Additive Manufacturing 22 (2018) 153–156

Contents lists available at ScienceDirect

Additive Manufacturing
journal homepage: www.elsevier.com/locate/addma

Full Length Article

Electrochemical behavior of AISI316L stainless steel parts produced by laser- T

based powder bed fusion process and the effect of post annealing process
⁎
Ali Hemmasian Ettefagh , Shengmin Guo
Department of Mechanical and Industrial Engineering, Louisiana State University, Baton Rouge, LA, 70803, USA

A R T I C LE I N FO A B S T R A C T

Keywords: This paper presents the investigation of the corrosion behavior of AISI316L samples prepared by laser-based
Stainless steel powder bed fusion additive manufacturing (AM) method. Both AM and conventional stainless steel 316L samples
Additive manufacturing were examined in NaCl 3.5% solution before and after the annealing process using Tafel curves, Electrochemical
3D printing Impedance Spectroscopy, and X-ray diffraction. The results indicate that the AM parts have an improved cor-
Laser-based powder bed fusion
rosion behavior than the conventional wrought samples. Besides, the heat treatment process is found to further
Corrosion
decrease the corrosion rate of the AM parts through the relieving of the residual stress. In contrast, the post
EIS
Polarization annealing induced improvement to corrosion resistance for the wrought samples is due to the elimination of
Passive films martensite phase which almost always exists after the plastic deformation during their production process.

1. Introduction layer to form a 3D structure based on a computer aided design (CAD)

design [9–12]. This technique has been used to build physical models,
From the corrosion behavior point of view, stainless steels (SS) are a prototypes, tooling components, and functional parts with complex
key type of industrial alloys [1,2]. The corrosion resistance of stainless shapes [13,14].
steels is mainly attributed to the presence of at least 11 wt% of chro- Although laser-based powder bed fusion AM parts provide the
mium, which forms a protective layer of chromium oxide. The char- mentioned advantages compared to the other methods, high residual
acteristic of the SS316L protective layer can be found in open literatures stress due to the repeated heating and rapid solidification is one of the
[3]. Stainless steels also have many other alloying elements such as drawbacks of this method [15]. As a remedy, annealing process is al-
nickel, molybdenum, copper, and aluminum that could affect the most inevitable after the manufacturing step [16]. Despite the nu-
electrochemical behavior of SS parts under a variety of working en- merous works on determining the corrosion behavior of AISI316L, the
vironments, along with the mechanical properties and phase stabilities studies on the corrosion behaviors of laser AM parts are still limited. For
[1,4]. Among numerous types of SS, AISI316L is used extensively in laser-based powder bed fusion related corrosion studies, the effect of
different industries, such as pharmaceutical, petrochemical, offshore changing printing parameters to the resulting corrosion performance,
drilling, marine shipping, water desalination, etc. [5]. AISI316L SS is an the localized corrosion behaviors, and the passive layer formation are
austenitic stainless steel, which provides a good resistance to the interesting research topics [17–19]. In addition, the evaluation of the
chlorine attacks in marine environments [6]. effect of heat treatment is mainly focused on the mechanical properties
Additive manufacturing (AM) has the ability to produce complex in the previous works [13,16,20]. The aim of this paper is to examine
shaped parts. Small tolerances and automated process are the addi- the corrosion performances of laser-based powder bed fusion parts
tional advantages of this technique, in comparison to other conven- before and after the annealing process for marine applications. The
tional manufacturing methods like machining, welding, forging, and general corrosion rate of these AM parts is compared with conventional
casting [7,8]. Laser-based powder bed fusion process is a laser AM commercial AISI316L samples by the means of Tafel curves, as well as
method, which promises increased economy and speed when produc- Electrochemical Impedance Spectroscopy (EIS) measurements.
tion volumes and batch sizes are small. Using fine metal/alloy powders
(10–45 μm in diameter), laser-based powder bed fusion process can 2. Material and methods
produce complex shaped components that cannot be easily produced by
conventional methods. Employing a high power laser source, laser- A Concept-Laser Mlab-cusing-R system was used in this study to
based powder bed fusion process makes 3D metal/alloy parts layer by prepare the samples for the corrosion testing. Austenitic stainless steel

⁎
Corresponding author.
E-mail address: [email protected] (A. Hemmasian Ettefagh).

https://doi.org/10.1016/j.addma.2018.05.014
Received 4 September 2017; Received in revised form 3 May 2018; Accepted 7 May 2018
Available online 08 May 2018
2214-8604/ © 2018 Elsevier B.V. All rights reserved.
A. Hemmasian Ettefagh, S. Guo Additive Manufacturing 22 (2018) 153–156

Table 1 tendency to form metal ions and corrosion process.

