- Email: [email protected]
Nb-based alloy dissimilar joints without interlayer
Accepted Manuscript Electron beam welding of difficult-to-weld austenitic stainless steel/Nb-based alloy dissimilar joints without interlayer Ali Hajitabar, Homam Naffakh-Moosavy PII:
S0042-207X(17)30945-4
DOI:
10.1016/j.vacuum.2017.09.046
Reference:
VAC 7622
To appear in:
Vacuum
Received Date: 16 July 2017 Revised Date:
27 September 2017
Accepted Date: 28 September 2017
Please cite this article as: Hajitabar A, Naffakh-Moosavy H, Electron beam welding of difficult-toweld austenitic stainless steel/Nb-based alloy dissimilar joints without interlayer, Vacuum (2017), doi: 10.1016/j.vacuum.2017.09.046. 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 Electron Beam Welding of Difficult-to-Weld Austenitic Stainless Steel/Nb-based Alloy Dissimilar Joints without Interlayer Ali Hajitabar, Homam Naffakh-Moosavy∗
(TMU), Tehran, Iran Abstract
RI PT
Department of Materials Engineering, PO Box: 14115-143, Tarbiat Modares University
SC
Dissimilar butt welding of Nb-1Zr to the 321 stainless steel was studied using electron beam welding. The aim of this study was investigation of the offset and microdilution
M AN U
effects on the weldability, microstructure and mechanical properties of this dissimilar joint. The offset of 50-50 and welding parameters of 13mA and 1343V showed the best weldability and penetration for the joint. The microdilution of Nb in the weld changes by going far from the base metals. The microdilution of Nb is the minimum at root of the weld
TE D
(~ 0.2) leading to formation of less brittle phases. As a result, no cracks are observed at the root. The microdilution of Nb at the top of weld just near to Nb-based alloy at a 200µm narrow zone under surface increases (~ 0.8) which leads to formation of more brittle
EP
phases and subsequent microcracks. Applying 200µm thickness machining on the surface
AC C
of weld can give a completely free of crack welds that show strength more than the Nballoy base. It is concluded that a sound weld can be achieved by 50-50 offset and Nb microdilution less than 0.6. No interlayer and/or transition piece are required, and appropriate mechanical properties can be obtained.
∗
Corresponding Author: Telefax.:+98 21 82884928; E-mail address: [email protected] (H. Naffakh-Moosavy)
ACCEPTED MANUSCRIPT Keywords: Microdilution; Nb-1Zr Alloy; 321 Stainless Steel; Electron Beam Welding; Weldability.
1. Introduction
RI PT
Niobium is a refractory metal with melting point of 2468 ̊Ϲ [1]. Nb-based alloys have shown excellent properties at high-temperatures, high melting point and appropriate
SC
creep properties [2]. Pure niobium and the Nb–based alloys have been widely used in the nuclear, aerospace industries and biomedical applications [3]. Dissimilar metals welding is
M AN U
an important technology in the modern industry. It is characterized by the ability to maximize the use of a variety of advantages of materials for industrial production [4-8]. Welding of dissimilar alloys is attractive for the industry because of its potential reduction of cost and weight, increasing design flexibility and complexity, and also improving the
TE D
product performance [9, 10]. Problems in dissimilar welding are attributed to the differences in physical and chemical properties between the welding members and residual stresses resulting in the degradation of mechanical properties of welds [11].
EP
Electron beam welding (EBW), with high energy density, large depth to width ratio [6, 12],
AC C
is of high interest for the welding of dissimilar materials [13]. Niobium to stainless steel joints are used in fabrication of superconducting niobium cavities, which are widely used for particle accelerators [14]. But niobium forms brittle intermetallic phases (Fe2Nb and FeNb intermetallics) when alloyed with iron according to the binary Fe-Nb phase diagram (Fig. 1). It is known that weld brazing of Nb to steel has been made [16]. However, a continuous layer of Fe2Nb phase is formed at the boundary between niobium and steel. Zhao used niobium as interlayer in hot-roll bonding of Ti alloy and stainless steel [17].
ACCEPTED MANUSCRIPT First samples of Ti and Nb tubes explosion welding joint with stainless steel has been reported by Sabirov et al. [18]. Laser welding of niobium to 410 steel with a nickel interlayer produced by electro spark deposition has been reported by Baghjari et al. [19]. Offsetting the beam position results in different mechanical properties of welded joint
RI PT
[20].
All studies above proposed complicated and/or costly procedures for dissimilar welding of
SC
such difficult-to-weld joints. Therefore, the less complicated and/ or costly ideas are welcomed. In addition, in most cases the mechanical properties of the obtained welds are
M AN U
considerably lower than the base metals. So, in this study, butt welding of the Nb-1Zr niobium alloy to 321 stainless steel using an electron beam welding is investigated. The research tries to determine the optimum microdilution at which less fraction of brittle intermetallics are formed, and improved weldability is obtained. Effects of various electron
TE D
beam and process parameters on the weld profile regarding large difference in thermophysical properties of the weld counterparts are studied. The endeavor is directed towards establishing optimum process parameters with regard to the weld metal soundness and
EP
the mechanical properties of the joint.
AC C
2. Experimental procedure 2.1. Materials
Nb-1Zr alloy and 321 stainless steel in the form of plates with the dimension of 100 × 50 × 3 mm were used for bead on plate EBW. The chemical composition of base metals are shown in Table 1.
Table 1 Chemical composition (wt%) of base metals
ACCEPTED MANUSCRIPT Nb
Zr
W
Ta
Fe
Ni
Mo
Cr
C
Mn
Si
Nb-1Zr
Balance
0.8
0.6
0.5
0.04
0
0.14
0
0
0
0
AISI 321
0
0
0
0
Balance
10
0.05
18
0.08
2
1
RI PT
Element
Before welding, all specimens were cut with wire cut and then prepared carefully by fine polishing with sandpapers of various grit sizes (600#, 800# and 1000#) and surface
SC
cleaning with acetone in order to prevent weld metal contamination. Specimens were firmly fixed in the form of butt joint (Fig. 2).
