Acta Metall. Sin. (Engl. Lett.

), 2014, 27(6), 1018–1030

DOI 10.1007/s40195-014-0116-5

Characterization of Microstructure, Strength, and Toughness

of Dissimilar Weldments of Inconel 625 and Duplex Stainless Steel
SAF 2205
R. Sridhar • K. Devendranath Ramkumar • N. Arivazhagan

Received: 25 January 2014 / Revised: 21 February 2014 / Published online: 3 August 2014
Ó The Chinese Society for Metals and Springer-Verlag Berlin Heidelberg 2014

Abstract The dissimilar combinations of Inconel 625 and duplex stainless steel SAF 2205 obtained from manual GTA
welding process employing ER2209 and ERNiCrMo-3 filler metals have been investigated. Formation of secondary phases
at the HAZ of Inconel 625 and grain coarsening at the HAZ of SAF 2205 were witnessed while using these filler wires. The
average hardness of ER2209 weldments was found to be greater than ERNiCrMo-3 weld. Tensile fracture was observed at
the weld zones for both the fillers. Impact test trials showed brittle mode of fracture on employing ER2209 filler and mixed
(ductile–brittle) mode of fracture while using ERNiCrMo-3 filler. Further optical microscopy and SEM/EDS analysis were
carried out across the weldments to investigate the structure–property relationships.

KEY WORDS: Inconel 625; Duplex stainless steel; SAF 2205; Dissimilar metal welding; Microstructure;
Mechanical characterization

1 Introduction bimetallic joints on Inconel and stainless steel series had

been reported also [8–10].
Nickel-based superalloys and stainless steels are widely One of the key issues in the welding of Ni-based
used in high temperature environments such as steam superalloy with stainless steels is the selection of appro-
generators, reformer, and pyrolysis tubes in oil refineries priate filler metal. Naffakh et al. [8] employed gas tungsten
[1–3]. In addition, these dissimilar combinations are used arc welding (GTAW) for joining Inconel 657 and AISI 310
in petrochemical factories and nuclear power plants, where using Inconel 82, Inconel A, Inconel 617, and 310 au-
a combination of the strength and resistance to corrosion is stenitic stainless steel filler wires. It was reported that the
required [4–6]. van Wijngaarden and Chater [7] reported weld microstructure was found to have fully austenitic
the use of these joints in the geothermal power plants structure for all the filler wires. Similar observations were
where Inconel 625 operates in high temperature and made by Shah Hosseini et al. [9] who carried out the
Duplex 2205 in sub-zero temperature. The applications of investigations on the mechanical properties of Inconel 617
and AISI 310 stainless steel.
Lee et al. [11] reported that the increase of Nb in filler
metal composition led to the formation of equiaxed den-
Available online at http://link.springer.com/journal/40195
drites. The authors also reported that Nb addition increases
R. Sridhar
K. D. Ramkumar (&)
N. Arivazhagan the yield strength and reduces the elongation with no
School of Mechanical & Building Sciences, VIT University, apparent influence in tensile strength while welding nickel-
Vellore 632014, India base alloy 690 to SUS 304L. Welding of super alloy with
e-mail: deva@vit.ac.in
stainless steel has a number of challenges such as ductility-
R. Sridhar dip cracking (DDC), hydrogen cracking, and formation of
Saraswathy Velu College of Engineering, Vellore, India brittle phases in the weld and weld interface which could