- Chemical composition (wt%)of AISI316L stainless steel. On the other hand, for the Tafel curves for all four cases, a wide
Cr Ni Mo Mn Si P C S Fe range of potential with almost constant values in current density can be
detected, which represents an active-passive behavior. The constant
Min 16.5 10.0 2.0 0.0 0.0 0.0 0.0 0.0 Balance current density suggests the same corrosion mechanism on the surfaces
Max 18.5 13.0 2.5 2.0 1.0 0.045 0.030 0.030 Balance
of all samples, even though the stability range can be different. The
mentioned range is identified on Fig. 1 as ΔE on the heat-treated AM
sample as an example and the related values are reflected in Table 2. In
316L powders from Concept-Laser (CL 20ES) were used, together with
addition, it can be seen that there exists a sudden increase of the current
the Concept-Laser’s “Speed-cusing” processing parameters. The process
density just out of the ΔE range (passive layer break down point). This
profile “Speed-cusing” is optimized by Concept-Laser for CL 20ES
occurs on a more positive value of potential for the heat-treated samples
SS316L powders. Being a proprietary file, general users are not allowed
compared to the as received ones, and for the AM samples compared to
to view and modify the settings. The chemical compositions of
the wrought ones. These results suggest a more stable passive film for
AISI316L are listed in Table 1 (particle size: 10–45 μm). Upon creation
the AM samples, which has become more stable after the annealing
of a CAD model of the desired geometry, two-dimensional slicing and
process.
support structure information are created using the AutoFab software
EIS measurements were subsequently performed in order to confirm
for input into the Mlab system. Before the building process, the Mlab
this finding. The Nyquist plots of the samples are depicted in Fig. 2.
system is purged with argon (Ar) down to an oxygen level less than
According to these plots, typical capacitive semi-loop shapes are similar
0.7%. This oxygen level is maintained throughout the AM process. For
for all samples but with different diameters. The trend of these curves is
each layer, after lower the building platform by 25 microns, a re-coater
caused by a combined effect of electric double layer capacitance and
blade moves across the build platform and deposits an even layer of SS
the charge transfer resistance. To find the values of these parameters, it
316L powders onto the build platform. After the rectangular shaped
is necessary to fit these curves with the equivalent circuit. Fig. 2 also
plates were built, a Mitsubishi wire EDM machine (MV series) was used
shows the proposed circuit for this situation, which best fits the results
to cut out a small section for corrosion study.
and consists of RΩ (electrolyte resistance), RP (passive layer resistance),
Wrought samples with the same chemical compositions were bought
CP (capacitive properties of passive layer), RCT (charge transfer re-
from a commercial source, in order to perform a comparative study.
sistance of electrochemical double layer), and CDL (capacitive proper-
Corrosion studies were carried out in a standard 3-electrode cell in-
ties of electrochemical double layer). This type of equivalent circuit
cluding a reference saturated calomel electrode (SCE), a platinum
represents an electrode with a protective coating layer. The parameters
electrode as the counter electrode, and the working electrode with the
for such an equivalent circuit can be found in Table 3 and their trend
exposed surface area of 1 cm2. The solutions used for all of the tests
will be discussed in the next paragraph. Based on the Nyquist plots, it is
were 3.5 wt% NaCl with naturally dissolved oxygen at room tempera-
confirmed that the mechanism of corrosion reaction for all samples are
ture. Annealing process was performed on both AM and wrought
the same and mainly controlled by the charge transfer phenomena
samples in a Nabertherm GmbH N 7/H Furnace under argon atmo-
through the protective layer.
sphere. Samples were annealed in 800 °C for 2 h and then cooled down
The results indicate that RΩ is low and almost identical for all
