Springer,
J. Manuf. Process. 41 (2019) 111–118. 2011, pp. 48–57.
[6] K. Dai, H. Peng, Z. Li, F. Zhang, F. Chen, Z. Xian, W. Xiao, Effect of solution [20] Gaurav Dak, Chandan Pandey, Experimental investigation on microstructure,
treatment on microstructure and cryogenic toughness of 316LN austenite stainless mechanical properties, and residual stresses of dissimilar welded joint of
steel weld metal welded by NG-MAG arc welding, Fusion Eng. Des. 125 (2017) martensitic P92 and AISI 304L austenitic stainless steel, Int. J. Press. Vessels Pip.
178–188. 194 (2021), 104536.
[7] T.E. Abioye, A. Medrano-Tellez, P.K. Farayibi, P.K. Oke, Laser metal deposition of [21] S. Kumar, C. Pandey, A. Goyal, Role of dissimilar IN617 nickel alloy consumable on
multi-track walls of 308LSi stainless steel, Mater. Manuf. Process. 32 (2017) microstructural and mechanical behavior of P91 welds joint, Arch. Civ. Mech. Eng.
1660–1666. 20 (2020) 3.
[8] B. Teaa, C. Oea, C. Tio, B. Pkf, C. Oog, Effects of welding speed on the [22] M. Rahmani, A. Eghlimi, M. Shamanian, Evaluation of microstructure and
microstructure and corrosion behavior of dissimilar gas metal arc weld joints of mechanical properties in dissimilar austenitic/super duplex stainless steel joint,
AISI 304 stainless steel and low carbon steel, Mater. Today.: Proc. 17 (2019) J. Mater. Eng. Perform. 23 (2014) 3745–3753.
871–877. [24] K.S. Siow, T.Y. Song, J.H. Qiu, Pitting corrosion of duplex stainless steels, Anti
[9] Anup Kumar Maurya, Chandan Pandey, Rahul Chhibber, Dissimilar welding of Corros. Methods Mater. 48 (2001) 31–37.
duplex stainless steel with Ni alloys: a review, Int. J. Press. Vessels Pip. 192 (2021), [25] Z. Brytan, J. Niagaj, Ł. Reiman, Corrosion studies using potentiodynamic and EIS
104439. electrochemical techniques of welded lean duplex stainless steel UNS S82441,
[10] Gaurav Dak, Chandan Pandey, A critical review on dissimilar welds joint between Appl. Surf. Sci. 388 (2016) 160–168.
martensitic and austenitic steel for power plant application, J. Manuf. Process. 58 [26] J. Kangazian, M. Shamanian, A. Ashrafi, Dissimilar welding between SAF 2507
(2020) 377–406. stainless steel and Incoloy 825 Ni-based alloy: the role of microstructure on
[11] S. Sirohi, C. Pandey, A. Goyal, Role of the Ni-based filler (IN625) and heat- corrosion behavior of the weld metals, J. Manuf. Process. 29 (2017) 376–388.
treatment on the mechanical performance of the GTA welded dissimilar joint of [27] P.A. Molian, Solidification behaviour of laser welded stainless steel, J. Mater. Sci.
P91 and SS304H steel, J. Manuf. Process. 65 (2021) 174–189. Lett. 4 (1985) 281–283.
[12] N. Arivazhagan, S. Singh, S. Prakash, GM Reddy. Investigation on AISI 304 [28] J. Verma, V.Ravindra Taiwade, R. Kataria, A. Kumar, Welding and electrochemical
austenitic stainless steel to AISI 4140 low alloy steel dissimilar joints by gas behavior of ferritic AISI 430 and austeno-ferritic UNS 32205 dissimilar welds,
tungsten arc, electron beam and friction welding, Mater. Des. 32 (2021) J. Manuf. Process. 34 (2018) 292–302.
3036–3050. [29] L. Jie, J. Ma, Y. Gao, Q. Zhai, Research on solidification structure refinement of
[13] T.E. Abioye, Electrochemical corrosion study of Inconel 625 laser coatings by wire pure aluminum by electric current pulse with parallel electrodes, Mater. Sci. Eng. A
and powder feedstock, Int. J. Mater. Eng. Innov. 8 (2017) 173. 490 (2008) 452–456.
[14] X. Zhang, G. Mi, C. Wang, Microstructure and performance of hybrid laser-arc [30] S.H. Bae, H.W. Lee, Effect of Mo contents on corrosion behaviors of welded duplex
welded high-strength low alloy steel and austenitic stainless steel dissimilar joint, stainless steel, Met. Mater. Int. 19 (2013) 563–569.
Opt. Laser Technol. 122 (2020), 105878. -105878. [31] N.N. Khobragade, M.I. Khan, A.P. Patil, Corrosion behaviour of
[15] J. Verma, R.V. Taiwade, Dissimilar welding behavior of 22% Cr series stainless chrome–manganese austenitic stainless steels and aisi 304 stainless steel in
steel with 316L and its corrosion resistance in modified aggressive environment, chloride environment, Trans. Indian Inst. Met. 67 (2013) 263–273.
J. Manuf. Process. 24 (2016) 1–10. [32] R. Walker, Principles and prevention of corrosion: Denny A. Jones Macmillan,
[16] B.O. Okonkwo, H. Ming, Z. Zhang, J. Wang, E. Rahimi, S. Hosseinpour, A. Davoodi, 1992, ISBN 0–02-361215–0, 16.95. Materials & Design, 1993, 14:207.
Microscale investigation of the correlation between microstructure and galvanic [33] S. Wang, Q. Ma, L. Yan, Characterization of microstructure, mechanical properties
corrosion of low alloy steel A508 and its welded 309/308L stainless steel overlayer, and corrosion resistance of dissimilar welded joint between 2205 duplex stainless
Corros. Sci. 154 (2019) 49–60. steel and 16MnR, Mater. Des. 32 (2011) 831–837.
[17] D. Waynereitz, A review of: welding metallurgy and weldability of stainless steel, [34] M. Atapour, H. Sarlak, M. Esmailzadeh, Pitting corrosion susceptibility of friction
Adv. Manuf. Process. 21 (2016) 219. -219. stir welded lean duplex stainless steel joints, Int. J. Adv. Manuf. Technol. 83 (2016)
[18] D.K. Mondal, R.M. Dey, Effect of grain size on the microstructure and mechanical 721–728.
properties of a C Mn V dual-phase steel, Mater. Sci. Eng.: A 149 (1992) 173–181.

11