123
R. Sridhar et al.: Acta Metall. Sin. (Engl. Lett.), 2014, 27(6), 1018–1030 1019

cause inferior mechanical and metallurgical properties. 2 Experimental

Ductility-dip cracking susceptibility was observed to be
one of the major problems during the multi-pass welding of The nominal chemical compositions of the as-received base
Inconel 625 [12]. Belloni et al. [13] reported that the use of metals and filler metals have been studied using wet spec-
filler metal employing higher Nb constituent enhances the troscopic method and are presented in Table 1. The base
mechanical properties but however the segregation of Nb metals with the dimension of 125 mm 9 50 mm 9 5 mm
forms the Nb-rich phase, which is a brittle compound that were employed in this study. Two filler metals such as
has detrimental effect on weldability and weld mechanical ER2209 and ERNiCrMo-3 have been chosen as these filler
properties such as ductility, fracture toughness, fatigue, and metals are compatible with the base metals employed. The
creep rupture. process parameters were ascertained based on the previous
Further it was reported that the presence of 3 wt% Nb studies and also by iterative studies based on the bead on
stabilizes the austenitic matrix in the weld zone [9]. Also plate welding and are represented in Table 2.
the presence of iron in nickel-based superalloys lowers the Standard V-groove butt joints (single V-groove having a
niobium solubility in austenite phase and the presence of root gap of 2 mm, size land of 1 mm, and included angle of
Niobium not only lowers the melting point constitutionally, 70°) were employed for welding these dissimilar metals. A
but also forms low-melting carbide–austenite eutectics specially designed fixture with a copper back plate that
during solidification [8]. could clamp the base metals firmly so as to avoid distortion
Dissimilar weld joints between Inconel 625 and Duplex and bending was employed. The weldments were charac-
stainless steel have particular use in high temperature terized for any defects using gamma ray radiography
applications where the combination of enhanced mechan- technique. Subsequent to the results, the welded samples
ical properties and corrosion resistance is required. As were cut into different coupons for further analysis.
evident from the open literature, very limited studies have The macro and microstructure studies were performed
been reported on these combinations of joints. Continuous on the coupons termed as ‘‘composite regions’’ which
current gas tungsten arc welding technique is employed covers all the zones [parent metals, heat affected zones
using two different filler metals such as ER2209 and ER- (HAZs), and weld zone]. Standard metallographic proce-
NiCrMo-3 to investigate and compare the penetration, dures were adopted for examining the microstructure of the
grain refinement, and segregation effects which influence weldments. Electrolytic etching (10% oxalic acid solution,
the weld quality. Further, the studies reported in the article 6 V DC supply of 1 A/cm2) was employed to examine the
addressed the effect of filler wire on metallurgical and microstructure of Inconel 625 and SAF 2205. Further, the
mechanical properties of these dissimilar weldments. SEM/EDS analysis was performed on the various zones of

Table 1 Chemical compositions of the base/filler metals (in wt%)

Materials C Mn Cr Mo Fe Nb Ni N Others

Inconel 625 0.022 0.169 22.80 9.33 4.24 2.98 59.65 – Si 0.104; Cu 0.157; P 0.004;
S 0.010; Al 0.101; Co \ 0.005; Ti 0.235; Ta \ 0.02
UNS 2205 0.028 1.32 23.16 3.19 Bal. – 5.20 0.149 Si 0.372; P 0.026; S 0.005
ER2209 0.024 1.6 22.19 3.1 Bal. – 9.0 0.16 P 0.016; S 0.005
ERNiCrMo-3 0.05 0.18 22.4 9.2 0.8 3.8 Bal – Ti 0.22; Cu 0.25; S 0.004;
P 0.024; Si 0.20

Table 2 Process parameters employed in the GTA welding using different filler metals
Filler Welding pass Voltage (V) Current (A) Shielding gas flow Filler wire
rate (L/min) dia. (mm)

ER2209 Root pass 10.9–11.1 151–152 15 2.4

Remaining passes 16.1–16.4 170–172 15
ERNiCrMo-3 Root pass 11.1–11.4 151–152 15 2.4
Remaining passes 12.4–14.6 152–170 15

123
1020 R. Sridhar et al.: Acta Metall. Sin. (Engl. Lett.), 2014, 27(6), 1018–1030