M AN U
2.2 Welding methods
Lots of previous studies [6, 7, 21, 22] have shown that non-centered welding was useful for dissimilar metals welding, but it is necessary to be further optimized. Electron beam
TE D
welding process was conducted with welding parameters as shown in Table 2.
Table 2 Electron beam welding fixed parameters for joining Nb-1Zr to 321 steel sheets.
[mm/S]
85
Filament
Focus current
Chamber
current (A)
(mA)
vacuum (m bar)
46
518.8
6×10-4
AC C
10
Voltage (kV)
EP
Welding speed
2.3 Offset and Microdilution control The welded joints were prepared with different beam offsets (BOFs) in order to obtain different dilutions (i.e., overall participation of Nb entering from the base to the entire of weld, WNb,weld/WNb,base) and microdilutions (i.e., local participation of Nb entering from the base to the weld, WNb,weld/WNb,base). Schematic diagram of the welding process were shown in Fig. 3. At first, optimization of welding parameters on each of the base metals were
ACCEPTED MANUSCRIPT analyzed, individually. The optimal parameters such as depth of penetration and weld quality for each sheet of Nb-1Zr alloy and 321 stainless steel were determined. The obtained welding parameters for individual base metals were used to apply to the dissimilar joining (Table 2). The next set of experiments dealt with the dissimilar joining of
RI PT
Nb-1Zr alloy to 321 stainless steel. The dissimilar joining was performed without use of any intermediate layer. For this purpose, the dissimilar joining is designed in such a way that
SC
the spot of the electron beam irradiation is adjusted on the welding interface to obtain the area ratio of 50-50, 25-75 and 75-25 percent (Fig. 3). The set of tests are given in Table 3.
M AN U
These different ratios are lead to different dilution and microdilution levels of Nb in the weld.
Table 3
Electron beam welding variable parameters for joining Nb-1Zr to 321 steel sheets. beam current [mA]
bias voltage [V]
SN1
24
1130
electron beam alignment (BOFs) [%] 75 Nb alloy 25 321
13
1312
75 Nb alloy – 25 321
13
1343
50-50
13
1388
25 Nb alloy – 75 321
SN3
AC C
SN4
EP
SN2
TE D
sample
2.4 Characterization methods To study the weld profile and details of melted zones on both sides of the weld line, the welded samples were sectioned transversely, polished and etched for metallographic examinations. Marble etchant (50 ml H2O, 50 ml HCL and 10 gr CuSO4) was used for the etching of dissimilar electron beam welded samples.
ACCEPTED MANUSCRIPT Microstructural investigations were performed through an Olympus BX51M optical microscope and a Philips XL30 scanning electron Microscope (SEM) equipped with an energy-dispersive X-ray spectrometry (EDS) system for spot and line weight analysis and fractography, and a Philips X’Pert MPD X-ray diffraction analysis were applied for the
RI PT
phase analysis of samples. Micro Vickers hardness of the joint was measured using a micro hardness device with 100 g load level and 15s loading time. Tensile tests were conducted according to ASTM E8M by a Santam STM-50 universal testing machine at room
3. Results and discussions
M AN U
were tested and the average value was recorded.
SC
temperature. The strain rate was set at 10-3 s-1. Three tensile samples of each kind of joints
3.1. Appearance and soundness of welds
According to Table 3, the beam offset on the joint interface for samples are different.
TE D
Other fundamental parameters were fixed in accordance with Table 2. According to Table 3, for sample SN1 in which 75% of beam was concentrated on niobium based alloy, the 321 stainless steel was superheated and the joint was failed due to applying high heat
EP
input to it (current 24 mA). Sample SN4 that 75% of electron beam is focused on 321
AC C
stainless steel, the niobium based alloy was not melted and the joint was failed due to applying low heat input. According to Table 4, the failed welding joints, can be attributed to the remarkable difference in melting temperature of the two sides of the joint (2407°C for niobium alloy and about 1400°C for 321 steel), as well as high thermal conductivity of niobium alloy in comparison to 321 stainless steel (about three times). Accordingly, niobium alloys needs to heat input greater than 321 stainless steel to melt due to high thermal conductivity and high melting temperature.
ACCEPTED MANUSCRIPT The cross-section of the SN2 and SN3 samples of Nb-1Zr alloy to 321 stainless steel dissimilar welds can be seen in figures 4 (a) and (b), that have been welded by the parameters shown in Tables 2 and 3. Fig. 4a shows the cross-section of the dissimilar welding of sample SN3. In this joint, the electron beam is applied to engage each base
RI PT
metals equally (i.e., 50-50 offset). It is seen that full depth of penetration is obtained and heat input has been able to make a complete weld metal. According to Table 4, because of
SC
high melting point and high thermal conductivity of niobium than 321 stainless steel, molten pool is closer to the 321 steel, and steel side shows more participation in producing
M AN U
the weld metal (weld pool asymmetric).
Table 4 Comparison of thermo-physical properties of Nb-1Zr and 321 steel [23]. Melting
Density
Specific
Thermal
Coefficient
point
(g/cm3)
heat
conductivity
of linear
(J/g K)
(W/m k)
thermal
at 20 °C
at 25 °C
expansion
TE D
Material
2407
AC C
Nb-1Zr
EP
(°C)
321 steel
1375-
(µm/m K)
8.59
0.27
41.9
7.54
8
0.5
16.3
16.6
1400
Figure 4b shows a cross-section of the dissimilar welding of sample SN2. In this dissimilar joint, 75% of the electron beam is focused on niobium alloy. In this dissimilar joint, the penetration depth is incomplete due to the low heat input. In both images a and b of Fig.
ACCEPTED MANUSCRIPT 4, heat-affected zone (HAZ) in niobium alloy is much broader than the 321 steel due to the high thermal conductivity of niobium alloy. 3.2. Microstructural examinations and Weldability According to the binary phase diagram of Fe/Nb in Fig. 1, two types of intermetallic
RI PT
compounds including FeNb and Fe2Nb can be formed. Also, in certain composition of Fe and Nb, eutectic structure can be formed. So, weld metal can include brittle intermetallic
SC
compounds, solid solution, and eutectic structures. To determine the phases in the welds, the XRD analysis was used and the results are shown in Fig. 5. These results suggest that
M AN U
weld metal contains austenite solid solution of iron, niobium, chromium, nickel, and intermetallic compounds of Fe2Nb and maybe FeNb.
