Materials Science & Engineering A 854 (2022) 143809

Contents lists available at ScienceDirect

Materials Science & Engineering A

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

A specially-designed super duplex stainless steel with balanced ferrite:

austenite ratio fabricated via flux-cored wire arc additive manufacturing:
Microstructure evolution, mechanical properties and corrosion resistance
Yiqi Zhang a, Shaojie Wu a, b, *, Fangjie Cheng a, b, **
a
School of Materials Science and Engineering, Tianjin University, Tianjin, 300350, China
b
Tianjin Key Laboratory of Advanced Joining Technology, Tianjin, 300350, China

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

Keywords: The balanced ferrite:austenite (α/γ) ratio is essential for duplex stainless steels (DSSs) to achieve superior
Duplex stainless steel strength and corrosion resistance. Nevertheless, the chemical compositions of commercial welding wires are not
Wire arc additive manufacturing especially designed for the wire arc additive manufacturing (WAAM) process, leading to excessive austenite
Microstructure evolution
formation (~74%) in their as-built DSSs. In this research, the chemical composition of a flux-cored wire arc
Corrosion resistance
Strength
additively manufactured super duplex stainless steel (FCWA-AM SDSS) was especially designed based on its
thermal-history to achieve the balanced α/γ ratio. Mechanisms describing the microstructure evolution of α and γ
during the FCWA-AM process were also proposed. Cr2N was the secondary phase formed in the as-built SDSS,
which acted as the nucleating sites of intragranular austenite during the cooling of the deposition layer and
precipitated at Widmanstätten austenite boundaries in the subsequent intrinsic heat treatment (IHT). The
corrosion tests suggested that a thicker and more stable passive film was formed on the surface of the as-built
SDSS than the hot-rolled 2205 DSS but their pitting corrosion resistance were comparable because the precip­
itation of Cr2N hindered the improvement of pitting corrosion resistance. The as-built SDSS showed higher
strength than previously reported WAAM DSSs because of the enhanced solid solution strengthening effect and
the balanced α/γ ratio.

1. Introduction hardness of the laser powder bed fusion sample. R. Casati et al. [4],
added Si powder to 7xxx Al powder and developed an Al-6.8Zn-6.5­
Additive manufacturing (AM) is a process of joining materials to Si–2Mg-1.3Cu alloy with high strength and hardness through SLM.
make parts from 3D model data, usually layer upon layer, as opposed to Proper alloying design of the wires for the specific WAAM process is also
subtractive manufacturing and formative manufacturing methodologies required to achieve the expected microstructure and the customized
[1]. Its “bottom-up” approach gives it great advantage over traditional performance [5]. However, the fabrication of the especially designed
“top-down” approach when manufacturing components with complex solid alloy wire is challenging because the work hardening and precip­
shape. The commonest metallic additive manufacturing processes itation strengthening is extremely strong during the drawing wire pro­
include powder-based AM such as selective laser melting (SLM) and cess [6]. Thus, most of the investigations on the WAAM use commercial
wire-based AM such as wire arc additive manufacturing (WAAM) [2]. In welding wire as the filler material and there have been limited wires
recent years, an increasing number of works have been dedicated on the specially designed and optimized for the WAAM process. The modifi­
design and modification of alloy powders used for powder-based AM cation of microstructure and improvement of properties were majorly
through in-situ alloying of powder mixtures to achieve the expected realized through the optimization of processing parameters [7,8] and
microstructure and properties. Emilie Hørdum Valente et al. [3], post manufacturing treatments [9,10].
modified the AISI 316L powder by the addition of 2.5 wt% Cr2N powder, Duplex stainless steel (DSS) is composed of nearly equal proportions
which resulted in a significant improvement of corrosion resistance and of austenite (γ) and ferrite (α) and it is widely used in environments that

* Corresponding author. School of Materials Science and Engineering, Tianjin University, Tianjin, 300350, China.
** Corresponding author. School of Materials Science and Engineering, Tianjin University, Tianjin, 300350, China.
E-mail addresses: [email protected] (S. Wu), [email protected] (F. Cheng).

https://doi.org/10.1016/j.msea.2022.143809
Received 22 June 2022; Received in revised form 8 August 2022; Accepted 14 August 2022
Available online 18 August 2022
0921-5093/© 2022 Elsevier B.V. All rights reserved.
Y. Zhang et al. Materials Science & Engineering A 854 (2022) 143809