Fig. 1 Photographs of the GTA weldments employing ER2209 a, ERNiCrMo-3 b fillers

the weldments to determine the presence of various ele-

ments and also helpful to assess the structure–property
correlations.
Further to estimate the mechanical properties of dissimilar
joints of Inconel 625 and Duplex stainless steel, various tests
were performed. Three samples were tested for tensile and
impact test for determining the repeatability of the results.
Tensile studies were carried out on the weldments which were
fabricated as per ASTM E8/8M standards. These samples
were tested at a strain rate of 2 mm/min at room temperature.
Charpy V-notch impact test samples were prepared and tested
according to ASTM E23-12C standards of sub-sized speci-
mens by simple beam method. Notches were made such that
fracture occurred only within the weld fusion zones. Fur-
thermore the fractured samples were characterized for SEM
analysis to determine the mode of fracture. Hardness mea-
surement was carried out on the composite region of the
weldments across three distinct zones vis-à-vis cap, filler pass,
and root of the weldments using Vicker’s micro-hardness Fig. 2 Macrostructure of the dissimilar weldments of Inconel 625
tester with a load of 4.9 N for a dwell time period of 10 s at and SAF 2205 Duplex stainless steel employing ER2209 a, ERNi-
regular intervals of 0.25 mm. The following chapter addres- CrMo-3 b fillers
ses the results and discussion of the experimental work.
fusion had occurred with the base metals on using GTA
welding process, as shown in Fig. 2. No physical surface
3 Results and Discussion defects like arc strike, cracks, and undercut have been
observed in any of these weldments. It is witnessed that the
3.1 Macrostructure selection of filler wire and process parameters are appropriate
for both cases. Moreover, proper weld pool morphology
The photographs of the weldments shown in Fig. 1 clearly without any macro defects obtained in the macro graph con-
indicated that the joining of dissimilar metals involving firmed for the proper weld fusion. Also gamma ray non-
Inconel 625 and SAF 2205 could be possible on employing destructive testing (NDT) inspection clearly corroborated that
these filler metals and the welding technique. It is understood the weldments were free from subsurface weld defects such as
from the cross-sectional macrostructure studies that apposite inclusions, porosity, and lack of fusion etc.

123
R. Sridhar et al.: Acta Metall. Sin. (Engl. Lett.), 2014, 27(6), 1018–1030 1021

Fig. 3 Microstructure showing dissimilar Inconel 625 and SAF 2205 weldments employing ER2209 a, b; ERNiCrMo-3 c, d fillers

3.2 Microstructure weld, a remarkable layer of precipitation zone was noticed

at the fusion zone and in Inconel 625; however, the width
The interface microstructure between filler ER2209 weld of the precipitation zone is meager in comparison with
metal and Inconel 625 clearly witnessed the considerable ER2209 welds which could be probably due to the proper
precipitation/un-mixed zone which is represented in Fig. 3a dilution of the filler metal with the base metal Inconel 625.
and b. This could be acquainted to the differences in the Further, the reason could be attributed to the similarities
melting temperature of filler and base metal which causes existing in the melting temperature and chemical compo-
poor dilution in the re-solidification stage. Interestingly, well sition of ERNiCrMo-3 and Inconel 625.
defined fusion line is noticed at the interface of ER2209 and In addition, the columnar growth occurs at the weld zone
SAF 2205 due to compatibility of filler and base metal. It is adjacent to the Inconel 625 side. On the other hand, combi-
also observed that more grain growth has occurred in the SAF nation of equiaxed and cellular dendrite was noticed in the
2205 HAZ due to the temperature increase during various weld center. Figure 3d represents the interface of ERNiCrMo-
welding passes. On the other hand, grain boundary thickening 3 weld metal, and SAF 2205 inferred the slight thickening of
with epitaxial growth could be seen in the HAZ of the Inconel fusion line which could be attributed to less compatibility of
625. This may be due to formation of low melting secondary the base metal with ERNiCrMo-3 filler wire compared to
phases at these boundaries. The weld microstructure adjacent ER2209 filler. The grain growth with fine precipitates along
to Inconel 625 side consists of austenitic matrix with some the solidification grain and sub-grain boundaries was observed
inter- and intra-granular precipitates in the equiaxed long in the HAZ of SAF 2205 side. However, this phenomenon is
dendritic arms. In contrast, the cellular grain growths were found to be absent on using ER2209 filler wire.
observed near the SAF 2205 side on employing ER2209 filler. Formation of secondary phases or the segregation
The interfacial micrograph of the filler ERNiCrMo-3 effects at the HAZ of Inconel 625 and grain coarsening at
weld metal and the parent metals of Inconel 625 and SAF the HAZ of SAF 2205 for both the filler metals is inferred
2205 is represented in Fig. 3c and d. Similar to the ER2209 from the micrographs (Fig. 3a–d). This could be attributed

