Available online at www.sciencedirect.

com

ScienceDirect
http://ees.elsevier.com/stmat
Science and Technology of Materials 30 (2018) 51–59

Evaluation of the effect of the thermal cycle on the characteristics of welded

joints through the variation of the heat input of the austhenitic AISI 316L
steels by the GMAW process
R. Demarque, E.P. dos Santos, R.S. Silva, J.A. de Castro
Vila Santa Ceciclia, Volta Redonda, Brazil
Received 5 August 2017; accepted 25 September 2018

Abstract
Welding processes are generally used in almost every industrial segment, but there are some applications that demand special attention to the
product obtained, due to the security it must offer against possible failures. In the present study, the influence of the heat input variation on the weld
bead geometry, the austenitic grain size in the HAZ, the microhardness in the HAZ (heat affect zone), and the ␦ ferrite volumetric fraction using
the gas metal arc welding (GMAW) performed with pulsed chain spray transfer, on AISI316L steel plates. The samples had their microstructure
characterized by optical microscopy, in their different regions, performing a comparative study of these regions for the different welding conditions.
From the variations in the thermal contributions it was demonstrated that the choice of the ideal heat input for a given process has an important
effect on the phenomena that occur during welding. The increase of the contribution caused a variation of almost 300% in the austenitic grain size,
approximately 100% in the ␦ ferrite volumetric fraction, and about 240% in the area of the fusion zone (FZ). The temperature distributions of the
plates were recorded using a thermal camera in order to monitor the maximum temperatures in each region of the welded joint, the microstructures
resulting from the process were later compared with the results of the thermal cycles obtained through numerical simulations. The results showed
good agreement and also contributed to the validation of the proposed numerical method, which allowed a better understanding of the thermal
history during welding and, consequently, of their effects on the microstructural transformations developed in the AISI 316L steel.
© 2018 Sociedade Portuguesa de Materiais (SPM). Published by Elsevier España, S.L.U. All rights reserved.

Keywords: GMAW welding; Thermal cycle; Stainless steel; Heat affected zone; Melted zone

1. Introduction Austenitic stainless steels contain between 6 and 26% Ni,

16 and 30% Cr and less than 0.30% carbon, with a total alloy
Stainless steels can be defined as ferrous alloys containing content of at least 26% [2]. Its characteristics are low flow
Fe as the base element, chromium as the main alloying element, limit, high ductility, high coefficient of thermal expansion and
additions of nickel and, in some cases, other alloying elements mechanical resistance, low thermal diffusivity and are among
[1]. The designation of stainless steel corresponds only to Fe–Cr the stainless, which have better weldability and general resis-
or Fe–Cr–Ni alloys containing more than 10–12% chromium [2]. tance to corrosion. However, they may suffer from pit corrosion
The chemical composition of these steels, together with the in environments containing chlorides.
thermomechanical processing to which it is subjected, results in Considering the purpose of the present study, austenitic stain-
different microstructures at room temperature, so that through less steels can be applied because they have higher hardness and
these microstructures resulting from the process stainless steels ductility than most other steels and are kept at very low tempera-
can be classified into four classes: ferritic stainless steels 400 tures. Therefore, these steels are considered for using in welded
series (200 series and 300, including superaustenitic) and some structures to be used at the helium condensation temperature
others, such as duplex stainless steels and precipitation curable (4 K or −269 ◦ C), such as the magnet structure in experimental
steels [3]. nuclear fusion reactors.

https://doi.org/10.1016/j.stmat.2018.09.001
2603-6363/© 2018 Sociedade Portuguesa de Materiais (SPM). Published by Elsevier España, S.L.U. All rights reserved.
52 R. Demarque et al. / Science and Technology of Materials 30 (2018) 51–59

However, when subjected to a severe thermal cycle, these

properties can be altered and even lost, compromising the appli-
cation of steel. For these reasons there are a constant need to
conduct research and applied studies and continuously collab-
orating to increase the theoretical and experimental knowledge
of the issue addressed in the present study.

