Accepted Manuscript

Microstructure and Hardness Studies of Electron Beam Welded Inconel 625 and
Stainless Steel 304L

M. Shakil, M. Ahmad, N.H. Tariq, B.A. Hasan, J.I. Akhter, E. Ahmed, M. Mehmood,
M.A. Choudhry, M. Iqbal

PII: S0042-207X(14)00297-8
DOI: 10.1016/j.vacuum.2014.08.016
Reference: VAC 6412

To appear in: Vacuum

Received Date: 19 April 2014

Please cite this article as: Shakil M, Ahmad M, Tariq NH, Hasan BA, Akhter JI, Ahmed E, Mehmood M,
Choudhry MA, Iqbal M, Microstructure and Hardness Studies of Electron Beam Welded Inconel 625 and
Stainless Steel 304L, Vacuum (2014), doi: 10.1016/j.vacuum.2014.08.016.

This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to
our customers we are providing this early version of the manuscript. The manuscript will undergo
copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please
note that during the production process errors may be discovered which could affect the content, and all
legal disclaimers that apply to the journal pertain.
 ACCEPTED MANUSCRIPT

Microstructure and Hardness Studies of Electron Beam Welded Inconel 625

and Stainless Steel 304L

M. Shakil*,a,b, M. Ahmadb, N.H. Tariqc, B.A. Hasanc, J.I. Akhterb, E. Ahmedb, M.

Mehmoodc, M.A. Choudhrya, M. Iqbalb

PT
a
Department of Physics, The Islamia University of Bahawalpur, Bahawalpur 63100,

Pakistan.

RI
b
Physics Division, Pakistan Institute of Nuclear Science & Technology, P.O. Nilore,

SC
Islamabad, Pakistan
c
DMME, Pakistan Institute of Engineering and Applied Sciences, P.O. Nilore, Islamabad,

U
Pakistan
AN
Abstract:

In this study, electron beam welding of dissimilar Inconel 625 and SS 304L alloys was
M

successfully performed by employing optimized electron beam welding parameters. The welded

joint was characterized using SEM/ EDS, XRD and micro-hardness tester. The welded joint was
D

found homogeneous, well bonded and defect free. Two types of microstructure i.e. columnar
TE

dendritic and cellular dendritic were observed in the fusion zone. The development of different

microstructures in the fusion zone was attributed to the localized cooling effects during
EP

solidification. Few micro-cracks along with dendrites splitting were observed in the vicinity of
C

end crater that was mainly due to the segregation of S element. A significant variation of Ni and
AC

Fe was observed across Inconel/FZ and FZ/SS interfaces due to their fast diffusion in the melt

pool. Micro-hardness measurements across Inconel/FZ and FZ/SS interfaces showed an

increasing trend in the FZ from SS 304L towards Inconel base alloy.

Key words: Inconel 625; SS 304L; Electron Beam Welding (EBW); Fusion zone; Scanning
electron microscopy (SEM)
*Corresponding Author: Contact no. +92 321 6828243,
E-mail addresses: [email protected], [email protected] (M. Shakil)
1
 ACCEPTED MANUSCRIPT

1. Introduction

Welding of dissimilar materials has been the subject of interest of many researchers due to their

high demand for various industrial applications. Dissimilar materials are welded in order to

increase flexibility in design and application. However, this practice may often create problems

PT
that severely affect the performance during service. The welding of two dissimilar alloys with

different thermal expansion coefficient, melting temperature is a complicated task [1, 2]. Inconel

RI
625 is a Nickel base superalloy strengthened by a solid solution of molybdenum and niobium in

SC
its nickel-chromium matrix. It has widespread applications in aerospace, marine, chemical,

petrochemical industries due to its superior physical and chemical properties i.e high strength,

U
corrosion and creep resistance at elevated temperatures. The alloy has also some applications in
AN
nuclear power plant reactor-core and control rod components of pressurized water reactor. It is

also being used as a heat exchanger tube in ammonia cracker plants of heavy water production
M

[3, 4]. Stainless steel 304L is a less expensive structural material that found applications within

general corrosive environments, reprocessing plants and nuclear power plant components.
D

During welding process its lower carbon content reduces the precipitation of carbide in the
TE

fusion zone. Nickel- based alloys and stainless steel (SS) have extensive applications in the

manufacturing of components used in pressurized water reactors and boiling water reactors in
EP

nuclear power plants [5]. To fabricate these parts traditional welding techniques such as shield
C

metal arc welding (SMAW), gas tungsten arc welding (GTAW) and gas metal arc welding
AC

(GMAW) are normally used. However, in conventional welding techniques large numbers of

welding passes are required to join thick parts for achieving sufficient strength in the weld metal.