123
1022 R. Sridhar et al.: Acta Metall. Sin. (Engl. Lett.), 2014, 27(6), 1018–1030

Fig. 4 Weld microstructures of the dissimilar Inconel 625 and SAF 2205 weldments employing ER2209 a, ERNiCrMo-3 b fillers; SEM
photographs at the weld zone ER2209 c, ERNiCrMo-3 d fillers

to the higher heat inputs developed during GTA welding results showed that there is enrichment of Mo and Fe in the
resulting in slow cooling rate. Columnar dendrites with weld interface adjacent to Inconel 625. This is attributed to
inter-dendritic arms were observed at the weld zone migration of Fe from SAF 2205 and Mo from Inconel 625
employing ER2209 filler. Fine dispersed secondary phases during the welding process, which is caused due to the
were also witnessed in between the inter-dendritic regions differences in elemental concentration gradients of base
across the entire zone, as shown in Fig. 4a and c. The and filler metal. The presence of secondary phases was
dendrite network in dark clearly appears in back-scattered observed, and these phases could be enriched with Fe–Ni–
electrons due to the segregation of heavy elements Nb and Cr as evident from the EDS results.
Mo in the inter-dendritic spaces which appear in a lighter Figure 5b represents the SEM and corresponding line
tone, as shown in Fig. 4b and d. According to the existing mapping results at the interface of ER2209 weld and SAF
literatures, the elements Mo, Si and, more particularly, Nb 2205. Similar to earlier discussion, there is an enrichment
are rejected in the inter-dendritic liquid during the solidi- of Ni, Mo and Fe in the weld zone. However the segre-
fication process, Ni and Fe being concentrated in the gation of Mo in the boundaries of the dendrite is highly
dendrite cores and are evident from the EDS analysis predominated as compared to other interface. Further, the
reported in the subsequent sections. results inferred the depletion of Mo and Ni in the weld
interface and this is evident from the migration of these
3.3 SEM and EDS Analysis elements as discussed in the earlier section. Figure 4d
shows the presence of austenitic matrix containing chains
3.3.1 SEM and EDAX Analysis on ER2209 Weldments of secondary phases in a skeleton structure. Different ori-
entation and colors of these phases could be ascribed to the
SEM and EDAX line mapping results of GTA ER2209 difference in solidification rate due to atomic weight of the
weld interface of Inconel 625 are represented in Fig. 5a. concentrated elements. Further the EDS results also wit-
SEM images witnessed the micrograph of the weldment nessed that there is enrichment of elements like Fe, Cr, Mo
which is discussed in the earlier Sect. 3.2. EDAX analysis and C.