SEM image of the samples SN3 (50-50 offset) and SN2 (75 Nb alloy – 25 321 offset) are presented in figures 6 and 7. Three different regions are observed. According to the binary
TE D
phase diagram of the Fe/Nb (Fig. 1), the results of XRD analysis in Fig. 5, as well as EDS analysis points in these regions in Fig. 8, it can be predicted that the region near the base metal 321, is two-phases region includes a proeutectic phase enriched in iron and an
EP
eutectic phase includes γ-Fe and intermetallic compounds of Fe2Nb. In a region away
AC C
enough from the base metal (central area), a eutectic region consisting of Fe-rich and Fe2Nb phases are formed. The third region includes a proeutectic phase rich in niobium and a eutectic phase composed of Fe–rich phase and Fe2Nb. In Fig. 6, the brighter regions represent locations containing more niobium, and mainly are intermetallic compounds of Fe2Nb. Formation of the brittle phases and exertion of thermal stresses originating from thermal shrinkages in fusion welding, can lead to immediate cracking of the welded joint [24]. According to Fig. 6b, microcracks in the white regions can be seen. The cracks show brittleness of these areas. As noted above, these regions contain large amounts of FeNb
ACCEPTED MANUSCRIPT and/or Fe2Nb brittle intermetallic compounds. The microcracks caused by stresses arisen from contractions of solidification and cooling. The microdilution of Nb in the weld changes by going far from the base metals especially Nb-based alloy. The microdilution of Nb is the minimum at the root of the weld (~0.2)
RI PT
leading to formation of less brittle phases. As a result, no cracks are observed at the root. The microdilution of Nb at the top of weld just near to Nb-based alloy at a 200µm narrow
SC
zone under surface increases (~0.8) which leads to formation of more brittle phases and subsequent microcracks. This high microdilution of Nb at the top of weld is attributed to
M AN U
wide upper geometry of key hole, which causes more portion of Nb-based alloy to melt and enter to the weld. In the root of weld, the intensity of beam is enough for melting a thin layer of Nb alloy only. It leads to less dilution of Nb to the weld, and consequently less Fe-Nb phase to form.
TE D
Fig. 8 illustrates the EDS analysis of iron and niobium changes with the distance from the 321 stainless steel base. It shows that by moving from the steel base metal to the center of weld, iron content decreases and the amount of niobium (i.e., microdilution of Nb)
EP
increases, respectively. Fig. 9 shows the increase in the atomic percent of niobium and a
AC C
decrease in atomic percent of 321 steel elements (i.e., iron, chromium and nickel) by moving from bottom of the weld to the top. It is known that the electron beam power density decreases by penetrating inside the keyhole depth. Thus, it seems that in the deeper parts of the weld the beam intensity is enough for melting of a thin layer of Nb-1Zr alloy, only. The above mentioned experiments show that at 50% offset, a complete weld penetration for both 321 alloy and Nb-1Zr alloy is obtained. Fig. 9 shows that a thin layer of Nb-1Zr base metal is molten. According to EDS analysis at point 9, it is observed that amount of
ACCEPTED MANUSCRIPT niobium in this area is significantly higher than other regions of the weld, that is more microdilution occurred. According to the Nb microdilution of this point (~0.8) and the Fe/Nb binary phase diagram in Fig. 1, it can be predicted that FeNb intermetallic compound is formed at this region (Fig. 6b).
RI PT
3.3. Evaluation of mechanical properties
Among the above mentioned experiments the best process conditions with regard to weld
SC
penetration and soundness i.e., SN3 was selected to evaluate the mechanical properties of Nb-1Zr alloy/321 stainless steel dissimilar weld. The hardness profile of the weld is shown
M AN U
in Fig. 10 in two different parallel lines, on top, and root weld – at the bottom of the key hole. On the basis of Fig. 10, it is observed that hardness of the base metal, Nb-1Zr is less than the 321 stainless steel. Although the melting point of Niobium is high, but it has low hardness and it is known as a soft metal [25]. Impurities, rolling and annealing processes
TE D
will greatly affect the mechanical properties of niobium plates, especially its tensile strength. Myneni and coworkers have studied the mechanical properties of niobium with different levels of purity and production processes [26].
EP
The results of micro hardness show that hardness in the weld metal is more than Nb-1Zr
AC C
and 321 stainless steel base metals. This increase of hardness is attributed to strengthening by solid solution and formation of Fe2Nb, FeNb and eutectics in the weld. In the weld, the hardness profile increases from the 321 steel to Nb-1Zr alloy due to the increase in niobium content. The hardness reaches its maximum value (766 HV) at region just near to the Nb-based alloy where is related to formation of FeNb at point 9 in Fig. 9. The hardness profile of root weld can also be observed in Fig. 10 with a red line. At this profile, the hardness in the weld is considerably less than hardness profile of the top part of weld. As mentioned, the power density of electron beam decreases as penetration
ACCEPTED MANUSCRIPT increases in the depth of weld, and as a result, in the deeper parts of weld, less amount of Nb-based alloy is molten and participates in the weld i.e., less microdilution of Nb-based alloy occurs. This is the key controlling factor for minimization of brittle intermetallics to form in the weld. That is, by decreasing the melting and participation of the Nb-based alloy
RI PT
in the weld, a sound and crack-free weld can be achieved. This decrease in melting confirms the EDS results in Fig. 9. EDS analysis in Fig. 9 shows that amount of niobium
SC
considerably decreases in the depth of weld. This decrease in niobium causes decrease in formation of hard Fe2Nb and FeNb phases in the weld and as a result, the hardness
M AN U
decreases. Wang et al. [27] found that there were two hardness sub-regions in the FZ that were due to different chemistry of base metals.
The tensile test from dissimilar welded samples is performed for investigating the strength of the joint. At first, the tensile strength of base metals is measured, where for Nb-1Zr
TE D
alloy and 321 stainless steel is 525 MPa and 657 MPa, respectively. For dissimilar joint tensile test, three samples were prepared and tested. The stress-strain curves for the dissimilar joints are shown in Fig. 11. The samples are broken at the top of weld metal just
EP
near to the niobium side, where the high microdilution for Nb (~80) is recorded (Fig. 12).