in the furnace atmosphere.
testing cases, which is consistent with the observed reproducibility of
The Tafel curves were obtained by scanning the potential from -0.75
the test setup. The fact that CP has decreased for the AM annealed
VSCE to +0.5 VSCE with the scan rate of 1.67 mV/s with a CHI-604C
sample refers to a better protection of the passive layer. Both RP and RCT
instrument. Before the tests, samples were abraded by the grit size
are higher for the as received and annealed AM parts, which also in-
320–1000 sand papers and then washed and immersed freely in the salt
dicate more resistance for the mass transfer in the diffusion and double
solution for 15 min. EIS measurements were performed after obtaining
layer (passive film) respectively. Specifically, the higher value of RCT
Tafel curves at E = EOC in the range of 10 mHz to 100 kHz and the
for the AM samples compared to the wrought ones suggests a better
equivalent circuits were generated using the electrochemical analyzer
quality of passive film. This protective layer is formed on the AM parts’
software of CHI-604C device. The phases of the samples were examined
surface. This finding agrees with the Tafel curves where the passivation
by XRD (PANalytical Empyrean XRD) using CuKα radiation with the
ranges of the AM samples are wider (ΔE in Table 2). Similar to the Tafel
scan step size of 0.039 degrees and 2θ ranging from 40° to 95° at room
curves, the differences between the parameters of annealed and as re-
temperature.
ceived AM parts are relatively small. This suggests that the mechanism
associated with the corrosion rate decrease after the heat treatment
3. Results and discussion process is not dominated by the quality of passive layer or phase
change. In contrast, the as received and the heat-treated wrought
Tafel curves for the wrought and AM samples before and after the samples show a large difference.
annealing process are presented in Fig. 1. It clearly indicates that the Fig. 3 depicts the XRD profiles of the wrought and AM samples
corrosion rate of the AM sample is less than that of the wrought one before and after the annealing process. It is observable that the wrought
before the post heat treatment process and the trend is the same after sample contains two phases, which are γ (Austenite) and α´ (marten-
annealing with a lower intensity. In other words, as shown in Fig. 1, the site), whereas the AM samples have no martensite phase. The formation
AM parts’ curve is shifted to lower values of current density compared of martensite phase can be attributed to the plastic deformation during
to the wrought sample. On the other hand, although the curves for both the manufacturing process of the wrought sample. It is well known that
samples are shifted to lower values of current density after the an- plastic deformation can induce the martensite phase and the martensite
nealing process, the changes are more pronounced for the wrought phase can deteriorate the corrosion behavior along with the passivation
sample. The related key parameters measured from these polarization capability of the surface as a result of the formed dislocations and in-
curves can be found in Table 2. Those results are attained by Tafel ternal stress [21]. After the heat treatment of the wrought sample, the
extrapolation. As can be seen in Table 2, icorr, which represents the intensity of the (111) martensite phase decreased and (201) and (211)
corrosion rate of the samples, decreased for both wrought and the AM peaks disappeared, suggesting the reduction of non-equilibrium phase
samples after heat treatment and the best result belongs to the heat quantity in the structure, which can be considered as an explanation for
treated AM sample. In addition, the corrosion potential (Ecorr) is more a better corrosion behavior of heat-treated wrought sample compared
positive after heat treatment process for both wrought and printed to the as received one. But in case of AM samples, the detected phases
samples compared to the as-received ones, which indicates the less for both annealed and as received specimens are the same.