demand high strength, good corrosion resistance or both [11], such as in acceptable range [23].
ship building, marine industries and chemical engineering. The balance In this work, a flux-cored wire was especially designed for the WAAM
of the α/γ is one of the first factors taken into consideration in the process of SDSS with modified chemical composition to achieve the
manufacturing of DSS component because it ensures the optimum balanced α/γ ratio in the as-built state. The thermal history, micro­
combination of good corrosion resistance and mechanical properties structure evolution, tensile properties and corrosion resistance of the
[12]. Excessive ferrite reduces the corrosion resistance and toughness of FCWA-AM SDSS were reported.
the DSS component by precipitating large amount of intermetallic sec­
ondary phases at its interior while too much austenite forming in DSS 2. Materials and methods
component decreases its hardness and strength because it is the softer
phase than ferrite. NORSOK M − 630 [13] specifies that the acceptable The setup of the experiment was shown in Fig. 1a. A KENNPPI FAST
austenite ratio in DSS is 35–65%. MIG welder was employed as the power source in this experiment, and
The investigation on the AM of DSS has just started in recent years the shielding gas was the mixture of 80%Ar+20%CO2. The pulsed
and is still in its infancy. Currently, investigations on the microstructure GMAW arc mode was selected to achieve a projected spray metal
and properties of the additively manufactured DSSs using commercial transfer mode during the deposition process [24]. Such arc mode can
powder and welding wire as the feedstock materials were conducted. generate significantly higher arc force than DC arc mode, resulting in
However, most of them showed upset α/γ balance in the as-built DSS and greater axial pressure, higher energy density and better arc stability. As
the effects of the adjustments of processing conditions on the α/γ bal­ a result, the depression of the deposition pools using pulsed GMAW arc
ance had also been proved to be limited [5]. In most powder based mode is generally deeper than the pools using DC arc mode and often
additively manufactured DSSs, excessive ferrite was usually found while showed a finger-shaped penetration. The detailed deposition parameters
austenite phase could hardly be observed because of the small heat input used in this experiment were listed in Table 1.
and the rapid cooling rate [5,14,15]. For WAAM DSSs, however, the The diameter of the especially-designed flux-cored wire was 1.2 mm,
successive reheating and the 2D heat dissipation condition caused with the structure of 304L stainless steel as the outer layer and the metal
greater heat accumulation than welding, leading to a significantly powder as the flux core (Fig. 1b). The weight ratio that the flux core
longer duration in the austenite formation temperature range [16]. As a accounted for the wire, namely the filling rate, was 24%. Details about
result, excessive austenite (~74%) was usually found in WAAM DSSs how the flux-cored wire was manufactured were shown in the supple­
using commercial welding wire as the deposition material [17–20]. mentary data related to this article. A 304 plate with dimensions of
Several investigations were conducted on modifying the α/γ ratio in 125ⅹ250ⅹ9 mm3 was used as the substrate and a single pass multi-layer
AM DSSs through post manufacturing heat treatment. According to wall with totally 38 layers was deposited, as presented in Fig. 1c and d.
Hengsbach et al. [15], an increasing of austenite ratio from 1% to 34% The chemical composition of the specially-designed FCWA-AM DSS was
was achieved in the SLM 2205 DSS after the heat-treatment at 1000 ◦ C. measured using a 5100VDV Inductively Coupled Plasma Optical Emis­
In our previous investigation [10], the austenite ratio in the E2209 sion Spectrometer (ICP-OES) and an ELTRA ONH2000 Oxygen Nitrogen
WAAM DSS was modified to ~50% by the 1300 ◦ C heat-treatment and Hydrogen Analyzer, as listed in Table 2. The influence of chemical
found that the sample’s pitting corrosion resistance was comparable composition on the pitting corrosion resistance of stainless steel could
with the hot-rolled 2205 DSS. However, the high temperature generally be evaluated using the pitting resistance equivalent number
heat-treatment was costly and limited the application of AM DSSs. Thus, (PREN) as shown in Eq (3).
some recent investigations managed to achieve a balanced α/γ ratio in
PREN=Cr%+3.3Mo%+16 N% (3)
the as-built DSS by modifying the alloy elements of the feedstock ma­
terial. Alloying elements in DSSs can generally be divided into ferrite The PREN of the specially-designed FCWA-AM DSS was as high as
promoting elements and austenite promoting elements. Their effects on 46.141. According to Lippold et al. [11], DSS with PREN higher than 40
the phase transformation can be evaluated using the chromium equiv­ could be classified into super duplex stainless steel (SDSS) grade.
alent (Creq) and nickel equivalent (Nieq) equations shown in Eq (1) and Therefore, the specially-designed FCWA-AM DSS in this research will be
Eq (2) [11]. named FCWA-AM SDSS in the follow.
Creq = Cr%+Mo%+0.7Nb% (1) The thermal history during the FCWA-AM process was investigated
using a non-contact pyrometer (LUMASENSE TECHNOLOGIES, spectral
Nieq = Ni%+35C%+20 N%+0.25Cu% (2) range 1 μm, effective range 400–1400 ◦ C) and a K-type thermocouple.
The pyrometer was used to record the cooling curves of the 1st to 10th
Marie Luise Köhler et al. [21], used three powder mixtures of the
deposition layer and the K-type thermocouple was spot welded to the
ratios AISI 316L + AISI 318LN 70:30, 50:50, 30:70 as the feedstock 15th deposition layer to record the typical successive reheating a
material for the SLM DSS. A balanced α/γ ratio was achieved in the SLM
deposition layer experiences during the FCWA-AM process.
DSS using the 70:30 mixture at the energy volume density of 48J/mm3. The positions of the samples cut from the FCWA-AM SDSS were
In order to overcome the difficulty in the fabrication of high-quality shown in Fig. 1c. To investigate the microstructure evolution, the X-Z
alloy wires, we recently proposed a novel WAAM method using plane of the sample was grounded to 2000 grit sandpaper, polished with
flux-cored wire as the feedstock material, namely flux-cored wire arc diamond paste to 0.25 μm and then electrolytically etched in 40 wt%
additive manufacturing (FCWA-AM) [17]. The flux-cored wire consists NaOH solution at 3V for 5–10s. The transmission electron microscopy
of a metal outer sheath filled with metal and mineral flux powder core (TEM, Philip-CM200) was used to characterize the secondary phase
[22]. The major purpose of using the flux is that various materials can be precipitation in the FCWA-AM SDSS. The TEM foil was twinjet electron
added to it to allow many performance enhancements to be made. For polished in a mixture of 5 vol% perchloric acid and 95 vol% alcohol at
example, different types and propositions of alloy powders can be added − 30 ◦ C and 32V.
to the flux of welding wires to improve their matching degree with the 3 samples in horizontal (Y) and vertical (Z) direction were prepared
newly developed base metal. Slag formers can also be added to the flux for the tensile tests. The experiments were performed on a INSTRON
of welding wires to shield the weld pool and shape and support the weld. 5848 MicroTester, with the cross-head speed of 2 mm/min. Their frac­
In the development of the feedstock material for WAAM, we recently ture surface morphologies were characterized using a JSM-7800F
modified the proportion of metal powder added to the flux core and scanning electron microscope after the tensile test.
successfully fabricated a FCWA-AM DSS with the composition of The corrosion tests were conducted on a CHI-660e potentiostat. The
Fe-27.88Cr-6.44Ni-5.18Mo-1.51Mn-0.41Si-0.16N-0.08C (wt%). Its test solution was 3.5 wt% NaCl at 25 ± 0.1 ◦ C and all of the specimens
austenite ratio was around 37%, which was at the lower limit of the

2
Y. Zhang et al. Materials Science & Engineering A 854 (2022) 143809

Fig. 1. Experimental of this work (a) Experimental set up for the FCWA-AM process; (b) photograph of the newly develop flux cored wire especially designed for
FCWA-AM of DSS and its structure; (c), (d) side-view and cross-section of the FCWA-AM SDSS.

3. Results
Table 1
Deposition parameters.
3.1. Thermal history and its induced intrinsic heat treatment in FCWA-
Parameter Value AM process
Average deposition current (I) 145 A
Average deposition voltage (U) 30 V The microstructure evolution of DSS depends on its manufacturing
Deposition speed (v) 5.4 mm/s
processes and chemical compositions [5]. Thus, the alloy design and the
Shielding gas and flow rate 80%Ar+20%CO2, 15L/min
Arc mode Pulsed GMAW
microstructure evolution investigations of the FCWA-AM SDSS had to be
Inter-pass temperature ≤50 ◦ C based on the thermal history it experienced during manufacturing, as
presented in Fig. 2.
With regard to manufacturing process of DSS, the duration time in
1200–800 ◦ C, t12/8, is used as an index to reflect the cooling rate in the
Table 2
Chemical composition of the FCWA-AM SDSS. temperature range of α→γ transformation. A longer t12/8 generally leads
to more austenite in DSS [5,16]. The cooling curves from the 1st to 10th
C Si Mn Cr Mo Ni N Fe
layer were shown in Fig. 2a and their t12/8 were presented in Fig. 2b. The
FCWA-AM 0.07 0.43 1.52 29.89 3.47 6.54 0.30 Balance results suggested that the t12/8 of the 1st layer was only 2.6s but raised to
SDSS
5.7s in the 10th layer. The drastic increase of t12/8 was noticed in the
first four layers and stabilized at ~5.5s after five deposition layers,
were immersed in the test solution for 1 h to achieve a stabilized open indicating that the cooling of the deposition layers higher than the 5th
current potential (OCP) before each experiment. Electrochemical would no longer be influenced by the heat dissipation of the substrate.
impedance spectroscopy (EIS) tests were conducted at OCP in a partic­ As presented in Fig. 2b, the stabilized t12/8 of the FCWA-AM DSS was
ular range of frequency (100kHz-0.01Hz) with an amplitude of 10 mV. approximately double the length of the t12/8 of the pulsed arc welded
ZSimpwin software was used to fit the EIS experiment data. The scan­ SDSS joint [16]. In other words, the FCWA-AM process had a signifi­
ning rate of potentiodynamic polarization tests was 10 mV/min. Each cantly slower cooling rate than the welding process, especially in the
corrosion test was conducted two or three times to ensure repeatability temperature range where α→γ happened. Thus, a higher Creq/Nieq
[25]. should be designed for the FCWA-AM SDSS than the joints to achieve the
α/γ balance.
The successive reheating during the FCWA-AM process can act as an
intrinsic heat treatment (IHT) on the deposition metal, triggering the
“in-situ” solid-state phase transformation such as the growth of austenite