AC C
The microcracks are formed in a 200µm length zone due to formation of high amount of Fe-Nb phase. As it was shown in the micro- hardness test, the maximum value for hardness in the weld was at regions near to the Nb-1Zr alloy, which showed the considerable amount of FeNb intermetallic. The tensile strength of tested samples was 170 MPa which is 62% of the autogenously welded base metal Nb-1Zr alloy obtained without any additional operations such as mechanical grinding of surface. Also, the elongation of 1.2% is obtained for the dissimilar joint. By removing this 200µm narrow zone by mechanical
ACCEPTED MANUSCRIPT grinding, the higher strength and elongation in comparison to the Nb-based alloy can be obtained. The fractography of tensile sample of dissimilar weld is shown in Fig. 13. It indicates that fracture of joints occurs in brittle failure mode by the obvious cleavage steps. EDS analysis
RI PT
of the fracture surface reveals presence of high amount of niobium, which can be attributed to brittle intermetallic compounds of FeNb, just close to Nb -1Zr alloy base
SC
metal. Conclusions
M AN U
Dissimilar welding of Nb-1Zr alloy to 321 stainless steel was studied using an electron beam welding for the first time. The high lights below are addressed: 1. Multiple offsets and heat inputs are investigated. The offset of 50-50 and heat input giving complete penetration for 321 stainless steel (i.e., 13mA, 1343V)
TE D
showed the best weldability and penetration for the dissimilar joint. 2. The microdilution of Nb in the weld changes by going far from the base metals. The microdilution of Nb is the minimum at the root of the weld leading to formation of
EP
less brittle phases. As a result, no cracks are observed at the root. The
AC C
microdilution of Nb at the top of weld just near to Nb-based alloy increases which leads to formation of more brittle phases and microcracks. A narrow zone of 200µm from the weld surface showed microcracks.
3. In the weld, the hardness profile increases from the 321 steel to Nb-1Zr alloy due to increase in niobium content resulting from increase of Fe2Nb and FeNb in weld. 4. In the root of weld, the intensity of beam is enough only for melting a thin layer of Nb alloy due to high melting temperature and high heat conductivity of niobium
ACCEPTED MANUSCRIPT alloy. It leads to less microdilution of Nb to the weld, and consequently less Fe-Nb phase to form. 5. In the as-welded joint, the tensile strength of tested samples was 170 MPa which is 62% of the autogenously welded base metal Nb-1Zr alloy.
RI PT
6. Applying 200µm thickness machining on the surface of weld can give a completely free of crack welds that show strength more than the Nb-alloy base.
SC
7. Finally, it is concluded that a sound Nb-1Zr alloy to 321 stainless steel weld can be achieved by 50-50 offset and Nb microdilution less than 0.6. No interlayer and/or
obtained. Acknowledgment
M AN U
transition piece are required, and appropriate mechanical properties can be
The authors would like to thank Mr Ahmad Reza Mohammad and Mr Seyed Abolfazl
References
TE D
Hosseini for their assistance.
1- Mishra, Brajendra; “Review of Extraction, Processing, Properties & Applications of
EP
Reactive metals”, 2013, john Wiley & Sons.
AC C
2- Jianshu Zheng; Xinmei Hou; “Isothermal oxidation mechanism of a newly developed Nb–Ti–V–Cr–Al–W–Mo–Hf alloy at 800–1200 °C”, Refractory Metals and Hard Materials, 2016, 322-329 .
3- Barry O’Brien; Jon Stinson; “Development of a new niobium-based alloy for vascular stent Applications”, The Mechanical Behavior of Biomedical Materials, 2008, 303-312. 4- Q. Zheng, X. Feng, Y. Shen, et al., Dissimilar friction stir welding of 6061 Al to 316 Stainless steel using Zn as a filler metal, J. Alloys Compd. 686 (2016) 693–701. 5- S. Chen, L. Li, Y. Chen, et al., Joining mechanism of Ti/Al dissimilar alloys during laser
ACCEPTED MANUSCRIPT welding-brazing process, J. Alloys Compd. 509 (3) (2011) 891–898. 6- S. Guo, Q. Zhou, J. Kong, et al., Effect of beam offset on the characteristics of copper/ 304 stainless steel electron beam welding, Vacuum 128 (2016) 205–212. 7- T. Soysal, S. Kou, D. Tat, et al., Macrosegregation in dissimilar-metal fusion welding, Acta
RI PT
Mater. 110 (2016) 149–160.
8- W. Guo, G. You, G. Yuan, et al., Microstructure and mechanical properties of dissimilar
SC
Inertia friction welding of 7A04 aluminum alloy to AZ31 magnesium alloy, J. Alloys Compd. 695 (2017) 3267–3277.
M AN U
9- M.J. Torkamany, F. Malek Ghaini, R. Poursalehi, Dissimilar pulsed Nd:YAG laser Welding of pure niobium to Ti6Al4V, Mater. Des. 53 (2014) 915–920.
10- P. Mastanaiah, G.M. Reddy, K.S. Prasad, C.V.S. Murthy, An investigation on microstructures and mechanical properties of explosive cladded C103 niobium alloy
TE D
over C263 nimonic alloy, J. Mater. Process. Technol. 214 (2014) 2316–2324. 11- Anawa EM, Olabi AG. Control of welding residual stress for dissimilar laser welded materials. J Mater Process Technol 2008;204:22–33.
EP
12- M.S. Węglowski, S. Błacha, A. Phillips, Electron beam welding-techniques and Trends —
AC C
review, Vacuum 130 (2016) 72–92. 13- Study on strengthening mechanism of Ti/Cu electron beam welding. Shun Guo, Qi Zhou, Yong Peng. Materials and Design. 2017. 14- A. Kumar, P. Ganesh, et al., A new vacuum brazing route for niobium-316 L stainless steel transition joints for superconducting RF cavities, J. Mater. Eng. Perform. 24 (2015) 952–963. 15- ASM Hand book, volume 3, alloy phase diagrams.