2. Experimental work

2.1. Materials and methods

The materials used in the experimental welding process of Fig. 1. Example of weld bead deposition on the sheet.
the present work were three sheets of austenitic stainless steel
AISI 316L. Tables 1 and 2 show, respectively, the dimensions Table 4
Subdivision of the sheets to be welded in the GMAW process with pulsed metal
of the plates analyzed and the chemical analysis of the steel. transfer for different bands of thermal input (predetermined values).
It is important to mention that the materials, after the man-
ufacturing process, were subjected to adequate solubilization Material Welding process GMAW Thermal input bands (kJ/mm)
heat treatment in order to dissolve formed precipitates during AISI Metal transfer 0.4–0.7
the working and shaping processes. However, no information is 316L Pulsed spray 1.3–1.7
available from suppliers regarding temperatures and solubiliza- 3 Plates 2.4– 2.7
tion times. Therefore in this study we carried out as received
analysis for understanding the initial microstructure and propri-
eties. 2.3. Welding

Considering as main objective of this study the microstruc-

2.2. Welding consumables tural analysis and comparison resulting from the GMAW
welding process with pulsed metal transfer for the same heat
The addition wire used was ER316L, which is an extra low input range and for the same steel, the welding of the materials
carbon electrode, which is know as avoiding the sensitization was carried out following the ranges shown in Table 4.
phenomenon, type CrNiMo 19/12/3 for welding of austenitic The heat input ranges presented in Table 4 were chosen in
stainless steels type AISI 316L. Table 3 shows the chemical order to study a lower limit than that technological recommenda-
composition of the wire. tion for the welding of these materials (0.4–0.7 kJ/mm), an upper
The welding of all plates was performed in a single pass and limit to the recommended one (1.3–1.7 kJ/mm) and a value well
with deposition of the addition material, forming a bead with a above the recommended range (2.4–2.7 kJ/mm). In addition, the
mean extension of 160 mm for AISI 316L along the length of parameters used in the welding of the three plates of the AISI
the plates, as shown in Fig. 1. 316L were all the same, the variation of the contribution being
obtained only by the variation of the welding speed. There are
Table 1 three samples, welded with GMAW process with pulsed metal
Dimensions of AISI 316L austenitic stainless steels. transfer, being AISI 316L steel one with low, one with interme-
Length (mm) Width (mm) Thickness (mm) diate and one with high heats input. In this way, comparisons
can be established among thermal welding cycles, cooling rates
AISI 316L 215 60 10
and final obtained microstructures, Vickers microhardness pro-
files, melt zone dimensions, delta ferrite quantification in the
Table 2 heat affected zone and melt zone, and grain size in the heat
Chemical composition (% by weight) of AISI 316L stainless steels (supplied by affected zone (HAZ), parameters and technological aspects that
the manufacturer Arcelor Mittal). qualifies the weldment procedures and quadrate the safety of the
AISI C Mn Si P S structures.
316L 0.0126 1.37 0.443 0.0343 0.0016 In addition to these parameters, were maintained constant the
Cr Ni Mo Cu N following conditions established initially using a synergic and
16.67 9.91 1.96 0.0931 0.0467
controlled equipment:

Table 3 • Stick-out: has been manually adjusted, having a fixed value

Range of chemical composition of the addition wire ER316L according to ESAB of 17 mm.
(Elektriska Svetsnings Aktie Bolaget) (% by weight). • Gas flow: Adjusted through the cylinder valve, having a fixed
C Cr Ni Mo Mn Si P and S value of 17 l/min.
• Welding speed: it is one of the main parameters, because it
<0.03 18.0–20.0 11.0–14.0 2.0–3.0 1.0–2.5 0.25–0.6 ≤0.3
directly influences the value of the heat input. The welding
 R. Demarque et al. / Science and Technology of Materials 30 (2018) 51–59 53