3
Y. Zhang et al. Materials Science & Engineering A 854 (2022) 143809

Fig. 2. Thermal history of the FCWA-AM SDSS: (a) Cooling curve of the 1st-10th deposition layer; (b) t12/8 of the 1st-10th deposition layer; (c) The typical successive
reheating a deposition layer experiences during FCWA-AM process; (d) Temperature range of the major solid state phase transformations that may happens during
the solidification and reheating of a DSS deposition layer.

and intermetallic phase precipitation in the FCWA-AM SDSS. In order to strength or both. Lastly, the precipitation of large amount of σ phase is
have a better understanding of the microstructure evolution of the expected to be avoided because its formation significantly reduces the
FCWA-AM SDSS, the typical reheating experienced by a deposition layer corrosion resistance of DSS component and increases the crack
was shown in Fig. 2c. Fig. 2d showed the typical solid-state phase susceptibility.
transformation temperature ranges in DSSs [26]. As presented in Fig. 2d In WAAM DSSs, a high Creq/Nieq increases the ferrite percentage,
that most of the solid-state phase transformations happened in the while a low Creq/Nieq promotes the formation of austenite. In WAAM
temperature range of 600–1100 ◦ C, thus reheating with peak tempera­ DSSs using commercial E2209 [10] (Creq/Nieq = 1.84) and ER2594 [20]
ture lower than 600 ◦ C was unlikely to cause further microstructure (Creq/Nieq = 1.89) welding wire, the austenite ratio in the as-built
evolution. Therefore, it could be predicted from Fig. 2c that the micro­ WAAM DSS component was ~74%. In our last investigation [23], a
structure in a deposition layer would be stabilized after being reheated FCWA-AM DSS with Creq/Nieq = 2.65 was developed and its austenite
by the following 2 passes. That is to say, the microstructure evolution ratio was successfully controlled to ~37%, which was at the lower limit
during the FCWA-AM process was recorded in the last three deposition of the acceptable range. In order to achieve the balanced α/γ ratio in the
layers, which experienced 0, 1 and 2 thermal cycles. The similar as-built state, the desirable Creq/Nieq for FCWA-AM SDSS must be lie
morphology in the bulk and in the antepenultimate layer reported in the within the range of 1.89–2.65 as shown in Fig. 3a. Thus, the Creq/Nieq =
following Section 3.2 and 3.3 further proved the above hypothesis. 2.23, which was at the middle of the range, was chose for the FCWA-AM
SDSS.
Elemental Cr plays an important role in the formation of passive film
3.2. Alloy design concept of the FCWA-AM SDSS and its solid solution strengthening effect can improve the strength of
DSS [10]. In order to achieve the superior strength and corrosion
The alloy design of the FCWA-AM SDSS took the following re­ resistance, a Cr content as high as 29.89% was selected for the
quirements into consideration. First of all, a balanced α/γ ratio is ex­ FCWA-AM SDSS. All of the elemental Ni in the FCWA-AM SDSS were
pected to be achieved in the as-built FCWA-AM SDSS as it ensures the from the 304L out layer (Ni content: ~8%) and the filling rate was 24%,
combination of good corrosion resistance and mechanical properties. thus, the Ni content in the FCWA-AM SDSS was 6.54%. Elemental Mo, N
Additionally, a high strength and good corrosion resistance was ex­ were used to balance the Creq/Nieq and the addition of other elements
pected to be achieved in the as-built FCWA-AM SDSS because it is mostly such as Si, Mn was determined by referring to the commercial DSS
used in applications that demands superior corrosion resistance,

4
Y. Zhang et al. Materials Science & Engineering A 854 (2022) 143809

Fig. 3. The alloy design concept of the FCWA-AM SDSS (a) plots of Creq vs Nieq; (b) 0.5%σ-phase theoretical CCT curve.

welding wires. et al. [29], and Liu et al. [27]. In Fig. 5, the fusion lines between the
The continuous cooling transformation (CCT) diagram calculated deposition layers were marked using the yellow dotted line and the
using JmatPro software was used to evaluate the possibility of σ phase interface between the FZ and AZ was marked using the pink dotted line.
precipitation. It was read in Fig. 3b that the cooling curve will not In the AZ of layer 38, the ferrite showed equiaxed grain structure at
intersect with the 0.5% σ-phase theoretical CCT curve when the cooling the surface area (Fig. 5b) while exhibited columnar grain structure at the
rate is higher than 10 ◦ C/s. As calculated from Fig. 2a, the cooling rate in interior range (Fig. 5c). Three typical austenite namely Widmanstätten
the σ precipitation range (600–1000 ◦ C) was as high as 40 ◦ C/s. austenite (WA), grain boundary austenite (GBA) and intragranular
Therefore, it could be expected that there wouldn’t be a serious pre­ austenite (IGA) were found in the last layer. Among them, IGA was the
cipitation of harmful secondary phase in cooling process. predominate type of austenite while only minor amount of GBA and WA
Fig. 4 showed the typical microstructure in the bulk of the FCWA-AM could be observed. The average austenite ratio in AZ of the 38th layer
SDSS. A more balanced α/γ ratio was achieved in the FCWA-AM SDSS was 43%. At the FZ of the 38th layer (Fig. 5d), column ferrite grains
than in WAAM DSSs using the commercial welding wire and the flux grew epitaxially from the bottom of the deposition pool towards the
cored wire in our previous investigation (Fig. 3a). As measured using the upper center and some equiaxed ferrite were observed at the interface
Feritscope, the average austenite ratio in the bulk was ~45%, suggesting between AZ and FZ. These equiaxed ferrite were remained in the bulk
that a balanced α/γ ratio was achieved in the FCWA-AM SDSS. (Fig. 4c) as they distribute at the bottom of the deposition layers and will
not be remelted by the subsequent pass. The volume fraction of austenite
3.3. Microstructure evolution and secondary phase precipitation in the FZ of the 38th layer was as low as 33% and only some GBA could
be found. As shown in Fig. 5e, coarsen column ferrite grains were found
The macrostructure of the last three deposition layers were shown in growing continuously across the fusion line between the AZ of the 37th
Fig. 5a. Based on their shape characteristics, each deposition layer could and 38th layer.
be partitioned into two regions: the wide upper arc zone (AZ) and the In the AZ of the one-pass reheated layer 37 (Fig. 5f), some WA in
narrow bottom finger zone (FZ). The formation of the AZ was the result their initialization was noticed nucleating on the GBA, thus the austenite
of the melting by the arc [27] and the deep and narrow FZ was formed ratio was slightly increased to 47%. The one-pass reheating also resulted
by the impact of the speedy fine droplet during the spray metal transfer in the formation of some WA and IGA in the FZ of layer 37, raising its
and the high plasma jet pressure exerted at the center of the deposition austenite volume fraction to 38%. The two-passes reheating caused the
pool [28]. Similar hybrid morphologies have also been obtained by Wu rapid growth of WA (Fig. 5h), leading to the further increase of the