123
R. Sridhar et al.: Acta Metall. Sin. (Engl. Lett.), 2014, 27(6), 1018–1030 1023

Fig. 5 Line mapping analysis on the dissimilar weldments employing ER2209 filler: a interface at Inconel 625 side; b interface at SAF 2205
side

EDS point analysis on ER2209 weld interface of both point analysis that the presence of Cr is 20.89 wt% which
SAF 2205 and Inconel 625 is represented in Fig. 6. The is lower at the voids as compared to dendritic core (Fig. 6a,
presence of voids has been observed at the weld zone b). Segregation has been observed as white precipitates at
adjacent to the interface of SAF 2205 employing ER2209. the weld fusion zone adjacent to the Inconel 625 side
This could be attributed to the depletion of Cr which (Fig. 6c). As evident from the EDS results, these segre-
evaporated during welding. It is also confirmed from the gations are enriched with elements such as Nb (5.81 wt%),

123
1024 R. Sridhar et al.: Acta Metall. Sin. (Engl. Lett.), 2014, 27(6), 1018–1030

Fig. 6 SEM/EDAX point analysis on the weld zone of the dissimilar weldments of Inconel 625 and SAF 2205 employing ER2209 filler: a,
b weld zone at the SAF 2205 side; c, d weld zone at the Inconel 625 side

123
R. Sridhar et al.: Acta Metall. Sin. (Engl. Lett.), 2014, 27(6), 1018–1030 1025

Fig. 7 Line mapping analysis on the dissimilar weldments employing ERNiCrMo-3 filler: a interface at Inconel 625 side; b interface at SAF
2205 side

Ti (0.22 wt%), and Mo (13.91 wt%). As these elements are in weld fusion zones of Ni–Cr–Mo alloys leads to a situ-
present in the inter-dendritic regions of the weld zone, they ation where dendrite cores are impoverished in elements
would be absent or found less in the matrix (Fig. 6d). This like Mo, and inter-dendritic regions are enriched in Mo,
could partially deteriorate the mechanical properties of the Nb, and sometimes Cr [13, 14]. Greater amounts of Nb
weldment. Similar observations were reported elsewhere; distribution occur principally in the inter-dendritic area and
the authors reported that the micro-segregation that occurs take the form of Nb-rich precipitates. These Nb-rich phases

123
1026 R. Sridhar et al.: Acta Metall. Sin. (Engl. Lett.), 2014, 27(6), 1018–1030

Fig. 8 SEM/EDAX point analysis on the weld zone of the dissimilar weldments of Inconel 625 and SAF 2205 employing ERNiCrMo-3 filler:
a weld interface—weld zone adjacent to SAF 2205; b–d weld zone

123
R. Sridhar et al.: Acta Metall. Sin. (Engl. Lett.), 2014, 27(6), 1018–1030 1027

Fig. 9 Hardness profile of the dissimilar weldments of Inconel 625 and SAF 2205 employing ER2209 a, ERNiCrMo-3 b fillers

would be either NbC and/or Laves phase. As reported by 2.98 wt% Nb, and 0.022 wt% C and hence tends to form
Dupont et al. [15, 16], a higher concentration of Nb tends the TiC and NbC during high-temperatures exposure.
to increase the volume of Nb-rich precipitates, while a These results confirmed by EDS point analysis have been
higher C content leads to the formation of NbC. However, discussed in detail in the next section.
the Laves phase forms readily with a high Nb and a low C Figure 8 represents the results of SEM/EDS point ana-
composition, particularly when Si and Fe are also present. lysis on ERNiCrMo-3 weld interface of both base metals.
The white precipitates were observed at the weld zone shown
3.3.2 SEM/EDAX Analysis on ERNiCrMo-3 Weldments in Fig. 8a and b, which were enriched with Nb and Ti in the
Inconel 625 side. This is attributed to the formation of phases
SEM and EDAX line mapping results of GTA-ERNiCrMo- such as NbC, (Nb,Ti) C as reported by Hosseini et al. [9].
3 weld interface of Inconel 625 and SAF 2205 are repre- Some segregation of secondary phases was noticed in the
sented in Fig. 7a and b. The results showed the enrichment SAF 2205 side (Fig. 8c, d). However, TEM and XRD
of Cr, Mo, and Ni in the weld zone with the meager analyses are required to confirm the presence of exact Nb-
depletion of Fe. This could be reasoned to the migration of rich precipitates formed in the weld zone.
elements during welding. However, there is a negligible
elemental movement experienced from the weld zone to 3.4 Hardness Test
Inconel 625 side or vice versa. SEM/EDAX analysis at the
weld zone envisaged the presence of macro-voids and the It is evident from the hardness profile that the average
presence of secondary phases in the voids. The SEM image hardness of GTA weldments employing ER2209 was found
of ERNiCrMo-3 weld and Inconel 625 shows the fully to be higher as compared to ERNiCrMo-3 weldments, as
austenitic matrix containing several types of precipitates. shown in Fig. 9. The average hardness at the cap, middle,
These precipitates were found in the dendrite and along the and root pass in the weld zone of ER2209 filler was found
grain boundaries. Inconel 625 contains 0.235 wt% Ti, to be 243.7, 265.2, and 280.2 HV and for the filler