Table 5 microstructural and mechanical properties changes. In this way,

Welding parameters. the thermal cycles obtained by means of this model is later com-
Sample Plate Chain (A) Voltage (V) Speed (mm/s) Heat input bands (kJ/mm) pared to results of the microstructures obtained experimentally,
1 0516 239 26.6 6.69 0.7
analyzing the theoretical effects of cooling rates and dynamic
2 1516 239 26.6 2.30 1.7 developments during the experiment. The software used for
3 2516 239 26.6 1.52 2.7 modeling was developed by the authors is termed Welder-3D.
The software is based on the numerical solution of the heat equa-
tion coupled with dynamic development of local thermophysical
speed was adjusted by the speed of the car by adjusting the properties [4]. The model equations were programmed in Visual
frequency of the car’s motor in the frequency inverter. FORTRAN 95 running on windows platform.
In the computational modeling process, heat transfer phe-
The following parameters are considered fixed for all weld- nomena (radiation, convection and conduction) coupled to mass
ment procedures: Stick-out equal to 17 mm, gas flow equal to transfer phenomena (melting and solidification) are considered.
17 l/min, protection gas composed of 98% argon and 2% oxygen In order to simplify and make feasible the elaboration of the
and addition wire ER316L with a diameter of 1.2 mm. Table 5 model for computational simulation, some considerations were
presents an overview of the other welding parameters used in the made [4–9]:
welding of the three plates, referring to AISI 316L. The nomen-
clature used is according to the range of the heat input practiced, • the loss of heat on the surface of the plate occurs by convection
the type of material. In this way, the first two numbers indicate and radiation;
the range of the heat input practiced, and the second two num- • torch speed is constant’
bers define the type of material. Thus, for the plate named 2516 • deposition of only one weld bead;
the range of the heat input is between 2.4 and 2.7 kJ/mm. • welding electrode running through the plate;
• the distance between electrode and plate is constant.
2.4. Microstructural characterization
This model is based on the resolution of the differential equa-
After cutting and preparation of the sample as received,
tion of energy conservation (Eq. [1]), the temperature (T) being
for comparison purposes it was polished using silicon carbide
the dependent variable, accordingly:
grades of the following granulometries: 100, 220, 320, 400, 600,
800, 1200, 2500 and 4000 mesh. Subsequently, the mechanical ∂       
polishing of the samples was carried out using as an abrasive ρcp T + div ρcp (
u) T = div k (grad (T )) + S (1)
∂t
diamond paste of 1 ␮m granulometry and then the sample was
attacked by immersion with the reagent Beraha, with composi- where:
tion according to Table 6.
After the welding process, a sample of each plate was taken ρ: density of steel (kg/m3 );
for microstructural characterization (phase analysis, measure- cp : specific heat (J/(kg K));
ment of austenitic grain size, quantification of ␦ ferrite in the k: thermal conductivity (W/(m K));
heat affected zone and in the melt zone), quantification using  : vector speed (m/s);
u
Image Pro Plus 4.0 software, dimensioning of the molten zone T: temperature (K).
and Vickers Microhardness measurement in the different zones
(heat affected zone, melt zone and base metal). The variable, S is the heat source, which corresponds to all
input sources or heat loss in the process (energy/heat added
2.5. Numerical methodology in welding, phase transformation, melting and solidification),
which in this work will be given by the double-ellipsoid model
The computational modeling applied in this evaluation aims proposed by Goldak et al. [10].
to simulate the effects of the variation of the welding Heat input The Goldak model (Eqs. (2) and (3)) is also known as double
on the thermal cycle of the welded material (AISI 316L). The ellipsoid, since the heat distribution is modeled (in the volume)
model is capable of predicting the transient temperature distri- as two ellipses, one ahead of the torch heat source and the other
bution during welding. Thus, it predicts temperature and times behind, as shown in Fig. 2. To model the welding process and to
that the material is exposed to thermal cycles, which resulted in predict the temperature in the plate, the same parameters of the
experimental process were considered. In addition, some addi-
Table 6 tional considerations were made in order to make feasible the
Solution used to attack materials as received, Beraha. computational simulation: loss of heat on the surface of the plate
AISI 316L-reagent by convection and radiation, no diffusion of chemical elements
between the added metal and the base metal and chemical com-
Distilled water 80 ml
Hydrochloric acid 37% 25 ml
position of the metal. The calculation domain is assumed the
Potassiun metabissulfite 0.5 g plate and moving torch with heat input given by the measured
current and voltage.
54 R. Demarque et al. / Science and Technology of Materials 30 (2018) 51–59

Fig. 2. Double ellipsoid heat source model.