ACCEPTED MANUSCRIPT 16- Yu. V. Budkin (2011) Welding joints in dissimilar metals, Welding International, 25:7, 523-525. 17- D. Zhao, J. Yan, Y. Liu, Z. Ji, Interfacial structure and mechanical properties of hot-roll bonded joints between titanium alloy and stainless steel using niobium interlayer, Trans.
RI PT
Nonferrous Metals Soc. China 24 (2014) 2839–2844. 18- First Samples of Ti and Nb Tubes Explosion Welding Joint with Stainless Steel for ILC 1.8 K Cryomodule. B. M. Sabirov, J. A. Budagov, and G. D. Shirkov. Physics of Particles and
SC
Nuclei. 2013.
19- Laser welding of niobium to 410 steel with a nickel interlayer produced By electro
M AN U
spark deposition. Seyed Hamzeh Baghjari, Farshid Malek Ghaini. Materials and Design.
TE D
2016.
20- Effect of beam offset on microstructure and mechanical properties of dissimilar electron beam welded high temperature titanium alloys, V. Esfahani Yeganeh, Peijie Li.
EP
Materials and Design. 2017.
AC C
21- W. Liu, G. Chen, B. Zhang, et al., Investigation on process optimization of Cu/Ti electron beam welding, Trans. China Weld. Inst. 5 (2008) 022. 22- G.Q. Chen, B.G. Zhang, L. Wei, et al., Influence of electron-beam superposition welding on intermetallic layer of Cu/Ti joint, Trans. Nonferrous Metals Soc. China 22 (10) (2012) 2416–2420. 23- ASM Hand book, volume 2, properties and selection: nonferrous alloys and specialpurpose materials.
ACCEPTED MANUSCRIPT 24- H. Chen, G. Bi, B.Y. Lee, C.K. Cheng, Laser welding of CP Ti to stainless steel with different temporal pulse shapes, J. Mater. Process. Technol. 231 (2016) 58–65. 25- Grill R, Gnadenberger A. Niobium as mint metal: production–properties– processing. Int J Refract Metal Hard Mater 2006;24:275–82.
RI PT
26- Myneni GR, Umezawa H. Variation of mechanical properties of high RRR and reactor grade niobium with heat treatments. Mater Tech 2003;7–9:19–22.
SC
27- S.Q. Wang, J.H. Liu, D.L. Chen, Tensile and fatigue properties of electron beam welded dissimilar joints between Ti–6Al–4V and BT9 titanium alloys, Mater. Sci. Eng. A 584 (Nov.
AC C
EP
TE D
M AN U
2013) 47–56.
ACCEPTED MANUSCRIPT Figure captions Fig. 1. Binary phase diagram of Fe/Nb [15] Fig. 2. Schematic illustration of the Nb/321 welding set up. Fig. 3. Schematic drawing of the electron beam alignment (BOFs) (a) the welding with 50%
RI PT
of the beam diameter positioned on each side (SN3), (b) the welding with 75% of the beam positioned on the Nb-1Zr side (SN1 and SN2) and (c) the welding with 75% of the beam
SC
positioned on the 321 steel side (SN4). Fig 4. Weld profile of: (a) 50-50 (SN3) and (b) 75-25 (SN2).
M AN U
Fig. 5. XRD results of the dissimilar welding of SN3. (a) 50-50 (b) 75-25
Fig. 6. Cross section of dissimilar electron beam welding of SN3. (a) 321 stainless steel side; (b) Nb-1Zr alloy side.
Fig. 7. Microstructure of offset 75-25 on the 321 steel (SN2).
TE D
Fig. 8. EDS results of the points marked. (a) The SEM image of the SN3 weld. (b) The corresponding EDS analysis of the SN3 weld
Fig. 9. The SEM image and corresponding EDS analysis of the SN3 weld.
EP
Fig. 10. Hardness variations in upper and lower parts of the FZ for SN3 weld.
AC C
Fig. 11. Tensile test of dissimilar welded Nb-1Zr to 321 stainless steel Fig. 12. The sample of tensile test of dissimilar welded Nb-1Zr to 321 steel. The test sample broken from Nb side.
Fig. 13. The fractography and EDS analysis of tensile sample of dissimilar joint.
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
ACCEPTED MANUSCRIPT
Highlights
Dissimilar butt welding of Nb-1Zr to the 321 stainless steel is studied.
•
An electron beam welding equipment is used.
•
The offset and Nb microdilution effects on the weldability are studied.
•
The offset of 50-50 and 13mA, 1343V showed the best weldability.
•
A sound weld can be achieved by 50-50 offset and Nb microdilution less than 0.6.
•
No interlayer and/or transition piece are required.
AC C
EP
TE D
M AN U
SC
RI PT
•
S0042-207X(17)30945-4
DOI:
10.1016/j.vacuum.2017.09.046
Reference:
VAC 7622
To appear in:
Vacuum
Received Date: 16 July 2017 Revised Date:
27 September 2017
Accepted Date: 28 September 2017
Please cite this article as: Hajitabar A, Naffakh-Moosavy H, Electron beam welding of difficult-toweld austenitic stainless steel/Nb-based alloy dissimilar joints without interlayer, Vacuum (2017), doi: 10.1016/j.vacuum.2017.09.046. 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 Electron Beam Welding of Difficult-to-Weld Austenitic Stainless Steel/Nb-based Alloy Dissimilar Joints without Interlayer Ali Hajitabar, Homam Naffakh-Moosavy∗
(TMU), Tehran, Iran Abstract
RI PT
Department of Materials Engineering, PO Box: 14115-143, Tarbiat Modares University
SC
Dissimilar butt welding of Nb-1Zr to the 321 stainless steel was studied using electron beam welding. The aim of this study was investigation of the offset and microdilution
M AN U
effects on the weldability, microstructure and mechanical properties of this dissimilar joint. The offset of 50-50 and welding parameters of 13mA and 1343V showed the best weldability and penetration for the joint. The microdilution of Nb in the weld changes by going far from the base metals. The microdilution of Nb is the minimum at root of the weld
TE D
(~ 0.2) leading to formation of less brittle phases. As a result, no cracks are observed at the root. The microdilution of Nb at the top of weld just near to Nb-based alloy at a 200µm narrow zone under surface increases (~ 0.8) which leads to formation of more brittle
EP
phases and subsequent microcracks. Applying 200µm thickness machining on the surface
AC C
of weld can give a completely free of crack welds that show strength more than the Nballoy base. It is concluded that a sound weld can be achieved by 50-50 offset and Nb microdilution less than 0.6. No interlayer and/or transition piece are required, and appropriate mechanical properties can be obtained.