123
1028 R. Sridhar et al.: Acta Metall. Sin. (Engl. Lett.), 2014, 27(6), 1018–1030

employing ERNiCrMo-3, the hardness values were

observed to be 212.6, 207.3, and 253 HV, respectively.
Hardness plots clearly envisaged that the hardness at the
weld zone has been plummeted compared to the other
zones of the weldments. The average hardness at the root
pass of the ER2209 weld zone is found to be greater than
ERNiCrMo-3 weld. Lower hardness envisaged at the weld
interface of ER2209 weldments in different passes shall be
attributed to the absence of chromium which gets depleted
during welding and forms the voids as evident from the
SEM/EDAX analysis. The variations in the hardness
observed at the weld zone shall be attributed to the for-
mation of different grains such as columnar, cellular, and
dendritic due to solidification and multi-pass welding. The
hardness at the cap of the ERNiCrMo-3 weldments can be
featured due to the presence of higher amounts of Nb
which in turn increase Nb-rich precipitates. The hardness
values at the cap and root passes show not much variation
for ERNiCrMo-3 weldments. However, there exists a
slightly higher difference in the hardness values across the
different passes of the ER2209 weldments. This may be
explained by the fact that the cooling rate after the first pass
is the highest and so a finer dendrite structure results in the
root compared to that in the cap [17]. Also it has been
shown that the temperature difference that exists between Fig. 10 Tensile samples of GTA weldments of Inconel 625 and SAF
2205 employing ER2209 a, ERNiCrMo-3 b fillers
the cap and the root results in plastic deformation within
the root [18]. Lower hardness at the weld zones of these
dissimilar joints could be attributed to the higher heat ER2209 weldments is found to be greater than the ERNi-
inputs developed during welding which resulted in slower CrMo-3 weldments. This is also supporting the results
cooling rate and hence contributed for the coarse grains in obtained from the hardness studies that the average hard-
the weld zone. It is known fact that the coarse grains ness at the weld zone employing ER2209 is found to be
normally result in lower hardness. higher than the ERNiCrMo-3 weld zone. The reason could
be the formation of richer amount of carbides and the
3.5 Tensile and Impact Studies presence of nitrogen in the filler and parent metal of SAF
2205. SEM fractographs at the cap zone of the ER2209
It is inferred from the tensile studies on these dissimilar weldments showed the presence of less dimpled with fewer
joints that the fracture occurred at the weld zone for both ridges which confirmed that the fracture mode was shear
the filler metals in all the trials. Tensile test trials showed ductile. The cap fractograph presented in Fig. 11b of ER-
that there were no much variations in the tensile strength of NiCrMo-3 weldments envisaged the brittle fracture. The
the weldments for both the fillers. The fractured tensile presence of Nb increases the fragility of the welds. Similar
samples are shown in Fig. 10. The average tensile strength observations indicated that Nb addition reduces the for-
of 718.3 MPa for ERNiCrMo-3 and 725.7 MPa for mation of ductile tear ridges and therefore suggests that
ER2209 weldments has been observed. Tensile properties increasing Nb content encourages inter-dendritic fracture
of these dissimilar weldments are shown in Table 3. Ten- in the cap [19]. The specimens with high Nb addition did
sile fractures were appeared to be brittle as there were no not show large deformations, although there is evidence
evidence of appreciable plastic deformation and ductility. that the ruptured mechanism was still governed by void
Further, the SEM fractographs in Fig. 11 have shown the connection; i.e., the Nb-rich precipitates between the den-
presence of lesser amounts of voids with cracked bound- drites (Fig. 11b). Similar observations were made by Lee
aries for both the weldments. As evident from the hardness et al. [11] where the Nb additions have no significant role
studies, the weld hardness is lower than that of the parent in increasing the strength and in all the levels of Nb
metals. It could be well noticed that there is a proper match addition, the fracture occurred at the fusion zone.
of hardness results with weld strength of these bimetallic Charpy V-notch studies carried out on these dissimilar
joints. Also it is noticed that the average tensile strength of coupons clearly envisaged that the impact toughness of the