Adapted from [6]. Fig. 3. Microstructure of AISI 316L austenitic stainless steel. Matrix composed
of polygonal grains of austenite, presence of maclas and ␦ ferrite bands. Attack
In this work by Goldak et al. [10], the heat of the welding arc Beraha, increase 200×.
in motion is applied as a volumetric heat source with a double
ellipsoid distribution following Eqs. (2) and (3): The front half in time and space and through algebraic arrangements, the equa-
of the heat source is defined by Eq. [2] [11]: tion of energy conservation is obtained in the discretized form
√ (Eq. [5]).
6 3ff Q −3(x/a)2 −3(y/bf )2 −3(z/c)2
qf (x, y, z) = √ e e e (2) ap Tp = aE TE + aw Tw + aN TN + as Ts + aT TT + aB TB + b(5)
abf cπ π
The back half to the heat source is defined by Eq. [3]: The terms of Eq. [5] represent the flux of energy thought the
√ discretized control volume faces. The coefficients of Eq. [5] can
6 3fr Q −3(x/a)2 −3(y/br )2 −3(z/c)2 be obtained using the power law scheme, found in the literature
qr (x, y, z) = √ e e e (3)
abr cπ π [12]. The solution of the discretized equation is then obtained
using the line-by-line method described by Patankar combined
At where:
with the tri-diagonal matrix solution algorithm. This sequence
of operations were implemented and used previously [4–9].
• ff and fr : represent respectively the fractions of heat deposited
in the front and rear halves of the heat source;
3. Results and discussion
• Q: represents the magnitude of the welding energy per unit
time (Q = ηVI) where η is the process efficiency, V is the
3.1. Material microstructure in the received state
voltage in volts and I is the current in Ampere;
• v is the welding speed (m/s);
Fig. 3 shows the micrograph of the base metal (BM) of this
• t is the welding time (s);
study, AISI 316L steel. The microstructure of this steel consists
• af , ar , b and c: are parameters that are related to the charac-
of recrystallized austenite grains, with the presence of annealing
teristics of the welding arc, defining the size and shape of the
blades that are characteristic of this material. However, hetero-
ellipse and, consequently, the volumetric distribution of heat
geneity of the matrix was observed, evidenced by the different
source.
grain sizes, which affect the thermophysical properties [8]. In
addition, the presence of small bands of ␦ ferrite [13] can be
In the welding process, the loss of heat on the surface of the
observed.
plate occurs by convection and radiation, which considers the
The average value of the Vickers microhardness of AISI 316L
two effects, follows Eq. [4]:
  steel in the state as received with a 95% confidence interval is
q = qc + qr = h (T − T0 ) + εσ T 4 + T04 (4) 170 ± 3.0 HV.

where: 3.2. HAZ microstructure of welded joints

T0 (environment temperature) = 25 ◦ C; In this topic we will discuss the results concerning the
ε (emissivity) = 0.682; microstructure, austenitic grain size, ␦ ferrite volumetric frac-
σ (constant of Stefan–Boltzman) = 5.67 × 10−8 W/(m2 K4 ); tion and Vickers microhardness exhibited by welded joints of
h (convection coefficient) = 15 W/(m2 K). AISI 316L steel.
Fig. 4 shows the binary phase transformation diagram of the
The initial condition adopted assumes that the temperature 316L steel, the straight line segment AB highlighted in red. In
of the domain is same as the environment equal to 25 ◦ C. this region during the cooling process of the steel occurs the
The solution to the differential equation of energy conser- formation of ferrite ␦, because it is a diffusional reaction the for-
vation (Eq. [2]) can be obtained by applying the finite volume mation of the ferrite depends on the time, so the longer the alloy
technique to a generalized coordinate system. After integration remains in this field, the greater the fraction of ferrite ␦ formed.
 R. Demarque et al. / Science and Technology of Materials 30 (2018) 51–59 55

Fig. 4. Pseudo-binary diagram of the Fe [Cr–Ni]eq system presenting the green

line referring to the value of [Cr/Ni]eq for AISI 316L [9].

Fig. 7. Photograph taken with the thermal camera at 1 m of the joint welded at
According to Ronqueti [14], the heat affected zones, HAZ the end of the process with the contribution of 0.5 kJ/mm.
of AISI 316L austenitic stainless steels contains amounts of
ferrite ␦ and austenitic grains [14]. In addition, the transition
zone (ZT), between the molten zone (MZ) and HAZ, is mainly
differentiated by the ␦ ferrite morphology [14,15].
The curves represent the simulated thermal cycle in the region
indicated by the number at each curve. Figs. 5–7 show respec-
tively the microstructure of the BM/HAZ/MZ interfaces, the
thermal cycle calculated by the numerical model, and the ther-
mal image obtained after the welding process, for the welded
joint with heat input of 0.5 kJ/mm.
Figs. 8–10 show respectively the microstructure of the
interfaces BM/HAZ/MZ, The thermal cycle calculated by the
numerical model, and the thermal image obtained shortly after

Fig. 8. AISI 316L heat affected zone (HAZ), melt zone (MZ) and metal of
base (MB) for heat input of 1.5 kJ/mm. Image obtained with 200× in optical
microscope (OM), chemical attack of Beraha reagent.

