Академический Документы
Профессиональный Документы
Культура Документы
Article views: 12
Introduction
Arc welding processes such as gas metal arc welding
(GMAW) and ux-cored arc welding (FCAW) are promising in the manufacturing of duplex stainless steel (DSS)based pressure vessels, pipelines and industrial equipment
in general.1,2 The FCAW process generates a slag due to
the presence of the ux and, as a result, gaseous species
released when ux melting generates a protective atmosphere, eliminating partially or entirely the need for a protection gas to be used. The process is automated which is
an attractive feature for practical engineering applications. However, the weld bead needs to be cleaned at
each welding pass to remove the slag, thus reducing the
process productivity. If the cleaning procedure is not properly carried out, then non-metallic inclusions will contaminate the weld bead. In this case, it has to be
partially removed and repaired. According to Tuomi
et al.,1 approximately 2% of the welding procedures of
DSS are reproved and need to be repaired. Some regions
1
Department of Materials Engineering, Universidade Estadual de Campinas (UNICAMP), School of Mechanical Engineering, 13083-860 Campinas,
SP, Brazil
2
Universidade Federal do ABC (UFABC), Centro de Engenharia, Modelagem e Cincias Sociais Aplicadas (CECS), 09210-580 Santo Andr, SP,
Brazil
of the heat-affected zone (HAZ) are not removed to execute the repair cycle. Hence, the repaired zones suffer
repeated thermal cycles and this can markedly affect the
corrosion behaviour of the welded joint.3
Welding cycles are likely to produce microstructural
transformations of DSS in the HAZ and fusion zone
(FZ), leading to an unbalanced fraction of the austenite
and ferrite phases.4 In this respect, the corrosion resistance of welded joints can be seriously affected, especially
the pitting corrosion behaviour of the welded DSS.5 Furthermore, partitioning of alloying elements such as chromium, molybdenum and nickel can take place due to
different thermal cycles, giving rise to distinct localised
corrosion behaviours in the weld bead and its surroundings.6 In the same way, the precipitation of undesirable
phases such as chromium nitrides and carbides in the
HAZ or FZ can deteriorate the corrosion properties
too.7 Naghizadeh and Moayed8 showed that secondaryphase precipitation decreases the corrosion resistance of
DSS due to chromium depletion after welding cycles
and this behaviour can be monitored by evaluating the
critical pitting temperature (CPT) of the welded joint.
Garca-Rentera et al.9 showed that resistance to localised
corrosion of DSS joints made by GMAW was impaired
by the presence of detrimental phases such as chromium
carbides. Ferrite-stabilising elements such as molybdenum can give rise to the formation of -phase which,
2016
Experimental
Material and welding procedure
The base metal investigated in this work was a hot rolled
and annealed DSS 2304 (UNS 32304 ArcelorMittal)
sheet with 10-mm thickness. The ller metal was a DSS
2209 ux-cored tubular wire of 1.2-mm diameter (KESTRA 4462 GFP). Compositions of base and ller metals
are shown in Table 1.
The 2304 sheet was cut into four rectangular pieces
with dimensions of 200 100 mm, which were chamfered
in a universal milling machine and welded, in pairs, using
the FCAW process for both root and lling passes. Butt
weld was carried out using reverse polarity in the at
1G position. Even though the weld root was backed
using a ceramic tile, root repair welding was necessary
in the backside of the welded joint. The dimensions of
the welded specimens prior to welding are indicated in
Fig. 1. The protection gas used was a 75%Ar-25%CO2
mixture at a ow of 1619 L min1. Additionally, in the
root pass, 99.9%Ar was used in the backside of the joint
at a ow of 20 L min1. The detailed welding sequence
BM
FM
C (%)
Cr
Cu
Mn
Mo
Ni
Si
0.018
0.028
22.550
22.540
0.424
0.343
1.350
1.370
0.259
3.040
0.001
0.080
3.570
10.000
0.025
0.030
0.001
0.003
0.360
0.540
2016
Electrochemical measurements
is presented in Fig. 2 and the welding parameters are presented in Table 2. For all welding passes, the temperature
in-between was kept below 150C and controlled by
means of type-K contact thermocouple placed on the surface of the weld bead.