Fig. 4. Microstructure in the bulk of FCWA-AM SDSS: (a) column grains in AZ; (b) column grains at the root of FZ; (c) equiaxial grains at the AZ-FZ interface.

5
Y. Zhang et al. Materials Science & Engineering A 854 (2022) 143809

Fig. 5. Microstructure evolution of the FCWA-AM SDSS.

austenite volume fraction in the AZ of the 36th layer to 52%. The and mainly controlled by the charge transfer phenomena through the
austenite ratio in FZ of the 36th layer was also increased to 42% after the protective layer [32]. An equivalent electrical circuit (EEC) model pre­
two-passes reheating. The above investigation on the microstructure sented in Fig. 7b was employed to model and fit the EIS spectra [25]. The
evolution indicated that the balanced austenite-ferrite ratio in the Rs, Rct, CPE in this model represented the solution resistance, the charge
FCWA-AM SDSS was formed in the IHT induced by the subsequent two transfer resistance and the double charge layer capacitance, respec­
layers. tively. Electrochemical impedance parameters obtained from the fitting
Secondary phase precipitation is a common phenomenon in the DSS of EIS diagrams were shown in Table 3. The stability of the passive film
joints and WAAM DSSs because of the successive reheating. Thus, TEM was reflected by Rct, which was related to the radius of the semicircle. A
characterization in the 36th and 38th layer was conducted to investigate higher Rct and a larger radius of the semicircle indicates a higher sta­
the secondary phase precipitation in the FCWA-AM SDSS. After careful bility of the passive film [33]. As can be seen from Fig. 7a that the
observation, no secondary phase except for Cr2N was found in the Nyquist plot of the FCWA-AM SDSS had a larger radius than the 2205
FCWA-AM SDSS. In the last deposition layer (Fig. 6a), a rodlike Cr2N DSS and the fitted results also showed that the FCWA-AM SDSS had
particle was closely attached to an initial IGA and the diffraction pattern higher Rct. Therefore, the passive film formed on the surface of the
(Fig. 6b) showed that the Cr2N and the initial IGA had [0001]Cr2N//[1 FCWA-AM SDSS was more stable than the one formed on the standard
11]IGA, (1 1 00)Cr2N//(2 20)IGA orientation relationship. Hence, the Cr2N 2205 DSS.
particles acted as the heterogeneous nucleation site for IGAs [30] in the The capacitive behavior of the electrical double layer and the passive
38th deposition layer, which resulted in the IGA predominate micro­ film is represented as constant phase element (CPE) instead of a perfect
structure in the last layer. In a grown IGA in the 36th deposition layer capacitance to address the existing micro or nano surface in­
(Fig. 6c), Cr2N was not observed neither at its interior nor its grain homogeneities. The impedance value of the CPE can be described using
boundary, suggesting the growth of IGA caused the dissolution of Cr2N. Equation (4)
As shown in Fig. 6d, Cr2N were found nucleating at the grain boundaries
ZCPE = [Y0(jω)n]− 1
(4)
of WA. These particles were richer in Cr and N than the adjacent
1/2
austenite and ferrite (Fig. 6e,f,g) and their precipitation resulted in the Where Y0 is a CPE constant, j= (− 1) , ω is the angular frequency, and n
formation of Cr-depleted zone at the grain boundaries of WA (Fig. 6f), is the dispersion coefficient of the CPE [34]. Electrochemical impedance
which was adverse to the pitting corrosion resistance and ductility [31]. parameters, Y0 and n of the CPE were obtained from the fitting of EIS
diagrams and shown in Table 3.
3.4. Corrosion properties The thickness of the passive film, δ, was obtained by using Equation
(5)
Fig. 7 compared the passivation behavior of the FCWA-AM SDSS εε0
with the standard hot rolled 2205 DSS. The Nyquist plots (Fig. 7a) of δ = (5)
Ceff
both metals showed a typical single capacitive semicircle, suggesting
that the mechanism of corrosion reaction for both metals were the same

6
Y. Zhang et al. Materials Science & Engineering A 854 (2022) 143809

Fig. 6. Cr2N precipitation during the FCWA-AM process: (a) TEM image showing initial IGA nucleate on Cr2N particles; (b) The electron diffraction pattern; (c) TEM
image of a grown IGA grain; (d) Cr2N particles precipitated on the WA grain boundaries. (e) EDS result of the Cr2N at the WA boundary; (f) liner scanning result of Cr
across the Cr2N; (g) liner scanning result of N across the Cr2N.

where ε ~ 12 was the relative dielectric constant of the passive film of

duplex stainless steel. ε0 is the permittivity of vacuum with a value of ε0 Ceff = gY0(ρδεε0)1− n
(6)
= 8.8542ⅹ10− 14F/cm [25]. The Ceff is the effective capacitance of the
CPE, which was calculated using Equation (6) according to the ρδ is the resistivity and is assigned a value of ρδ = 500Ωcm in the
power-law model proposed by Orazem et al. [35]. power-low model. g is a function of n which can be expressed by

7
Y. Zhang et al. Materials Science & Engineering A 854 (2022) 143809

Fig. 7. EIS results of the FCWA-AM SDSS and the standard 2205 DSS plate: (a) Nyquist plots; (b) the Equivalent circuit model applied to simulate the EIS spectra.