154
A. Hemmasian Ettefagh, S. Guo Additive Manufacturing 22 (2018) 153–156

Fig. 1. Tafel curves obtained in 3.5% NaCl Solution. The indicated ΔE range refers to the passive layer stability potential range for the annealed AM (printed) sample.

Table 2 Table 3
Corrosion parameters attained from polarization curves and Tafel extrapola- Measured parameters from the simulation of Nyquist plots with an electric
tion. equivalent circuit shown in Fig. 2.
Sample Ecorr icorr ΔE Sample RΩ RP CP RCT CDL
(v) (μA/cm2) (v) (Ωcm2) (Ωcm2) (μF/cm2) (Ωcm2) (μF/cm2)

Wrought −0.471 4.16 0.463 Wrought 7.92 710 86.3 6514 93.9
Wrought (Heat-Treated) −0.434 2.69 0.564 Wrought (Heat-Treated) 8.34 1002 56.2 9636 31.4
Printed −0.362 1.29 0.609 Printed 8.83 1239 84.3 13038 177.0
Printed (Heat-Treated) −0.347 1.14 0.613 Printed (Heat-Treated) 10.71 1326 42.9 14285 70.2

Based on the information obtained in the XRD patterns, the Tafel the corrosion behavior despite its improvement compared to as re-
curves can be reinterpreted from the corrosion potential aspect. Fig. 1 ceived wrought sample. In contrast, the absence of this martensite
clearly shows that the corrosion potential of the AM samples are in the phase in the AM sample led to a more positive corrosion potential,
range of more positive values compared to wrought samples and this is along with a decrease in the corrosion rate. After annealing for 2 h in
more obvious for the heat-treated AM sample. This increase in corro- 800 °C, the residual stresses associated with the fast heat up and rapid
sion potential can be credited to the lack of martensite phase in the AM solidification during the AM process are relieved and this is the main
sample. As the corrosion potential is dominated by the fastest reaction, reason of the slight improvement of the corrosion behavior after the
in the case of commercial wrought AISI316L testing, the anodic reac- heat treatment of AM parts. In other words, the presence of this stress
tion is governed by the corrosion of a less resistant phase, which is leads to the formation of corrosion cells within the structure with the
martensite in the wrought sample. This phase has decreased after the stressed zones acting as the anodic regions. Despite this mechanism not
annealing process but still exists in the wrought sample and deteriorates being the controlling regime of the corrosion of AISI316 AM parts, the

Fig. 2. Nyquist plots of wrought and AM (printed) AISI316L samples before and after the annealing process in 3.5 wt% NaCl and the electric equivalent circuit used to
fit the curves.

155
A. Hemmasian Ettefagh, S. Guo Additive Manufacturing 22 (2018) 153–156

award #OIA-1541079.