The increased number of passes produces greater thermal cycling effect resulting in the

formation of carbide. The high heat input and large heat affected zone (HAZ) in afore-mentioned

2
 ACCEPTED MANUSCRIPT

conventional welding techniques result in reduced corrosion resistance and mechanical

properties of the welded joint [6-14].

Electron beam welding (EBW) is a high energy density beam welding that is capable of welding

thick parts in a single pass. EBW has a number of significant benefits over other welding

techniques i.e lower total heat input, smaller HAZ, a high depth to width ratio and low residual

PT
stress in the weld. EBW can produce deep weld as compared to other welding processes. These

RI
advantages thus provide defect free welded joint as well as reduce the welding stresses in the

fusion zone (FZ). In addition, the high efficiency of EBW reduces the welding time, improves

SC
the safety and reliability of welded parts [15-20]. In the present work, EBW has been

successfully employed to join Inconel 625 with SS 304L to achieve defect free dissimilar welded

U
part. It is anticipated that the results of this work would be very useful for advanced industrial
AN
applications of dissimilar Inconel 625/ SS 304L welded joints.
M

2. Experimental

The samples of Inconel 625 and SS 304L with dimensions 10 mm ×10 mm× 3 mm were cut
D

from the sheets using slow speed cutter. Nominal composition of as received Inconel 625 and SS
TE

304L is provided in Table 1. The samples were ultrasonically cleaned and polished on a lapping

machine using diamond paste down to 0.25 µm. Joining sides of both metals were also polished
EP

up to the same level. The samples were placed together in a locally designed special die to hold
C

them in such a way that they could not separate from each other during the course of welding.
AC

Electron beam welding was performed in a continuous mode, normal to the sample surface by

employing various combinations of welding parameters, the beam current and welding speed,

under vacuum and optimized values were selected. The welding parameters used are provided in

the Table 2.A schematic illustration of the design of the weldment is shown in Fig. 1.The

weldment was placed on a thick copper sheet (large heat sink) to ensure that the heat source of

3
 ACCEPTED MANUSCRIPT

electron beam is uniformly distributed along the Inconel 625 and SS 304L plate. The experiment

was performed using TECHMETA, France EBW machine having maximum voltage 60 kV and

maximum current 50 mA. After welding, the welded sample was again polished and etched

chemically. The microstructure and elemental distribution was investigated by Scanning Electron

Microscopy (SEM) equipped with Energy Dispersive Spectrometer (EDS). In order to determine

PT
the phases in the weld zone X-ray diffraction (XRD) scan was performed. Finally, microhardness

RI
was measured across the weld interface using Vickers hardness tester by using a load of 300g at

room temperature.

SC
3. Results and discussion

In order to achieve a crack and defect free weld interface, the beam current (I) and welding speed

U
(v) were optimized after conducting series of preliminary experiments with the aim to reduce
AN
welding residual stresses at weld interfaces. Welding residual stresses could be reduced by
M

controlling cooling rate (R) of the fusion zone which has following relationship with beam

current and welding speed [21];

R α (v / I )2
TE

According to the above equation, a moderate cooling rate can be achieved through the balance

between welding speed and beam current. Based on the observed microstructure, the operating
EP

parameters were optimized by a systematic increase in the beam current from 20 to 30 mA at

different welding speeds ranging from 400 to 700 mms-1. Fig. 2 shows the low magnification
C

SEM micrographs of the welded samples under different welding parameters. A sound and
AC

defect free joint was achieved at beam current of 25 mA and 600 mms-1 welding speed as shown

in Fig. 2(c).

3.1. Microstructure

Fig. 3 shows low magnification SEM micrograph of EB welded Inconel 625 and SS 304L.