Fig. 5. AISI 316L HAZ, heat input of 0.5 kJ/mm. Image obtained with increase
of 200× in optical microscope, chemical attack performed with reagent Beraha.

Fig. 9. Thermal cycles calculated by the numerical model for the AISI 316L
input of 1.5 kJ/mm.

the welding process, for the welded joint with heat input of
1.5 kJ/mm.
Figs. 10–12 show respectively the microstructure of the
BM/HAZ/MZ interfaces, the thermal cycle calculated by the
numerical model, and the thermal image obtained shortly after
Fig. 6. Thermal cycles calculated by the numerical model for the AISI 316L
the welding process, for the welded joint with heat input of
input of 0.5 kJ/mm. 2.5 kJ/mm (Fig. 13).
56 R. Demarque et al. / Science and Technology of Materials 30 (2018) 51–59

Fig. 10. Photograph taken with the thermal camera at 1 m of the joint welded at Fig. 13. Photograph taken with the thermal camera at 1 m of the joint welded at
the end of the process with the contribution of 1.5 kJ/mm. the end of the process with the contribution of 2.5 kJ/mm.

Fig. 14. Plot of grain size versus heat input in the HAZ region.

Table 8
␦ ferrite quantification in the HAZ of the welded joints of the AISI 316L with
Fig. 11. AISI 316L of HAZ, MZ and MB for heat input of 2.5 kJ/mm. Image different bands of heat input and metal transfer by pulsed spray (P).
obtained with increase of 200× in optical microscope, chemical attack per- Samples Volumetric fraction of ferrite ␦
formed with reagent Beraha.
0516 10.8
1516 11.5
2516 12.2

3.3. Size of austenitic grain in HAZ

Table 7 shows the values of the austenitic grain size that

the HAZ of the samples of the welded joints of the AISI
316L presented. When comparing the austenitic grain size
presented by the BM (22.0 ± 4.1 ␮m) and the austenitic grain
size presented in the HAZ, it is possible to observe that all the
samples evidenced the austenitic grain growth in the HAZ, as
Silva also stated in 2009 [16].
Fig. 12. Thermal cycles calculated by the numerical model for the AISI 316L Fig. 14 shows a plot with average grain size in HAZ as a
input of 2.5 kJ/mm. function of the heat input.
As expected due to the thermal cycle suffered by the region
Table 7 the austenitic grains suffered an increase when compared to the
Austenitic grain size within HAZ of AISI 316L.
material as received.
Samples Grain size (␮m)

0516 32 3.4. Quantification of delta ferrite in HAZ

1516 48
2516 63 Table 8 shows a quantitative analysis of the presence of delta
ferrite in the HAZ region, so that the connection between the
 R. Demarque et al. / Science and Technology of Materials 30 (2018) 51–59 57

Fig. 15. Plot of ferrite fraction ␦ versus heat input in the HAZ.
Fig. 18. AISI 316L MZ, heat input of 2.5 kJ/mm. Image obtained with increase
of 200× in optical microscope, chemical attack performed with reagent Beraha
200×.

Fig. 16. AISI 316L MZ, heat input of 0.5 kJ/mm. Image obtained with increase
of 200× in optical microscope, chemical attack performed with reagent Beraha.

Fig. 19. Thermal cycles calculated by the numerical model for the AISI 316L
input of 2.5 kJ/mm.