Table 3
Electrochemical impedance parameters obtained from the fitting of EIS results.
Specimen type Code Rs(Ωcm2) Y0 (Ω− 1
cm− 2 n
s ) n Rct (Ωcm2) Average Rct (Ωcm2) δ (nm) Average δ (nm)
− 5 5 5
FCWA-AM SDSS 1 4.32 4.27ⅹ10 0.883 4.64ⅹ10 5.12ⅹ10 3.00 2.78
5
2 6.24 3.93ⅹ10− 0.892 5.47ⅹ105 2.70
5
3 6.01 3.74ⅹ10− 0.895 5.24ⅹ105 2.65
5
Standard 2205 DSS plate 1 5.69 3.36ⅹ10− 0.916 3.84ⅹ105 3.64ⅹ105 1.89 1.78
5
2 6.87 3.55ⅹ10− 0.922 3.51ⅹ105 1.58
5
3 5.61 4.21ⅹ10− 0.906 3.57ⅹ105 1.86

Equation (7)
Table 4
g = 1 + 2.88(1-n)2.375 (7) Corrosion parameters of the polarization curves in 3.5 wt% NaCl solution.
Specimen type Code Ecorr (V/ Epit (V/ Ipass Data
Based on Equation (5)–(7), the thickness of the passivation film of
SCE) SCE) A/cm2 source
each sample was calculated and listed in Table 3. The result indicated
6
FCWA-AM SDSS 1 − 0.14 0.99 1.33ⅹ10− This
that a thicker passivation film was formed on the FCWA-AM SDSS than 6
2 − 0.19 1.05 1.12ⅹ10− work
the standard 2205 DSS. Standard 2205 DSS 1 − 0.12 0.94 1.90ⅹ10− 6
Ref [10]
The representative potentiodynamic polarization curves of the plate
FCWA-AM SDSS and the standard hot-rolled 2205 DSS [10], measured
under the same experimental condition, were shown in Fig. 8. Their
pitting corrosion resistance parameters were listed in Table 4. The passivation current density (Ipass) than the standard 2205 DSS, because
FCWA-AM SDSS showed comparable corrosion potential (Ecorr) and its passive film was thicker and had higher charge transfer resistance. In
pitting potential (Epit) with the hot-rolled 2205 DSS. In the passivation the potential range of 0.85–0.99V/SCE, several small current peaks were
potential range of 0.18–0.85V/SCE, the FCWA-AM SDSS had a lower noticed, indicating the appearance of the metastable pits on the
FCWA-AM SDSS surface. The occurrence of the metastable pits was

Fig. 8. Pitting corrosion results of the FCWA-AM SDSS (a) potentiodynamic polarization curves; (b) surface morphology after the potentiodynamic polarization test.

8
Y. Zhang et al. Materials Science & Engineering A 854 (2022) 143809

related to sensitization of the WA boundaries. As demonstrated in

σ ssDSS = σ ssα Vα + σ ssγ (1 − Vα ) (8)
Fig. 6d–f that the precipitation of Cr2N particles formed a Cr-deplete
zone at the WA grain boundaries. The stability and thickness of the
where Vα is the volume fraction of ferrite and σ ss ss
α , σ γ are the yield
passive film formed on Cr-deplete zone was worse than on the matrix. As
a result, when the sample was polarized to the potential higher than strength contributed from solid solution, which can be calculated using
0.85V/SCE, the passive film on the Cr-deplete zone were easier to be Equation (9) and Equation (10). All elements were calculated using the
attacked than the matrix, resulting in the formation of the metastable unit of mass percent.
pits on the sample and the sudden increase of the current density in the σ ssα = 1103.45C + 1103.45N + 25.8Si + 19.2Ni + 16.9Mn + 15.9Mo + 2.6Cr
potentiodynamic polarization curve. However, because of the high Cr (9)
content in the FCWA-AM SDSS, the metastable pits were immediately
repassived, leading to the sudden decrease of the current density [36]. σ ssγ = (1103.45C+1103.45N+25.8Si+19.2Ni+16.9Mn+15.9Mo+2.6Cr)2/3
When the potential was beyond 0.99V/SCE, the repassivation rate of the
(10)
pits was slower than the dissolution rate, resulting in the propagate of
the pits and the drastic current increase of the potentiodynamic polar­ Based on Equations (8)-(10), it was calculated that solid solution
ization curve. As presented in Fig. 8b, pits were found distributing at the contributed 422 MPa out of 809 MPa of the yield strength in horizontal
grain boundaries of the WA, which further confirmed the pitting direction of the FCWA-AM SDSS. In GMA-WAAM ER2209 DSS reported
mechanism. Pitting corrosion is a type of localized corrosion in which in Ref. [38], solid solution contributed 181 MPa out of 507 MPa of the
intense attack at localized areas on surface of components, while the yield strength in horizontal direction. Thus, the enhanced solid solution
remaining area of the surface corrodes at a much slower rate. For the effect and the balanced α/γ ratio contributed to approximately 80% of
FCWA-AM SDSS, the localized Cr depleted zone at the WA boundary the improvement of the yield strength. That is to say, the significant
were easier to be attacked and played a role in the initiation and improvement of strength was mainly attributed to the balanced α/γ ratio
propagation of corrosion pits. As a result, although a SDSS grade and the improved solid solution strengthening comparing with WAAM
component was additively manufactured in the research and its passive DSSs reported in former investigations.
film was thicker and stabler than the standard 2205 DSS, the improve­ Ref [17,19,38,39] have reported that the UTSs in vertical direction
ment of its pitting corrosion resistance was hinder by the Cr2N precip­ were about 10% lower than in horizontal direction. Zhang et al. [41],
itation, making its pitting potential comparable with the standard 2205 figured out that the interlayer bonding places in the additively manu­
DSS. Thus, our future investigations will be focused on the restriction of factured component easily become the fracture source during the tensile
Cr2N precipitation during the successive reheating of the FCWA-AM load in vertical direction, which leads to the weaker strength than in
process. horizontal direction. In this research, the robustness of the interlayer
bounding was enhanced by the deep penetration caused by the Pulsed
GMAW arc mode [42], resulting in a comparable strength in vertical and
3.5. Tensile properties
horizontal directions.