Interfaces of FZ with both of the alloys are clearly seen. It was observed that FZ was gradually
4
 ACCEPTED MANUSCRIPT

widened towards its end (bottom of Fig. 3). This phenomenon was mainly due to high heat

buildup at the end of FZ [22]. In order to further investigate the bonded region, Inconel/ FZ and

FZ/SS interfaces (marked by regions 1 and 2 respectively in the Fig. 3) were further magnified as

shown in Figs. 4 and 5. It is quite evident from Figs. 4 and 5 that the interfaces are quite

homogeneous, well bonded and defect free. Only few micro cracks along with some dendrite

PT
splitting were observed in the vicinity of end crater (marked by region-4 in Fig.3), otherwise, the

RI
entire weld was crack free. The formation of these micro cracks and separated grain boundaries

at the end of FZ is either due to the solidification shrinkage during solidification and/ or the

SC
segregation of minor alloying elements/impurities [23,24]. Kou et. al. reported thermal shrinkage

liable for these cracks/defects that exist in intergranular and dendritic morphology in the FZ

U
during the end stage of solidification [24] (Fig. 7). These micro cracks were also reported by
AN
Patterson et. al. during the welding of austenitic stainless steels and nickel alloys due the
M

segregation of S, P, Ti and Nb elements [25]. However, in the present scenario, segregation of S

was only witnessed around cracks as evident by EDS point analysis shown in Fig. 7. The
D

chemical composition of various regions of the weld is summarized in the Table 3. The
TE

composition determined in the cracks showed an increased amount of Sulphur as compared to

the base alloys composition.

Fig. 8 shows the high magnification SEM micrograph of FZ area (marked by region-3 in Fig.3)

showing dendritic microstructure composed of two types namely, columnar dendrites and
C

cellular dendrites. This microstructure is attributed to the different thermo-physical properties of

Inconel 625 & SS-304L including thermal conductivity and specific heat which vary with

temperature. For instance the thermal conductivities of Inconel 625 & SS-304L at 100°C are

11.4 and 16.3 Wm-1K-1 that changes to 17.5 and 21.4 Wm-1K-1 at 500 °C. It is obvious that in the

weldment, the Inconel 625 side would have a significantly steeper temperature gradient as

compared to SS-304L starting from the centre line of the FZ to the end to the specimen at either
5
 ACCEPTED MANUSCRIPT

side. S. Kou et al. [1] indicated that solidification structure of welded products is determined by

the degree of constitutional supercooling (i.e. the ratio of the temperature gradient (G) to the

growth rate (R), G/R) during solidification. Furthermore, the product G×R is equivalent to the

cooling rate. Larger thermal gradients during EBW favor the formation of complex

microstructures due to the non-equilibrium conditions during the rapid solidification of FZ. The

PT
thermal gradients are larger for high energy density welding processes such as EBW and laser

RI
beam welding. It is lowest along the weld centerline and found maximum at the fusion line when

moving outward from centerline along the solidification front [26]. In Fig. 8, the area consisted

SC
of columnar dendrites is very small and located near the edge of the FZ towards the Ni-alloy

solid surface. The possible reason for the formation of columnar dendrites may be the much

U
lower thermal conductivity of Inconel 625 in comparison to SS 304L which results in small heat
AN
dissipation in this area. As a result, slightly higher values of G/R, associated with lower
M

constitutional supercooling, were achieved. Similar dendritic microstructure was observed by

other researchers in Ni-base single crystal alloy during the EB treatment [20]. The dendritic
D

growth and latent heat dissipation in FZ are anti-parallel in the case of columnar dendritic
TE

structure, whereas, it is parallel in cellular structure [20]. The average composition taken in the

core of dendrites and at inter dendritic area is listed in the Table 3. The dendritic core was found
EP

to be enriched in Ni, Mo, and Nb as compared to the other areas around the core. On the other

hand uniform equiaxed cellular structure (not shown) was observed in the central region of the
C

FZ. The occurrence of equiaxed structure in the central area is attributed to the uniform and
AC

symmetric cooling as compared to the edges of the FZ. The equiaxed cellular structure is highly

depending upon the alloy composition, turbulence and thermal gradients at S/L interface. As

revealed by EDS discussed later (Table 2), the chemical composition of FZ was based on a

mixture of both Inconel 625 and SS 304L. So, FZ may have wide spread between Liquidus and

Solidus temperatures which often favours formation of he equiaxed cellular structure. Secondly,
6
 ACCEPTED MANUSCRIPT

the equiaxed cellular structure might be formed due to the low thermal gradients at S/L interface

towards SS-304 side. In other words, much lower G/R ratio was achieved at central region.