123
R. Sridhar et al.: Acta Metall. Sin. (Engl. Lett.), 2014, 27(6), 1018–1030 1029

Table 3 Tensile properties of the dissimilar Inconel 625 and SAF 2205 GTA weldments with different fillers
Filler Trial Maximum UTS Average Young’s Elongation Fracture
load (kN) (MPa) UTS (MPa) modulus (GPa) at break (%) zone

ER2209 1 22.98 751 725.7 58.73 26.94 Weld region

2 22.27 728 65.16 32.24
3 21.35 698 57.91 8.8
ERNiCrMo-3 1 22.20 726 718.3 49.91 16.59 Weld region
2 20.03 663 56.72 10.89
3 23.45 766 56.53 17.92

Fig. 12 Impact test samples of GTA weldments of Inconel 625 and

SAF 2205 employing ER2209 a, ERNiCrMo-3 b fillers

Fig. 11 SEM fractographs on the tensile coupons of GTA weldments weldments; micro-voids and dimples with the clusters of
of Inconel 625 and SAF 2205 employing ER2209 a, ERNiCrMo-3 ductile tear ridges were found spread in the ERNiCrMo-3
b fillers weldments. Charpy V-notch impact studies inferred that the
weldments undergo brittle fracture for ER2209 and ductile–
ERNiCrMo-3 weldments is found to be almost five times brittle mode of fracture for ERNiCrMo-3 filler. However, the
better than ER2209 weldments (Fig. 12). In both the trials, impact toughness is found to be greater for the ERNiCrMo-3
the ER2209 weldments have shown poor resistance to weldments (42.7 J) in comparison with the ER2209 weld-
impact loading. The fractured surfaces are then examined ments (9.8 J). SEM fractographs however showed the pre-
using scanning electron microscopy and depicted in Fig. 13. sence of dimples with the ductile tearing ridges on the
The voids were found to be lesser and ridges appearing on the fractured surface of ERNiCrMo-3 weldments. As reported
surface with cracked boundaries were witnessed for ER2209 by Vernot-Loier and Cortial [19], the lower toughness values

123
1030 R. Sridhar et al.: Acta Metall. Sin. (Engl. Lett.), 2014, 27(6), 1018–1030

precipitates in the form of segregation at the HAZ of

Inconel 625.
(5) Tensile results showed that the fracture occurred at
the weld zone in both the cases of the weldments. The
average ultimate tensile strength of the ER2209
weldments was found to be greater compared to
ERNiCrMo-3 weldments.
(6) Weld hardness was found to be plummeted compared
to other zones of the weldments.
(7) The presence of Nb-rich phases at the inter-dendritic
regions rather than the core deteriorates the mechan-
ical properties of these weldments.
(8) Impact toughness of the ERNiCrMo-3 weldments is
greater than that of the ER2209 one, and the mode of
fracture of the both weldments is ductile–brittle in
nature.