3.5. Welded joints MZ microstructure

Figs. 16 and 17 shows the microstructure of weld metal

welded with heat input of 0.5 kJ/mm and the calculated ther-
mal cycle that was submitted respectively. The tangent line to
the cooling curve indicates the calculated cooling rate of the
Fig. 17. Thermal cycles calculated by the numerical model for the AISI 316L microstructure after the welding process.
input of 0.5 kJ/mm. Figs. 18 and 19 show the weld metal weld metal microstruc-
ture with heat input of 2.5 kJ/mm and the calculated thermal
cycle, which was subjected respectively. The tangent line to
the cooling curve indicates the calculated cooling rate of the
increase of the thermal contribution and the presence of this microstructure after the welding process.
phase in the HAZ region can be correlated. It can be seen that the line of the cooling rate indicated in
Fig. 15 shows a graph with the behavior of the quantification Fig. 17 shows a steeper slope than the line indicated in Fig. 19.
of the ferrite delta HAZ as a function of the heat input. This is due to the higher heat input in Fig. 18, which provides
As Kumar observed in 2011 [17], an increase in the amount more energy for welded joint resulting in a Temperature, so its
of ␦ ferrite was expected due to the increase of the heat input, cooling rate is lower.
the cooling rate decreased, causing the material to pass more The microstructural characterization and quantification indi-
slowly through the ferrite ␦ formation region, then the volumet- cated that the lower cooling rate resulted in thinner dendrites
ric fraction increased when compared to that of the material as with smaller dendritic spacing (Fig. 16), while the higher cooling
received. rate provided dendrites with larger spacings (Fig. 18).
58 R. Demarque et al. / Science and Technology of Materials 30 (2018) 51–59

Table 9
Average values of Vickers microhardness (HV) of welded joints of AISI 316L
steel.
Samples Welded joint zone Microhardness Vickers (HV)

0516 MZ 191
HAZ 186
BM 175
1516 MZ 178
HAZ 185
BM 174
Fig. 20. Methodology used to size the weld metal in the macrograph of the AISI
2516 MZ 174 316L weld bead.
HAZ 173
BM 169

3.6. Vickers Microhardness

As previously mentioned, the analysis of the Vickers micro-

hardness profile of the welded joint is of fundamental importance
when it is desired to evaluate the characteristics of the welded
region, since this property is directly related to phenomena of
phase and microstructural transformations due to the different
thermal cycles developed. Table 9 shows the mean MZ, HAZ
and BM Vickers Microhardness values of all AISI 316L plates.
Considering that the average Vickers microhardness of the
AISI 316L steel in the state as received is 170 ± 3.0 HV, it can Fig. 21. Area of the weld metal versus heat input.
be seen that the Vickers microhardness values for the differ-
ent zones (MZ and HAZ) and base metal (BM) Did not show
significant variations after the welding process (considering the to increase. This is due to the increased heat supplied to the
Vickers scale ranging from 100HV to approximately 900 HV). weld joint, so the amount of melt material in the joint is higher
Thus, this result can be explained as a result of the measure- resulting in an increase in the area of the weld metal.
ments made in the austenitic matrix, where the thickness of the
␦ ferrite dendrites is very thin, making it difficult to measure 4. Conclusions
the Vickers microhardness only at this stage. A similar result
was obtained by Pessanha [18] in the study of welded austenitic The increase in the heat input provokes an increase in the
stainless steel. dimensions of the fusion area, increase of ␦ ferrite volumetric
fraction and the austenitic grain size in the HAZ. The ␦ ferrite
3.7. Dimensions of the MZ and HAZ morphology revealed in the HAZ was classified as vermicu-
lar and the presence of unwanted precipitates or intermetallic
In order to evaluate the influence of the increase of the heat phases was not observed in the welded joint. The values of
input used in geometry and weld bead dimensions, with the help Vickers microhardness measured in the different zones showed
of the Geogebra and Profileometer software, the regions of the no significant variations. The contribution of 2.5 kJ/mm showed
cross sections of the weld beads were designed for the three heat the highest volumetric fraction of ferrite ␦, which together with
inputs used In welding the AISI 316L, as shown in Fig. 20 and the excessive growth of the austenitic grain size in the HAZ is
Table 10. expected to deteriorates the properties of this region. The heat
Fig. 21 shows a graph with the behavior of the area of the inputs of 0.5 and 1.5 kJ/mm provided the best combined results,
weld metal as a function of the heat input. with lower ␦ ferrite volume fractions in the HAZ, as well as MZ.
As can be seen in Table 10 and in the graph of Fig. 20, when Furthermore, the dimensions and areas of the zones indicated
the heat input is increased the area of the weld metal tends that the welded joints are suitable for these heat inputs.