It is again clear from Table 3 that the dendritic cells boundaries are enriched in Mo and Nb as

compared to the inner areas of cellular structures. Since Mo and Nb have large radii in

comparison to the other elements in the melt pool, they cause segregation during the terminal

PT
solidification. It has been reported in the literature that in Nb- bearing superalloys, the terminal

RI
solidification is different due to the presence of Nb element which is a strong segregant

controlling terminal solidification reaction [27].

SC
In order to determine compositional variation across Inconel/FZ interface, FZ and FZ/SS

interface, EDS microanalysis (Fig. 9) was performed along the white dash line indicated in Fig 3.

U
A significant variation of Ni and Fe was observed across Inconel/FZ and FZ/SS interfaces. The
AN
large variation of Ni and Fe in FZ was mainly due to their fast diffusion in the melt pool, since
M

Ni and Fe have small atomic radii in comparison to other weld metal constituent elements.

3.2 X-ray diffraction

XRD pattern of Inconel 625, SS 304L, and welded zone is shown in the Fig. 10. It was also
TE

revealed that the XRD pattern of Inconel alloy was much similar with that of FZ. This indicates

that both FZ and Inconel alloy have almost same crystal structure. However, EDS analysis
EP

(Table 2) revealed that the chemical composition of FZ was based on a mixture of Inconel 625

and SS 304L. It can be clearly seen from Fig. 10 that the base Inconel 625 alloy as well as FZ
C

show the presence of Ni (fcc) solid solution with lattice parameter of 360.58 pm and 360.60 pm
AC

respectively. More or less same γ-Ni matrix lattice parameter of base Inconel 625 alloy as well as

FZ indicates that no intermetallic phases like γ', γ'' or δ were precipitated out in the matrix. Since

the cooling rate during EBW is very high, most of the strengthening elements like Nb and Mo

remained in the solid solution of γ-Ni matrix and formed no precipitates. It was also observed

that XRD peaks intensity of FZ was decreased and slightly shifted towards the lower angles
7
 ACCEPTED MANUSCRIPT

which might be attributed to the residual stresses induced by fast cooling of FZ [20] and large

variation of Ni and Fe content in the FZ zone (Table 3, Fig. 9).

3.3 Micro-hardness

The micro-hardness measurements were made by using Vickers hardness tester. Fig. 11 shows

the difference in hardness values between the Inconel 625, FZ, and SS 304L. Inconel 625 is a

PT
solution strengthened Ni-based superalloy with higher hardness values as compared to the SS

RI
304L. The approximate length of FZ was found to be 2.4 mm. The hardness measured in the FZ

showed overall an increasing trend moving from SS 304L side towards the Inconel alloy.

SC
The hardness in the weld increased from 150 to 225 Hv. This might be caused by alterations in

the structure across the weld cross-section. However, within the FZ the hardness values were

U
scattered that may be attributed to the convection effects in the weld metal as well as due to the
AN
variation of supersaturation of Nb and Mo strengthening elements in the γ-matrix.
M

4. Conclusions

Electron beam welding of dissimilar Inconel 625 and SS 304L alloys was achieved by employing
D

optimized electron beam welding parameters. The welded joint was found homogeneous, well
TE

bonded and defect free. Different types of microstructures were evolved in the fusion zone due to

the localized cooling effects during solidification. Few micro-cracks along with dendrites
EP

splitting were observed in the vicinity of end crater which was mainly due to the segregation of S
C

element. Micro-hardness measurements across Inconel/FZ and FZ/SS interfaces showed an

increasing trend in the FZ from SS 304L towards Inconel base alloy.

Acknowledgments

The authors gratefully acknowledged the financial support provided by the Higher Education

Commission (HEC) Pakistan for this work.

8
 ACCEPTED MANUSCRIPT

References

[1]. Sun Z, Karppi R. J Mater Process Technol 1996; 59: 257-267.

[2]. Richard Avery E. Pay Attention to Dissimilar-Metal Welds; Avery Consulting Associates

Inc. 2003, pp. 1-6.