References

[1] X.Q. Wu, H.M. Jing, Z.M. Zheng, Z.M. Yao, W. Ke, Z.Q. Hu,
Mater. Sci. Eng. A 293, 252 (2000)
[2] F. Wang, D. Northwood, Mater. Charact. 31, 3 (1993)
[3] M. Blair, T.L. Stevens, Steel Casting Handbook (Steel Founders
Society of America and ASM International, Materials Park,
2000), pp. 1–13
[4] M.J. Donachie, S.J. Donachie, Superalloys: a Technical Guide
(ASM International, Materials Park, 2002), pp. 10–30
[5] J.R. Daris, ASM Handbooks (ASM International, Materials Park,
1999), pp. 18–23
Fig. 13 SEM fractographs on the impact test samples of GTA [6] J.C. Lippold, Weld. J. 62, 91 (1983)
weldments of Inconel 625 and SAF 2205 employing ER2209 a, [7] M. van Wijngaarden, J. Chater, Stainless Steel World, 42 (2006)
ERNiCrMo-3 b fillers [8] H. Naffakh, M. Shamanian, F. Ashrafizadeh, J. Mater. Process.
Technol. 209, 3628 (2009)
indicated in the weldments could be attributed to the pre-
[9] H. Shah Hosseini, M. Shamanian, A. Kermanpur, Mater. Cha-
sence of M6C, M23C6 carbides which were also witnessed ract. 62, 425 (2011)
from the EDS results. [10] R. Dehmolaei, M. Shamanian, A. Kermanpur, Mater. Charact.
59, 1447 (2008)
[11] H.T. Lee, S.L. Jeng, C.H. Yen, T.Y. Kuo, J. Nucl. Mater. 335,
59 (2004)
4 Conclusions [12] D.J. Lee, Y.S. Kim, Y.T. Shin, E.C. Jeon, S.H. Lee, H.J. Lee,
S.K. Lee, J.H. Lee, H.W. Lee, Metall. Mater. Int. 16, 813 (2010)
(1) Sound joints of Inconel 625 and Duplex stainless steel [13] G. Belloni, G. Caironi, A. Gariboldi, A. Lo Conte, P. di Milano,
Trans. SMiRT 16, 1546 (2001)
could be made possible on employing GTA welding
[14] J.R. Crum, L.E. Shoemaker, S.D. Kiser, Special alloys and
employing these filler wires. overmatching welding products solve FGD corrosion problems.
(2) Microstructure observations showed the grain coars- Special Metals Welding Products, Technical Paper
ening at the HAZ of SAF 2205 and segregation [15] J. Dupont, J. Lippold, S. Kiser, Welding Metallurgy and Wel-
dability of Nickel-Base alloys (Wiley, New York, 2009)
effects at the HAZ of Inconel 625 and these micro-
[16] J.N. DuPont, C.V. Robino, A.R. Marder, M.R. Notis, J.R.
structural changes are attributed to the higher heat Michael, Metall. Mater. Trans. A 29, 2785 (1998)
inputs developed during welding. [17] J.N. DuPont, C.V. Robino, A.R. Marder, Metall. Mater. Trans. A
(3) The formation of the precipitation zone adjacent to 29, 2797 (1998)
[18] I. Gowrisankar, A.K. Bhaduri, V. Seetharaman, D.D.N. Verma,
the weld zone occurred due to the temperature
D.R.G. Achar, Weld. J. 66, 147s (1987)
differences between the base and filler metals. [19] C. Vernot-Loier, F. Cortial, Superalloy 718/625 (TMS, War-
(4) Line mapping analysis showed that the elements such rendale, 1991), p. 409
as Nb, Mo, Cr have been enriched and appeared as

123