The representative strain-stress curves of the FCWA-AM SDSS in

4. Discussion
horizontal direction (Y) and vertical direction (Z) were presented in
Fig. 9. The yield strength (YS) and the ultimate tensile strength (UTS)
4.1. Growth mechanism of ferrite grains during FCWA-AM process
were presented in Table 5. Dimples were found on the fracture surface of
both tensile samples, suggesting a ductile fracture feature in horizontal
The growth mechanism of ferrite grains during the deposition pro­
and vertical direction. Comparing with the standard 2205 DSS [37] and
cess of the nth layer is illustrated in Fig. 10. Fig. 10a shows the condition
the WAAM DSSs using commercial welding wire as the deposition ma­
where totally n-1 layers have been deposited. Ferrite shows column
terial [17,19,38,39], the FCWA-AM SDSS had drastically higher YS and
structure in the AZ but exhibits equiaxed structure at the upper surface
UTS.
of the n-1st layer and at the interface between AZ and FZ. During the
As reported by Jiang et al. [40], The yield strength contribution from
deposition process of the nth layer, as presented in Fig. 10b, the equi­
the solid solution of DSS, σ ss
DSS , can be estimated using Equation (8)
axed ferrite at the surface of the n-1st layer is remelted by the pulsed
GMAW arc but the equiaxed ferrite grains at the interface between AZ
and FZ remain unmelted as they are at the bottom of the pass. The speed
of the plasma jet at the center of the pulsed GMAW arc is significantly
higher than it is at the edge. Thus, drastically higher plasma jet pressure
is exerted at the middle of the deposition pool than at the edge, forming
a finger zone (FZ) at the bottom center of the deposition pool. During the
spray metal transfer, the fine droplets are speed up by the high energy
density plasma jet that the initial acceleration can be 60 times higher
than the gravity [43]. The continuous fine droplet impact on the depo­
sition pool further increases the depth and width of the FZ. As reported
in Refs. [27,29,44], the undercooling degree in FZ is much higher than
in AZ. Moreover, Liu et al. [27] indicated that the heat dissipation
conditions in FZ is much better than AZ, thereby resulting in rapider
cooling rate in FZ. The higher undercooling and faster cooling speed
leads to a much rapider solidification rate in the FZ than in AZ. There­
fore, the growth rate of ferrite in the FZ is remarkably faster than it is in
the AZ (Fig. 10 d1). Because the nitrogen solution of α phase is only
about 1/4 of the liquid [45], large amount of N atoms escape from the
solid FZ to the liquid AZ during the rapid solidification, resulting in a
lower N content in FZ than AZ (Fig. 10 d2). As a result, only minor
Fig. 9. The strain-stress curve of the FCWA-AM SDSS in Y and Z directions. The amount of GBA will form in FZ in the subsequent solid-state trans­
insets show the fracture surface of the tensile test samples in both directions. formation. The concentration of solute and nucleation particles at the

9
Y. Zhang et al. Materials Science & Engineering A 854 (2022) 143809

Table 5
UTS and YS of the FCWA-AM SDSS.
Manufacturing method Raw Austenite ratio (%) Direction YS (MPa) UTS (MPa) Anisotropic percentage Data source
Material

FCWA-AM SDSS flux-cored wire in this work 45 horizontal 809 ± 12 961 ± 5 1.8% This work
vertical 822 ± 14 978 ± 4
Hot-rolling 2205 50 N/A 450 620 N/A ASTM A240 [37]
GMA-WAAM ER2209 73 horizontal 507 770 11.4% Ref [38]
vertical 480 682
FCWA-AM E2209 74 horizontal N/A 784 10.2% Ref [17]
vertical N/A 704
CMT-WAAM ER2209 75 horizontal N/A 837 9.7% Ref [39]
vertical N/A 756
Top-TIG-WAAM ER2594 98 horizontal 548 895 9.4% Ref [19]
vertical 520 810

Fig. 10. Schematic illustration of the ferrite grain growth mechanism during the FCWA-AM DSS process: (a) totally n-1 layers have been deposited; (b) deposition
process of the nth layer; (c) the deposition pool of the nth layer; (d) solidification process of the deposition pool of the nth layer; (e) totally, n layers have
been deposited.

solid/liquid interface increases during the solidification. The high 4.2. Microstructure evolution of austenite during FCWA-AM process
overcooling degree in FZ also promoted the equiaxial nucleation [44].
As a result, equiaxed ferrite are found nucleating at the upper center of The microstructure evolution of austenite during the FCWA-AM
the FZ (Fig. 10 d3). Comparing with the FZ, more heat input is accu­ process was illustrated in Fig. 11. Fig. 11 a showed the thermal his­
mulated in the AZ, resulting in a longer time staying in the high tem­ tory a deposition layer experiences during the FCWA-AM process. Cr2N
perature region [44]. Misaligned grains are outgrown by the ferrite particles form in the temperature range of 700–900 ◦ C [26]. During the
grains with their <001> paralleled with the thermal gradient, resulting cooling period, the transformation of α→GBA occurs firstly in the tem­
in the coarse columnar structure in the AZ (Fig. 10 d4). The concen­ perature range of 1350–800 ◦ C [46] and lath shaped WA nucleates and
tration of the solute and nucleation particle will also increase during the grows from the GBA with further decreasing of temperature. As reported
growth of the coarse columnar ferrite in AZ. As a result, equiaxed by Abbas et al. [47], WA forms in the temperature range between 800
nucleation was triggered at the end of the solidification of the deposition and 650 ◦ C. The formation of IGA requires more undercooling as the
pool and equiaxed ferrite grains can be found at the upper surface of the driving force compared to the GBA and WA because of its higher acti­
nth layer (Fig. 10 d5, e). vation energy for lattice diffusion, thus its formation temperature is
lower than 650 ◦ C.

10
Y. Zhang et al. Materials Science & Engineering A 854 (2022) 143809

Fig. 11. Microstructure evolution of austenite during the FCWA-AM process.

As presented in Fig. 11b and c, the FCWA-AM SDSS solidifies as type The peak temperature of the 1st reheating is approximately 1000 ◦ C,
F and virtually 100% ferrite forms when the deposition pool is cooled which is higher than formation temperature range of WA. Therefore, as
under the liquidus line (~1460 ◦ C). When the temperature decreases shown in Fig. 11i and j, the 1st time reheating triggered the initialization
lower than 1350 ◦ C, GBA firstly nucleates at the α/α boundaries of WA at the GBA when the temperature is cooled lower than 800 ◦ C and
(Fig. 11d). Because of the coarse grain structure of ferrite, the distance the precipitation of Cr2N particles at the WA boundaries in the tem­
between the GBAs is long. A lot of N atoms that do not have sufficient perature range of 700–800 ◦ C. Thus, in Figs. 11k and 5f, the initializa­
time to diffuse into the GBA will react with Cr atoms and form Cr2N tion of WA can be observed on GBA. The peak temperature of the 2nd
particles in ferrite grains when the temperature is cooled below 900 ◦ C reheating is approximately 800 ◦ C, which is within the formation tem­
(Fig. 11e). When 800 ◦ C is reached, side-plate WA starts to initiate from perature range of WA and Cr2N (Fig. 11L, m). As a result, WA grows
the GBA and grows into the ferrite grain. The nucleation and growth of rapidly after the 2nd reheating in the temperature range of 700–800 ◦ C
the WA is suppressed comparing with it is in WAAM DSSs using com­ (Figs. 11n and 5h) and Cr2N are found precipitating at the WA bound­
mercial welding wire because of the higher Creq/Nieq in the FCWA-AM aries (Figs. 11n and 6d). The peak temperatures of the reheating latter
SDSS. As a result, only minor amount of WA forms in the cooling than the 2nd are lower than 600 ◦ C, which generally have no influence
period as illuminated in Fig. 11f. At temperature lower than 650 ◦ C, the on the microstructure of the deposition layer. Thus, the microstructure
undercooling is high enough for the nucleation of IGA and the Cr2N in the bulk (Fig. 11 o) is similar to it is in the 2-time reheated layer.
particles distributing at the interior of ferrite grain provide massive
nucleation sites for IGA (Fig. 11g). As a result, IGA is the predominate 5. Conclusion
type of austenite while only minor amount of GBA and WA can be
observed in the as-built layer as presented in Figs. 11h and 5c. In this research, the chemical composition of a FCWA-AM SDSS was