[3] Sims CT, Stollof NS, Hagel WC. Superalloys. 2nd ed. New York; Wiley-Inter science: 1987.

PT
[4] Shankar V, Rao KBS, Mannan SL. J Nucl Mater 2001; 288: 222-232.

RI
[5] Ishihiara T. Weld J 1989; 68: 209-216.

[6] Stiller K. Nilsson JO, Norring K. Metall Mater Trans A 1996; 27A: 327–341.

SC
[7] Kuo TY, Lee HT. Mater Sci Eng A 2002; 338: 202–212.

[8] Lee HT, Jeng SL. Sci Technol Weld Joining 2001; 6: 225–234.

U
[9] Lee HT, Kuo T. Sci Technol Weld Joining 1999; 4: 94–103.
AN
[10] Horn RM, Gordon GM, Ford FP, Cowan RL. Nucl Eng Des 1997; 9: 313–325.
M

[11] Jeng SL, Lee HT, Rehbach WP, Kuo TY, Mayer JP. Mater Sci Eng A 2005; 397: 229–238.

[12] Lee HT, Jeng SL, Yen CH, Kuo TY. J Nucl Mater 2004; 335: 59–69.
D

[13] Lee HT, Jeng SL, Kuo TY. Metall Mater Trans A 2003; 34: 1097–2005.
TE

[14] Kuo TY, Lee HT, Tu CC. Sci Technol Weld Joining 2003; 8: 39–48.

[15] Stone HJ, Roberts SM, Reed RC. Metall Mater Trans A 2000; 31A: 2261–2273.
EP

[16] Sun Z, Karppi R. J Mater Process Technol 1996; 59: 257–264.

[17] Lim YS, Kim JS, Kwon HS. Mater Sci Eng A 2000; 279: 192–200.
C

[18] Tariq NH, Shakil M, Hasan BA, Akhter JI, Haq MA, Awan NA. Vacuum 2014; 101: 98-
AC

101.

[19] Ahmad M, Ali G, Ahmed E, Haq MA, Akhter JI. Appl Surf Sci 2011; 257: 7405-7410.

[20] Ahmad M, Ali G, Akhter JI, Akhtar M, Chaudhry MA. J Alloy Compd 2008; 461: 102–105.

[21] Yokoyamaa Y, Abeb N, Fukaura K, Shinohara T, Inoue A. Mater Sci Eng A 2004; 375–

377: 422–426
9
 ACCEPTED MANUSCRIPT

[22] Schultz H. Electron beam welding; Woodhead Pub: 1993.

H. Shultz, Electron Beam Welding (Woodhead, 1993).

[23] Ahmad M, Akhter JI, Shahzad M. Akhtar M. J Alloy Compd 2008; 1–2: 131-134.

[24] Kou S. JOM 2003; 55: 37-42.

[25] Patterson R A, Milewisky J O. Weld Research Suppl 1985; 8: 227s-231s.

PT
[26] David SA, Vitek JM. Int Mater Rev 1989; 34:213-245.

RI
[27] Dupont JN, Banovic S, Marder A. Weld J 2003; 82:125s-135s.

U SC
AN
M
D
TE
C EP
AC

10
 ACCEPTED MANUSCRIPT

Table Captions

Table 1 Nominal composition (wt.%) of welded alloys.

Table 2 Electron beam welding parameters.

Table 3 Quantitative analysis of different regions in the FZ.

PT
RI
U SC
AN
M
D
TE
C EP
AC

11
 ACCEPTED MANUSCRIPT

Figure Captions

Fig. 1 A schematic illustration of the design of the weldment with a large heat sink of

copper.

Fig. 2 Low magnification SEM micrographs of Inconel 625/SS 304L welds obtained at

PT
different welding parameters. The regions marked by rectangles show weld defects.

RI
Fig. 3 SEM micrograph of Inconel 625/SS 304L weld at optimized welding conditions.

SC
Fig. 4 SEM micrograph showing the interface between Inconel 625 and FZ.

Fig. 5 SEM micrograph showing interface between Fusion zone and SS.

U
AN
Fig. 6 SEM micrograph showing the presence of cracks at the bottom of FZ.