References

[1] A.J. Sedriks, Corrosion of Stainless Steel, 2nd edition, Wiley-Interscience, New
York, 1996.
[2] P. Henry, J. Takadoum, P. Bercot, Tribocorrosion of 316L stainless steel and TA6V4
alloy in H2SO4 media, Corros. Sci. 51 (no. 6) (2009) 1308–1314.
[3] J.J. Kim, Y.M. Young, Study on the passive film of type 316 stainless steel, Int. J.
Electrochem. Sci. 8 (2013) 11847–11859.
[4] J.R. Davis, ASM Specialty Handbook - Stainless Steels, ASM International, 1994.
[5] H. Alves, D.C. Agarwal, H. Werner, Alloys Suitable for Phosphoric Acid
Applications, NACE International, San Diego, 2006.
[6] J. Chen, Q. Zhang, Q.A. Li, S.L. Fu, J.Z. Wang, Corrosion and tribocorrosion be-
haviors of AISI 316 stainless steel and Ti6Al4V alloys in artificial seawater, Trans.
Nonferrous Metals Soc. China 224 (2014) 1022–1031.
[7] W. Gao, Y. Zhang, D. Ramanujan, K. Ramani, Y. Chen, C.B. Williams, C.C.L. Wang,
Y.C. Shin, The status, challenges, and future of additive manufacturing in en-
gineering, Comput. Aided Des. 69 (2015) 65–89.
[8] D. Manfredi, F. Calignano, M. Krishnan, R. Canali, E.P. Ambrosio, S. Biamino,
D. Ugues, M. Pavese, Additive manufacturing of Al alloys and aluminium matrix
composites (AMCs), Light Metal Alloys Applications, InTech, 2014 Chapter 1.
[9] D. Herzog, V. Seyda, E. Wycisk, C. Emmelmann, Additive manufacturing of metals,
ActaMaterialia 117 (2016) 371.
[10] K. Zhang, W. Liu, X. Shang, Research on the processing experiments of laser, Opt.
Laser Technol. 39 (2007) 549.
[11] R.M. Mahamood, E.T. Akinlabi, Laser Additive Manufacturing Advanced
Manufacturing Techniques Using Laser Material Processing, IGI Global book series
Advances in Civil and Industrial Engineering (ACIE), Pennsylvania, 2016.
Fig. 3. XRD profile of the AISI316L AM (printed) and wrought samples, before
[12] D. Pham, S.S. Dimov, Rapid Manufacturing: the Technologies and Applications,
and after the annealing process. ■ bullets indicate α´ (martensite) phase while Springer, London, 2001.
▼ bullets represent the γ (Austenite) phase. [13] E. Yasa, J.P. Kruth, Microstructural investigation of selective laser melting 316L
stainless steel parts exposed to laser re-melting, Procedia Eng. 19 (2011) 389–395.
[14] Y. Zhong, L. Liu, S. Wikman, D. Cui, Z. Shen, Intergranular cellular segregation
elimination of the residual stress can lead to a less susceptible passive network structure strengthening 316L stainless steel prepared by selective laser
film in a non-stressed situation and a slight decrease of corrosion rate as melting, Nucl. Mater. 470 (2016) 170.
[15] P. Mercelis, J.P. Kruth, Residual stresses in selective laser sintering and selective
a result. This finding is in agreement with the previous researches on laser melting, Rapid Prototyp. 12 (no. 5) (2006) 254–265.
the characteristic of passive film of austenitic stainless steels [22–23]. [16] M.L. Montero Sistiaga, S. Nardone, C. Hautfenne, J. Van Humbeeck, Effect of heat
treatment of 316L stainless steel produced by selective laser melting (SLM), Austin,
Texas, USA, the 27th Annual International Solid Freeform Fabrication Symposium,
4. Conclusions
2016.
[17] G. Sander, S. Thomas, V. Cruz, M. Jurg, N. Birbilis, X. Gao, M. Brameld,
The corrosion behavior of AISI316L samples produced via laser- C.R. Hutchinson, On the corrosion and metastable pitting characteristics of 316L
stainless steel produced by selective laser melting, J. Electrochem. Soc. 164 (no. 6)
based powder bed fusion AM method was investigated before and after
(2017) 250–257.
the annealing process and compared to the conventional wrought [18] Y. Sun, A. Moroz, K. Alrbaey, Sliding wear characteristics and corrosion behaviour
samples of the same alloy and heat treatment cycle. Results of elec- of selective laser melted 316L stainless steel, J. Mater. Eng. Perform. 23 (2014) 518.
trochemical tests, including Tafel curves and EIS measurements, [19] M. Zietala, T. Durejko, M. Polanski, I. Kunce, T. Plocinski, W. Zielinski, M. Lazinska,
W. Stepniowski, T. Czujko, K. Kurzydlowski, Z. Bojar, The microstructure, me-
showed a decrease in the corrosion rate for the AM samples due to the chanical properties and corrosion resistance of 316L stainless steel fabricated using
ability to form a thicker and more stable protective layer on the surface laser engineered net shaping, Mater. Sci. Eng. A 677 (2016) 1.
despite the same corrosion mechanism of both types of samples. The [20] I. Tolosa, F. Garciandía, F. Zubiri, F. Zapirain, A. Esnaola, Study of mechanical
properties of AISI 316 stainless steel processed by ``selective laser melting’’, fol-
corrosion behavior of heat-treated AM sample is slightly better due to lowing different manufacturing strategies, Adv. Manuf. Technol. 51 (5-8) (2010)
the elimination of residual stress after the annealing process. Post heat 639–647.
treatment results in a lower corrosion rate in the wrought sample [21] L. Jinlong, Y. Meng, H. Miura, L. Tongxiang, The effect of surface enriched chro-
mium and grain refinement by ball milling on corrosion resistance of 316L stainless
mainly due to the reduction of non-equilibrium martensite phase in the steel, Mater. Res. Bull. 91 (2017) 91–97.
structure. Coupling Tafel curves, Electrochemical Impedance [22] A. Nazarov, V. Vivier, D. Thierry, F. Vucko, B. Tribollet, Effect of mechanical stress
Spectroscopy, and XRD results, it’s confirmed that, for both wrought on the properties of steel surfaces: scanning Kelvin probe and local electrochemical
impedance study, Electrochem. Soc. 164 (no. 2) (2017) 66–77.
and AM samples, heat treatment is a positive factor on the corrosion
[23] O. Takakuwa, H. Soyama, Effect of residual stress on the corrosion behavior of
behavior with two different mentioned mechanisms. austenitic stainless steel, Adv. Chem. Eng. Sci. 5 (2015) 62–71.

Acknowledgment

The current work is funded by the NSF EPSCoR CIMM project under

156