11
Y. Zhang et al. Materials Science & Engineering A 854 (2022) 143809

designed based on its thermal history to achieve a balanced ferrite- [5] D. Zhang, A. Liu, B. Yin, P. Wen, Additive manufacturing of duplex stainless steels-
A critical review, J. Manuf. Process. 73 (2022) 496–517.
austenite ratio in the as-built state. The FCWA-AM SDSS showed a
[6] X. Guo, H. Li, Z. Pan, S. Zhou, Microstructure and mechanical properties of ultra-
combination of superior strength and corrosion resistance. The mecha­ high strength Al-Zn-Mg-Cu-Sc aluminum alloy fabricated by wire+ arc additive
nisms describing the evolution of ferrite and austenite during the FCWA- manufacturing, J. Manuf. Process. 79 (2022) 576–586.
AM process were also proposed. The major conclusions are summarized [7] N. Knezović, I. Garašić, I. Jurić, Influence of the interlayer temperature on
structure and properties of wire and arc additive manufactured duplex stainless
as below: steel product, Materials 13 (24) (2020) 5795.
[8] V.A. Hosseini, M. Högström, K. Hurtig, M.A. Valiente Bermejo, L.-E. Stridh,
(1) A SDSS with Creq/Nieq = 2.23 was successfully developed using L. Karlsson, Wire-arc additive manufacturing of a duplex stainless steel: thermal
cycle analysis and microstructure characterization, Weld. World 63 (4) (2019)
FCWA-AM. The as-built FCWA-AM SDSS had a duplex structure 975–987.
with an austenite ratio ~45%. The balanced α/γ ratio was formed [9] K. Zhang, J. Xiong, G. Zhang, Effect of heat treatment on microstructure and
in the IHT induced by the subsequent two layers. corrosion properties of double-wire GTA additive manufactured 308L stainless
steel, Mater. Lett. 309 (2022), 131362.
(2) Cr2N was found precipitating in the FCWA-AM SDSS. It acted as [10] Y. Zhang, F. Cheng, S. Wu, Improvement of pitting corrosion resistance of wire arc
the heterogeneous nucleation site of IGA during the cooling of the additive manufactured duplex stainless steel through post-manufacturing heat-
deposition layer and precipitated at the WA boundaries during treatment, Mater. Char. 171 (2021), 110743.
[11] J.C. Lippold, D.J. Kotecki, in: John C. Lippold, Damian J. Kotecki (Eds.), Welding
the IHT. Metallurgy and Weldability of Stainless Steels, Welding Metallurgy and Weldability
(3) The pulsed GMAW arc formed a FZ at the bottom of the deposi­ of Stainless Steels, Wiley-VCH, March 2005, 2005, ISBN 0-471-47379-0, p. 376,
tion layer. Its faster solidification rate causes the nitrogen escape 376.
[12] A. Rokanopoulou, P. Skarvelis, G. Papadimitriou, Welding design methodology for
from FZ to AZ, resulting in a lower austenite ratio than in AZ.
optimization of phase balance in duplex stainless steels during autogenous arc
Some equiaxed ferrite nucleated at the interface of FZ and AZ welding under Ar–N2 atmosphere, Weld. World 63 (1) (2019) 3–10.
because of the increasing of the concentration of solution and [13] N. M-630, Material Data Sheets for Piping, 2004. NORSOK M-630.
nucleation particles during solidification. [14] K. Davidson, S. Singamneni, Selective laser melting of duplex stainless steel
powders: an investigation, Mater. Manuf. Process. 31 (12) (2016) 1543–1555.
(4) A thicker and more stable passive film was formed on the surface [15] F. Hengsbach, P. Koppa, K. Duschik, M.J. Holzweissig, M. Burns, J. Nellesen,
of the FCWA-AM SDSS than hot-rolled 2205 DSS. The FCWA-AM W. Tillmann, T. Tröster, K.-P. Hoyer, M. Schaper, Duplex stainless steel fabricated
SDSS showed significantly higher strength than hot-rolled 2205 by selective laser melting-Microstructural and mechanical properties, Mater. Des.
133 (2017) 136–142.
DSS and previously reported WAAM DSSs as the result of the [16] B. Wittig, M. Zinke, S. Jüttner, Influence of arc energy and filler metal composition
balanced α/γ ratio and the enhanced solid solution strengthen on the microstructure in wire arc additive manufacturing of duplex stainless steels,
effect. Weld. World 65 (1) (2021) 47–56.
[17] Y. Zhang, F. Cheng, S. Wu, The microstructure and mechanical properties of duplex
stainless steel components fabricated via flux-cored wire arc-additive
CRediT authorship contribution statement manufacturing, J. Manuf. Process. 69 (2021) 204–214.
[18] G. Posch, K. Chladil, H. Chladil, Material properties of CMT—metal additive
manufactured duplex stainless steel blade-like geometries, Weld. World 61 (5)
Yiqi Zhang: Conceptualization, Data curation, Writing – original (2017) 873–882.
draft. Shaojie Wu: Supervision, Writing – review & editing, Funding [19] X. Zhang, K. Wang, Q. Zhou, J. Ding, S. Ganguly, G. Marzio, D. Yang, X. Xu,
acquisition, Project administration. Fangjie Cheng: Supervision, P. Dirisu, S.W. Williams, Microstructure and mechanical properties of TOP-TIG-
wire and arc additive manufactured super duplex stainless steel (ER2594), Mater.
Writing – review & editing, Funding acquisition, Project administration. Sci. Eng., A (2019), 138097.
[20] P.P. Nikam, D. Arun, K.D. Ramkumar, N. Sivashanmugam, Microstructure
characterization and tensile properties of CMT-based wire plus arc additive
Declaration of competing interest manufactured ER2594, Mater. Char. 169 (2020), 110671.
[21] M.L. Köhler, J. Kunz, S. Herzog, A. Kaletsch, C. Broeckmann, Microstructure
analysis of novel LPBF-processed duplex stainless steels correlated to their
The authors declare that they have no known competing financial mechanical and corrosion properties, Mater. Sci. Eng., A 801 (2021), 140432.
interests or personal relationships that could have appeared to influence [22] J. Norrish, Advanced Welding Processes, Springer Science & Business Media, 1992.
[23] Y. Zhang, S. Wu, F. Cheng, A duplex stainless steel (DSS) with striking tensile
the work reported in this paper. strength and corrosion resistance produced through wire arc-additive
manufacturing (WAAM) using a newly developed flux-cored wire, Mater. Lett.
Data availability (2022), 131760.
[24] G. Liu, S. Han, X. Tang, H. Cui, Effects of torch configuration on arc interaction
behaviors and weld defect formation mechanism in tandem pulsed GMAW,
Data will be made available on request. J. Manuf. Process. 62 (2021) 729–742.
[25] L. Jinlong, L. Tongxiang, D. Limin, W. Chen, Influence of sensitization on
microstructure and passive property of AISI 2205 duplex stainless steel, Corrosion
Acknowledgement Sci. 104 (2016) 144–151.
[26] J. Verma, R.V. Taiwade, Effect of welding processes and conditions on the
This program is supported by the National Natural Science Founda­ microstructure, mechanical properties and corrosion resistance of duplex stainless
steel weldments—a review, J. Manuf. Process. 25 (2017) 134–152.
tion of China (Grant No. 51775372, Grant No. 52005366). [27] D. Liu, D. Wu, R. Wang, J. Shi, F. Niu, G. Ma, Formation mechanism of Al-Zn-Mg-
Cu alloy fabricated by laser-arc hybrid additive manufacturing: microstructure
evaluation and mechanical properties, Addit. Manuf. 50 (2022), 102554.
Appendix A. Supplementary data
[28] H.H. Zargari, K. Ito, M. Kumar, A. Sharma, Visualizing the vibration effect on the
tandem-pulsed gas metal arc welding in the presence of surface tension active
Supplementary data to this article can be found online at https://doi. elements, Int. J. Heat Mass Tran. 161 (2020), 120310.
org/10.1016/j.msea.2022.143809. [29] D. Wu, D. Liu, F. Niu, Q. Miao, K. Zhao, B. Tang, G. Bi, G. Ma, Al–Cu alloy
fabricated by novel laser-tungsten inert gas hybrid additive manufacturing, Addit.
Manuf. 32 (2020), 100954.
References [30] A. Ramirez, S. Brandi, J. Lippold, Secondary austenite and chromium nitride
precipitation in simulated heat affected zones of duplex stainless steels, Sci.
Technol. Weld. Join. 9 (4) (2004) 301–313.
[1] I.A. 52900, Additive Manufacturing—General Principles—Terminology, 2021.
[31] Z. Zhang, H. Jing, L. Xu, Y. Han, L. Zhao, X. Lv, J. Zhang, Influence of heat input in
[2] M. Köhler, L. Sun, J. Hensel, S. Pallaspuro, J. Kömi, K. Dilger, Z. Zhang,
electron beam process on microstructure and properties of duplex stainless steel
Comparative study of deposition patterns for DED-Arc additive manufacturing of
welded interface, Appl. Surf. Sci. 435 (2018) 352–366.
Al-4046, Mater. Des. 210 (2021), 110122.
[32] A.H. Ettefagh, S. Guo, Electrochemical behavior of AISI316L stainless steel parts
[3] E.H. Valente, V.K. Nadimpalli, T.L. Christiansen, D.B. Pedersen, M.A. Somers, In-
produced by laser-based powder bed fusion process and the effect of post annealing
situ interstitial alloying during laser powder bed fusion of AISI 316 for superior
process, Addit. Manuf. 22 (2018) 153–156.
corrosion resistance, Addit. Manuf. Lett. 1 (2021), 100006.
[33] A. Chen, W. Hu, D. Wang, Y. Zhu, P. Wang, J. Yang, X. Wang, J. Gu, J. Lu,
[4] R. Casati, M. Coduri, M. Riccio, A. Rizzi, M. Vedani, Development of a high
Improving the intergranular corrosion resistance of austenitic stainless steel by
strength Al–Zn–Si–Mg–Cu alloy for selective laser melting, J. Alloys Compd. 801
high density twinned structure, Scripta Mater. 130 (2017) 264–268.
(2019) 243–253.