∗
Corresponding Author: Telefax.:+98 21 82884928; E-mail address: [email protected] (H. Naffakh-Moosavy)
ACCEPTED MANUSCRIPT Keywords: Microdilution; Nb-1Zr Alloy; 321 Stainless Steel; Electron Beam Welding; Weldability.
1. Introduction
RI PT
Niobium is a refractory metal with melting point of 2468 ̊Ϲ [1]. Nb-based alloys have shown excellent properties at high-temperatures, high melting point and appropriate
SC
creep properties [2]. Pure niobium and the Nb–based alloys have been widely used in the nuclear, aerospace industries and biomedical applications [3]. Dissimilar metals welding is
M AN U
an important technology in the modern industry. It is characterized by the ability to maximize the use of a variety of advantages of materials for industrial production [4-8]. Welding of dissimilar alloys is attractive for the industry because of its potential reduction of cost and weight, increasing design flexibility and complexity, and also improving the
TE D
product performance [9, 10]. Problems in dissimilar welding are attributed to the differences in physical and chemical properties between the welding members and residual stresses resulting in the degradation of mechanical properties of welds [11].
EP
Electron beam welding (EBW), with high energy density, large depth to width ratio [6, 12],
AC C
is of high interest for the welding of dissimilar materials [13]. Niobium to stainless steel joints are used in fabrication of superconducting niobium cavities, which are widely used for particle accelerators [14]. But niobium forms brittle intermetallic phases (Fe2Nb and FeNb intermetallics) when alloyed with iron according to the binary Fe-Nb phase diagram (Fig. 1). It is known that weld brazing of Nb to steel has been made [16]. However, a continuous layer of Fe2Nb phase is formed at the boundary between niobium and steel. Zhao used niobium as interlayer in hot-roll bonding of Ti alloy and stainless steel [17].
ACCEPTED MANUSCRIPT First samples of Ti and Nb tubes explosion welding joint with stainless steel has been reported by Sabirov et al. [18]. Laser welding of niobium to 410 steel with a nickel interlayer produced by electro spark deposition has been reported by Baghjari et al. [19]. Offsetting the beam position results in different mechanical properties of welded joint
RI PT
[20].
All studies above proposed complicated and/or costly procedures for dissimilar welding of
SC
such difficult-to-weld joints. Therefore, the less complicated and/ or costly ideas are welcomed. In addition, in most cases the mechanical properties of the obtained welds are
M AN U
considerably lower than the base metals. So, in this study, butt welding of the Nb-1Zr niobium alloy to 321 stainless steel using an electron beam welding is investigated. The research tries to determine the optimum microdilution at which less fraction of brittle intermetallics are formed, and improved weldability is obtained. Effects of various electron
TE D
beam and process parameters on the weld profile regarding large difference in thermophysical properties of the weld counterparts are studied. The endeavor is directed towards establishing optimum process parameters with regard to the weld metal soundness and
EP
the mechanical properties of the joint.
AC C
2. Experimental procedure 2.1. Materials
Nb-1Zr alloy and 321 stainless steel in the form of plates with the dimension of 100 × 50 × 3 mm were used for bead on plate EBW. The chemical composition of base metals are shown in Table 1.
Table 1 Chemical composition (wt%) of base metals
ACCEPTED MANUSCRIPT Nb
Zr
W
Ta
Fe
Ni
Mo
Cr
C
Mn
Si
Nb-1Zr
Balance
0.8
0.6
0.5
0.04
0
0.14
0
0
0
0
AISI 321
0
0
0
0
Balance
10
0.05
18
0.08
2
1
RI PT
Element
Before welding, all specimens were cut with wire cut and then prepared carefully by fine polishing with sandpapers of various grit sizes (600#, 800# and 1000#) and surface
SC
cleaning with acetone in order to prevent weld metal contamination. Specimens were firmly fixed in the form of butt joint (Fig. 2).
M AN U
2.2 Welding methods
Lots of previous studies [6, 7, 21, 22] have shown that non-centered welding was useful for dissimilar metals welding, but it is necessary to be further optimized. Electron beam
TE D
welding process was conducted with welding parameters as shown in Table 2.
Table 2 Electron beam welding fixed parameters for joining Nb-1Zr to 321 steel sheets.
[mm/S]
85
Filament
Focus current
Chamber
current (A)
(mA)
vacuum (m bar)
46
518.8
6×10-4
AC C
10
Voltage (kV)
EP
Welding speed
2.3 Offset and Microdilution control The welded joints were prepared with different beam offsets (BOFs) in order to obtain different dilutions (i.e., overall participation of Nb entering from the base to the entire of weld, WNb,weld/WNb,base) and microdilutions (i.e., local participation of Nb entering from the base to the weld, WNb,weld/WNb,base). Schematic diagram of the welding process were shown in Fig. 3. At first, optimization of welding parameters on each of the base metals were
ACCEPTED MANUSCRIPT analyzed, individually. The optimal parameters such as depth of penetration and weld quality for each sheet of Nb-1Zr alloy and 321 stainless steel were determined. The obtained welding parameters for individual base metals were used to apply to the dissimilar joining (Table 2). The next set of experiments dealt with the dissimilar joining of
RI PT
Nb-1Zr alloy to 321 stainless steel. The dissimilar joining was performed without use of any intermediate layer. For this purpose, the dissimilar joining is designed in such a way that
SC
the spot of the electron beam irradiation is adjusted on the welding interface to obtain the area ratio of 50-50, 25-75 and 75-25 percent (Fig. 3). The set of tests are given in Table 3.
M AN U
These different ratios are lead to different dilution and microdilution levels of Nb in the weld.