Fig. 7 High magnification SEM micrograph of the region marked by a rectangle in Fig. 6.
M

Fig. 8 SEM micrograph of a FZ region (marked by region-3 in Fig. 3) showing columnar

and cellular dendrites.

Fig. 9 EDX point analysis across FZ/Inconel and FZ/SS interfaces.

Fig. 10 XRD pattern of (a) Inconel 625 (b) SS 304L and (c) Fusion zone.
EP

Fig. 11 Microhardness profile across the FZ.

12
 ACCEPTED MANUSCRIPT

Table 1

Alloys Ni Cr Mn Si Fe Mo Nb Al Ti S
Inconel 625 Bal. 22.76 0.20 0.37 4.18 7.96 2.86 0.36 0.39 0.28
SS 304L 9.72 19.32 1.78 0.47 Bal. -- -- -- -- 0.23

PT
RI
U SC
AN
M
D
TE
C EP
AC

13
 ACCEPTED MANUSCRIPT

Table 2

Parameters Values

Voltage 30 kV

PT
Current 20, 25, 30 mA,
Power 600, 750, 900 W
1.3×10-6 mbar

RI
Vacuum
Welding speed 400, 500, 600, 700 mm/min

SC
Welding Pass 1
Welding mode Continuous

U
AN
M
D
TE
C EP
AC

14
 ACCEPTED MANUSCRIPT

Table 3

Ni Cr Mo Nb Fe Si Al S Mn

Inconel 625(base metal) 59.97 22.76 7.96 2.86 4.18 0.37 0.36 0.28 0.20

SS 304L (base metal) 9.72 19.32 -- -- 68.48 0.47 -- 0.23 1.78

PT
Fusion Zone 27.83 22.55 2.50 1.03 43.98 0.88 0.49 -- 0.89

Cracks boundary 31.01 23.53 -- -- 40.00 -- -- 2.57 --

RI
Inside the cracks 28.61 26.17 -- -- 41.62 1.19 -- 1.89 --

SC
Cellular boundaries 28.26 22.70 2.88 1.13 42.26 0.98 -- -- --

Inside the cells 27.59 21.82 2.15 -- 46.35 -- -- -- --

U
Dendrite cores 36.66 22.69 3.79 2.00 32.73 -- -- -- --
AN
Inter dendritic area 29.94 22.46 2.21 -- 43.37 -- -- -- --
M
D
TE
C EP
AC

15
 ACCEPTED MANUSCRIPT

PT
RI
U SC
AN
M
D

Fig. 1
TE
EP
C
AC

16
 ACCEPTED MANUSCRIPT

PT
RI
U SC
AN
M
D
TE
EP
C
AC

Fig. 2

17
 ACCEPTED MANUSCRIPT

PT
RI
U SC
AN
M
D
TE
EP
C

Fig. 3
AC

18
 ACCEPTED MANUSCRIPT

PT
RI
U SC
AN
M
D
TE
EP

Fig. 4
C
AC

19
 ACCEPTED MANUSCRIPT

PT
RI
U SC
AN
M

Fig. 5
D
TE
EP
C
AC

20
 ACCEPTED MANUSCRIPT

PT
RI
U SC
AN
M
D
TE
EP

Fig. 6
C
AC

21
 ACCEPTED MANUSCRIPT

PT
RI
U SC
AN
M
D
TE

Fig. 7
EP
C
AC

22
 ACCEPTED MANUSCRIPT

PT
RI
U SC
AN
M
D
TE
EP
C

Fig. 8
AC

23
 ACCEPTED MANUSCRIPT

PT
RI
U SC
AN
M
D
TE
EP
C

Fig. 9
AC

24
 ACCEPTED MANUSCRIPT

PT
RI
U SC
AN
M
D
TE
EP
C

Fig. 10
AC

25
 ACCEPTED MANUSCRIPT

PT
RI
U SC
AN
M
D
TE
EP

Fig. 11
C
AC

26
 ACCEPTED MANUSCRIPT

Research Highlights

• Welding of Inconel 625 to SS 304L alloys was achieved by electron beam

• Welding was performed at accelerating voltage of 30 KV and 25 mA current

PT
Hardness increased in FZ from SS 304L towards Inconel 625 alloy.

RI
U SC
AN
M
D
TE
C EP
AC