The welded sheet was divided into three specimens of
60 mm. The cutting line was perpendicular to the welding
line. The rst and last 10 mm of the weld bead were separated and disposed. One piece was not repaired (R0).
One piece was repaired once (R1) and the last piece was
repaired twice (R2). The repair cycles were carried out
always on the same side of the plate which was backgauged to sound metal using an abrasive disk before the
new welding take place. Each repair cycle consisted of
two passes (passes 5 and 6 in Fig. 2). The specimens are
designated as R0, R1 and R2, where the digit indicates
the number of repair cycles carried out for each specimen.
The electrochemical measurements were conducted in
triplicate.
Table 2
R1
R2
Microstructure analysis
Quantication of relative fractions of austenite and ferrite phases was performed by OM. Specimens were electrolytically etched in a 30 wt-% KOH water-based
Specimen
R0
Step
1 (root)
2 (lling)
3.1 (nishing)
3.2 (nishing)
4 (root repair)
5 (lling repair)
6 (nishing repair)
5 (lling repair)
6 (nishing repair)
5 (lling repair)
6 (nishing repair)
Current/A
Voltage/V
Speed/mm s1
125
128
128
128
128
174
174
174
174
174
174
24
24
24
24
24
23
23
23
23
23
23
3.19
2.17
3.37
3.14
3.50
3.50
3.04
3.68
3.33
3.68
3.33
940
1418
911
979
878
1143
1315
1086
1201
1086
1201
2016
Table 3
Material
2304
R0
R1
R2
Icorr/108 A cm
3.6 1.2
37.4 26.6
3.7 2.0
11.4 7.5
4 Optical microscopy of the 2304 steel a showing a good balance the ferrite (dark) and austenite (bright) phases and SEM-back
scattered electrons image b of a non-metallic inclusion in the weld bead of one R0 specimen
2016
5 Optical microscopy of R0 specimens showing deleterious precipitates revealed by the oxalic acid etching
2016
7 Optical microscopy showing the presence of precipitates in the HAZ of an R2 specimen. The image on the right shows the
HAZ at a higher magnication
Table 4
Sample
R1
R2
Cr/wt-%
Ni/wt-%
Mo/wt-%
PREN
Ferrite
Austenite
Ferrite
Austenite
22.3 0.5
20.5 0.5
21.7 0.7
20.3 0.6
5.0 0.3
5.8 0.2
4.9 0.2
5.6 0.3
0.34 0.2
0.27 0.2
0.33 0.2
0.28 0.2
24.4
27.4
23.8
27.3
2016
a pit nucleation site, as shown in Fig. 9. R0 and R2 specimens, in turn, presented heterogeneous microstructure
characterised by the formation of widespread undesirable
precipitates detected by the oxalic acid test. Hence, the
corrosion resistance was depressed for these conditions.
Conclusions
The inuence of repair welding cycles on the corrosion
behaviour of the 2304 DSS has been evaluated. Besides
the FCAW process leading to the presence of non-metallic
inclusions in the weld bead if it is not properly cleaned
after welding, the results showed that it is possible to
obtain a weld bead with a homogeneous microstructure
after the repair cycle (R1 specimens), but it seems there
are a limited number of welding-repair operations that
result in an optimised microstructure.
The CPT was sensitive to the microstructural changes
during the welding process and can be used as a design
criterion to assess the quality of repaired welding in
DSS. A good agreement was found between the electrochemical behaviours observed in the potentiodynamic
polarisation curves and the CPT values.
Acknowledgements
The authors thankfully acknowledge the Brazilian
research funding agencies CAPES and CNPq for their
nancial support, KESTRA for providing ux cored
welding wire for this work, Prof. Rubens Caram (UNICAMP) for the SEM equipment and Montcalm Montagens, Brazil, for the welding equipment supplied and staff.
ORCID
Renato Altobelli Antunes
1540-6495
http://orcid.org/0000-0003-
References
1. Tuomi, A. Lfstrand and M. Harju: Increased usage of duplex
materials in manufacturing of pulping equipment, Duplex
America Conference 2000, Stainl. Steel World, Houston, TX, 29
February1 March 2000, 401408.
2016
2016
17.
18.
19.
20.
21.
22.
23.
24.
25.
26.
27.
28.
29.
30.
31.