Table 3
Electron beam welding variable parameters for joining Nb-1Zr to 321 steel sheets. beam current [mA]
bias voltage [V]
SN1
24
1130
electron beam alignment (BOFs) [%] 75 Nb alloy 25 321
13
1312
75 Nb alloy – 25 321
13
1343
50-50
13
1388
25 Nb alloy – 75 321
SN3
AC C
SN4
EP
SN2
TE D
sample
2.4 Characterization methods To study the weld profile and details of melted zones on both sides of the weld line, the welded samples were sectioned transversely, polished and etched for metallographic examinations. Marble etchant (50 ml H2O, 50 ml HCL and 10 gr CuSO4) was used for the etching of dissimilar electron beam welded samples.
ACCEPTED MANUSCRIPT Microstructural investigations were performed through an Olympus BX51M optical microscope and a Philips XL30 scanning electron Microscope (SEM) equipped with an energy-dispersive X-ray spectrometry (EDS) system for spot and line weight analysis and fractography, and a Philips X’Pert MPD X-ray diffraction analysis were applied for the
RI PT
phase analysis of samples. Micro Vickers hardness of the joint was measured using a micro hardness device with 100 g load level and 15s loading time. Tensile tests were conducted according to ASTM E8M by a Santam STM-50 universal testing machine at room
3. Results and discussions
M AN U
were tested and the average value was recorded.
SC
temperature. The strain rate was set at 10-3 s-1. Three tensile samples of each kind of joints
3.1. Appearance and soundness of welds
According to Table 3, the beam offset on the joint interface for samples are different.
TE D
Other fundamental parameters were fixed in accordance with Table 2. According to Table 3, for sample SN1 in which 75% of beam was concentrated on niobium based alloy, the 321 stainless steel was superheated and the joint was failed due to applying high heat
EP
input to it (current 24 mA). Sample SN4 that 75% of electron beam is focused on 321
AC C
stainless steel, the niobium based alloy was not melted and the joint was failed due to applying low heat input. According to Table 4, the failed welding joints, can be attributed to the remarkable difference in melting temperature of the two sides of the joint (2407°C for niobium alloy and about 1400°C for 321 steel), as well as high thermal conductivity of niobium alloy in comparison to 321 stainless steel (about three times). Accordingly, niobium alloys needs to heat input greater than 321 stainless steel to melt due to high thermal conductivity and high melting temperature.
ACCEPTED MANUSCRIPT The cross-section of the SN2 and SN3 samples of Nb-1Zr alloy to 321 stainless steel dissimilar welds can be seen in figures 4 (a) and (b), that have been welded by the parameters shown in Tables 2 and 3. Fig. 4a shows the cross-section of the dissimilar welding of sample SN3. In this joint, the electron beam is applied to engage each base
RI PT
metals equally (i.e., 50-50 offset). It is seen that full depth of penetration is obtained and heat input has been able to make a complete weld metal. According to Table 4, because of
SC
high melting point and high thermal conductivity of niobium than 321 stainless steel, molten pool is closer to the 321 steel, and steel side shows more participation in producing
M AN U
the weld metal (weld pool asymmetric).
Table 4 Comparison of thermo-physical properties of Nb-1Zr and 321 steel [23]. Melting
Density
Specific
Thermal
Coefficient
point
(g/cm3)
heat
conductivity
of linear
(J/g K)
(W/m k)
thermal
at 20 °C
at 25 °C
expansion
TE D
Material
2407
AC C
Nb-1Zr
EP
(°C)
321 steel
1375-
(µm/m K)
8.59
0.27
41.9
7.54
8
0.5
16.3
16.6
1400
Figure 4b shows a cross-section of the dissimilar welding of sample SN2. In this dissimilar joint, 75% of the electron beam is focused on niobium alloy. In this dissimilar joint, the penetration depth is incomplete due to the low heat input. In both images a and b of Fig.
ACCEPTED MANUSCRIPT 4, heat-affected zone (HAZ) in niobium alloy is much broader than the 321 steel due to the high thermal conductivity of niobium alloy. 3.2. Microstructural examinations and Weldability According to the binary phase diagram of Fe/Nb in Fig. 1, two types of intermetallic
RI PT
compounds including FeNb and Fe2Nb can be formed. Also, in certain composition of Fe and Nb, eutectic structure can be formed. So, weld metal can include brittle intermetallic
SC
compounds, solid solution, and eutectic structures. To determine the phases in the welds, the XRD analysis was used and the results are shown in Fig. 5. These results suggest that
M AN U
weld metal contains austenite solid solution of iron, niobium, chromium, nickel, and intermetallic compounds of Fe2Nb and maybe FeNb.
SEM image of the samples SN3 (50-50 offset) and SN2 (75 Nb alloy – 25 321 offset) are presented in figures 6 and 7. Three different regions are observed. According to the binary
TE D
phase diagram of the Fe/Nb (Fig. 1), the results of XRD analysis in Fig. 5, as well as EDS analysis points in these regions in Fig. 8, it can be predicted that the region near the base metal 321, is two-phases region includes a proeutectic phase enriched in iron and an
EP
eutectic phase includes γ-Fe and intermetallic compounds of Fe2Nb. In a region away
AC C
enough from the base metal (central area), a eutectic region consisting of Fe-rich and Fe2Nb phases are formed. The third region includes a proeutectic phase rich in niobium and a eutectic phase composed of Fe–rich phase and Fe2Nb. In Fig. 6, the brighter regions represent locations containing more niobium, and mainly are intermetallic compounds of Fe2Nb. Formation of the brittle phases and exertion of thermal stresses originating from thermal shrinkages in fusion welding, can lead to immediate cracking of the welded joint [24]. According to Fig. 6b, microcracks in the white regions can be seen. The cracks show brittleness of these areas. As noted above, these regions contain large amounts of FeNb
ACCEPTED MANUSCRIPT and/or Fe2Nb brittle intermetallic compounds. The microcracks caused by stresses arisen from contractions of solidification and cooling. The microdilution of Nb in the weld changes by going far from the base metals especially Nb-based alloy. The microdilution of Nb is the minimum at the root of the weld (~0.2)
RI PT
leading to formation of less brittle phases. As a result, no cracks are observed at the root. The microdilution of Nb at the top of weld just near to Nb-based alloy at a 200µm narrow
SC
zone under surface increases (~0.8) which leads to formation of more brittle phases and subsequent microcracks. This high microdilution of Nb at the top of weld is attributed to
M AN U
wide upper geometry of key hole, which causes more portion of Nb-based alloy to melt and enter to the weld. In the root of weld, the intensity of beam is enough for melting a thin layer of Nb alloy only. It leads to less dilution of Nb to the weld, and consequently less Fe-Nb phase to form.