12
Y. Zhang et al. Materials Science & Engineering A 854 (2022) 143809

[34] S. Salahi, A.V. Nemani, M. Ghaffari, J. Lunde, A. Nasiri, On microstructure, [40] D. Jiang, X. Gao, Y. Zhu, C. Hutchinson, A. Huang, In-situ duplex structure
crystallographic orientation, and corrosion properties of wire arc additive formation and high tensile strength of super duplex stainless steel produced by
manufactured 420 martensitic stainless steel: effect of the inter-layer temperature, directed laser deposition, Mater. Sci. Eng., A 833 (2022), 142557.
Addit. Manuf. 46 (2021), 102157. [41] K. Zhang, S. Wang, W. Liu, X. Shang, Characterization of stainless steel parts by
[35] M.E. Orazem, I. Frateur, B. Tribollet, V. Vivier, S. Marcelin, N. Pébère, A.L. Bunge, laser metal deposition shaping, Mater. Des. 55 (2014) 104–119.
E.A. White, D.P. Riemer, M. Musiani, Dielectric properties of materials showing [42] M. Benakis, D. Costanzo, A. Patran, Current mode effects on weld bead geometry
constant-phase-element (CPE) impedance response, J. Electrochem. Soc. 160 (6) and heat affected zone in pulsed wire arc additive manufacturing of Ti-6-4 and
(2013) C215. Inconel 718, J. Manuf. Process. 60 (2020) 61–74.
[36] B. Jegdić, B. Bobić, B. Radojković, B. Alić, L. Radovanović, Corrosion resistance of [43] A. Zabirov, Investigation of Droplet Formation and Transfer in GMA Welding,
welded joints of X5CrNi18-10 stainless steel, J. Mater. Process. Technol. 266 Dissertation, Rheinisch-Westfälische Technische Hochschule Aachen, 2016,
(2019) 579–587. p. 2016.
[37] A. A240, Standard Specification for Chromium and Chromium-Nickel Stainless [44] M. Gao, X. Zeng, J. Yan, Q. Hu, Microstructure characteristics of laser–MIG hybrid
Steel Plate, Sheet, and Strip for Pressure Vessels and for General Applications, welded mild steel, Appl. Surf. Sci. 254 (18) (2008) 5715–5721.
2013. [45] I. Woo, Y. Kikuchi, Weldability of high nitrogen stainless steel, ISIJ Int. 42 (12)
[38] F. Hejripour, F. Binesh, M. Hebel, D.K. Aidun, Thermal modeling and (2002) 1334–1343.
characterization of wire arc additive manufactured duplex stainless steel, J. Mater. [46] Z. Zhang, H. Jing, L. Xu, Y. Han, L. Zhao, C. Zhou, Effects of nitrogen in shielding
Process. Technol. 272 (2019) 58–71. gas on microstructure evolution and localized corrosion behavior of duplex
[39] K. Qi, R. Li, Z. Hu, X. Bi, T. Li, H. Yue, B. Zhang, Forming appearance analysis of stainless steel welding joint, Appl. Surf. Sci. 404 (2017) 110–128.
2205 duplex stainless steel fabricated by cold metal transfer (CMT) based wire and [47] A. Eghlimi, M. Shamanian, K. Raeissi, Effect of current type on microstructure and
arc additive manufacture (WAAM) process, J. Mater. Eng. Perform. (2022) 1–11. corrosion resistance of super duplex stainless steel claddings produced by the gas
tungsten arc welding process, Surf. Coating. Technol. 244 (2014) 45–51.

13