Table 10
Dimensions of weld seams and HAZ for different inputs.
Heat input (kJ/mm) Sample Width MZ (mm) Height MZ (mm) Dimensions HAZ (mm) Area MZ (mm2 )

0.4–0.7 0516 13.46 6.38 0.05 39

1.3–1.7 1516 19.53 8.9 0.18 91
2.4–2.7 2216 23.86 9.85 0.29 132
 R. Demarque et al. / Science and Technology of Materials 30 (2018) 51–59 59

References [10] J. Goldak, A. Chakravarti, M. Bibby, A new finite element model for
welding heat sources, Metall. Trans. B 15B (1984) 299–305.
[1] A.H. Ramírez, Influência dos íons de brometo e cloreto sobre a resistência [11] J.A. Goldak, M. Akhlaghi, Computational Welding Mechanics, Springer,
à corrosão por pite de diferentes aços inoxidáveis austeníticos e ferríticos, 2005.
Dissertação, Mestrado em Engenharia Metalúrgica e de Materiais – Escola [12] S.V. Patankar, Numerical Heat Transfer and Fluid Flow, McGraw-Hill,
Politécnica da Universidade de São Paulo, SP, 2011. USA, 1979.
[2] P.J. Modenesi, Soldabilidade dos Aços Inoxidáveis, SENAI-SP, São Paulo, [13] P. Tiedra, O. de Martín, Effect of welding on the stress corrosion cracking
2001. behaviour of prior cold worked AISI 316L stainless steel studied by using
[3] R.D. Campbell, Avoiding defects of stainless steels welds, Weld. J. 86 (5) the slow strain rate test, Mater. Des. (2013) 103–109.
(2007) 53–63. [14] L.A. Ronqueti, Efeito do modo de transferência metálica sobre o com-
[4] C.R. Xavier, H.G. Delgado Jr., J.A. Castro, A.F. Ferreira, Numerical pre- portamento térmico e as transformações metalúrgica na soldagem GMAW
dictions for the thermal history microstruture and hardness distributions at dos aços austeníticos AISI 304 e 316 utilizados em projetos de instalações
the HAZ during welding of low alloy steels, Mater. Res. (2016) 520–533. nucleares, Dissertação de Mestrado, Escola Engenharia Indústiral Metalúr-
[5] J.A. Castro, E.M. Oliveira, D.S.S. Almeida, G.S. Fonseca, C.R. Xavier, gica de Volta Redonda, PPGEM, UFF, 2014, pp. 1–163.
Effects of local heat input conditions on the thermophysical properties of [15] J. Yan, M. Gao, X. Zeng, Study on microstructure and mechanical pro-
super duplex stainless steels (SDSS), Mater. Res. (2018) 1–9. prieties of 304 stainless steel joints by TIG, laser and laser-TIG hybrid
[6] R. Demarque, J.A. Castro, C.R. Xavier, D.S.S. Almeida, C.J. Marcelo, E.P. welding, Opt. Lasers Eng. (2010) 512–517.
Santos, et al., Estudo Numérico e Experimental da Evolução Microestru- [16] C.C. Silva, J.P. Farias, H.B. Sant’ana, Evaluation of AISI 316L stainless
tural e das Propriedades de Juntas Soldadas de Vergalhões pelo Processo steel welded plates in heavy petroleum environment, Mater. Des. (2009)
GMAW, Soldagem Inspeção 20 (2015) 434–445. 1581–1587.
[7] G.S. Fonseca, L.O.R. Barbosa, E.A. Ferreira, C.R. Xavier, J.A. Castro, [17] S. Kumar, A.S. Shahi, Effect of heat input on the microstructure and
Microstructural, mechanical, and electrochemical analysis of duplex and mechanical properties of gas tungsten arc welded AISI 304 stainless steel
superduplex stainless steels welded with the autogenous TIG process using joints, Mater. Des. (2011) 3617–3623.
different heat input, Metals 7 (2017) 538–559. [18] C.E. de Pessanha, Quantificação da ferrita delta e avaliação da relação
[8] C.R. Xavier, M.F. Campos, J.A. Castro, Numerical method applied to microestrutura/propriedades de um açoinoxidável austenítico 347 soldado,
duplex stainless steel, Ironmak. Steelmak. 40 (2013) 420–429. Dissertação (Mestrado em Engenharia e Ciências dos materiais), Universi-
[9] C.R. Xavier, H.G. Delgado Jr., J.A. Castro, An experimental and numerical dade Estadual do Norte Fluminense/UENF, Campos dos Goytacazes, RJ,
approach for the welding effects on the duplex stainless steel microstruc- 2011.
ture, Mater. Res. 18 (2015) 489–502.