TE D
Fig. 8 illustrates the EDS analysis of iron and niobium changes with the distance from the 321 stainless steel base. It shows that by moving from the steel base metal to the center of weld, iron content decreases and the amount of niobium (i.e., microdilution of Nb)
EP
increases, respectively. Fig. 9 shows the increase in the atomic percent of niobium and a
AC C
decrease in atomic percent of 321 steel elements (i.e., iron, chromium and nickel) by moving from bottom of the weld to the top. It is known that the electron beam power density decreases by penetrating inside the keyhole depth. Thus, it seems that in the deeper parts of the weld the beam intensity is enough for melting of a thin layer of Nb-1Zr alloy, only. The above mentioned experiments show that at 50% offset, a complete weld penetration for both 321 alloy and Nb-1Zr alloy is obtained. Fig. 9 shows that a thin layer of Nb-1Zr base metal is molten. According to EDS analysis at point 9, it is observed that amount of
ACCEPTED MANUSCRIPT niobium in this area is significantly higher than other regions of the weld, that is more microdilution occurred. According to the Nb microdilution of this point (~0.8) and the Fe/Nb binary phase diagram in Fig. 1, it can be predicted that FeNb intermetallic compound is formed at this region (Fig. 6b).
RI PT
3.3. Evaluation of mechanical properties
Among the above mentioned experiments the best process conditions with regard to weld
SC
penetration and soundness i.e., SN3 was selected to evaluate the mechanical properties of Nb-1Zr alloy/321 stainless steel dissimilar weld. The hardness profile of the weld is shown
M AN U
in Fig. 10 in two different parallel lines, on top, and root weld – at the bottom of the key hole. On the basis of Fig. 10, it is observed that hardness of the base metal, Nb-1Zr is less than the 321 stainless steel. Although the melting point of Niobium is high, but it has low hardness and it is known as a soft metal [25]. Impurities, rolling and annealing processes
TE D
will greatly affect the mechanical properties of niobium plates, especially its tensile strength. Myneni and coworkers have studied the mechanical properties of niobium with different levels of purity and production processes [26].
EP
The results of micro hardness show that hardness in the weld metal is more than Nb-1Zr
AC C
and 321 stainless steel base metals. This increase of hardness is attributed to strengthening by solid solution and formation of Fe2Nb, FeNb and eutectics in the weld. In the weld, the hardness profile increases from the 321 steel to Nb-1Zr alloy due to the increase in niobium content. The hardness reaches its maximum value (766 HV) at region just near to the Nb-based alloy where is related to formation of FeNb at point 9 in Fig. 9. The hardness profile of root weld can also be observed in Fig. 10 with a red line. At this profile, the hardness in the weld is considerably less than hardness profile of the top part of weld. As mentioned, the power density of electron beam decreases as penetration
ACCEPTED MANUSCRIPT increases in the depth of weld, and as a result, in the deeper parts of weld, less amount of Nb-based alloy is molten and participates in the weld i.e., less microdilution of Nb-based alloy occurs. This is the key controlling factor for minimization of brittle intermetallics to form in the weld. That is, by decreasing the melting and participation of the Nb-based alloy
RI PT
in the weld, a sound and crack-free weld can be achieved. This decrease in melting confirms the EDS results in Fig. 9. EDS analysis in Fig. 9 shows that amount of niobium
SC
considerably decreases in the depth of weld. This decrease in niobium causes decrease in formation of hard Fe2Nb and FeNb phases in the weld and as a result, the hardness
M AN U
decreases. Wang et al. [27] found that there were two hardness sub-regions in the FZ that were due to different chemistry of base metals.
The tensile test from dissimilar welded samples is performed for investigating the strength of the joint. At first, the tensile strength of base metals is measured, where for Nb-1Zr
TE D
alloy and 321 stainless steel is 525 MPa and 657 MPa, respectively. For dissimilar joint tensile test, three samples were prepared and tested. The stress-strain curves for the dissimilar joints are shown in Fig. 11. The samples are broken at the top of weld metal just
EP
near to the niobium side, where the high microdilution for Nb (~80) is recorded (Fig. 12).
AC C
The microcracks are formed in a 200µm length zone due to formation of high amount of Fe-Nb phase. As it was shown in the micro- hardness test, the maximum value for hardness in the weld was at regions near to the Nb-1Zr alloy, which showed the considerable amount of FeNb intermetallic. The tensile strength of tested samples was 170 MPa which is 62% of the autogenously welded base metal Nb-1Zr alloy obtained without any additional operations such as mechanical grinding of surface. Also, the elongation of 1.2% is obtained for the dissimilar joint. By removing this 200µm narrow zone by mechanical
ACCEPTED MANUSCRIPT grinding, the higher strength and elongation in comparison to the Nb-based alloy can be obtained. The fractography of tensile sample of dissimilar weld is shown in Fig. 13. It indicates that fracture of joints occurs in brittle failure mode by the obvious cleavage steps. EDS analysis
RI PT
of the fracture surface reveals presence of high amount of niobium, which can be attributed to brittle intermetallic compounds of FeNb, just close to Nb -1Zr alloy base
SC
metal. Conclusions
M AN U
Dissimilar welding of Nb-1Zr alloy to 321 stainless steel was studied using an electron beam welding for the first time. The high lights below are addressed: 1. Multiple offsets and heat inputs are investigated. The offset of 50-50 and heat input giving complete penetration for 321 stainless steel (i.e., 13mA, 1343V)
TE D
showed the best weldability and penetration for the dissimilar joint. 2. The microdilution of Nb in the weld changes by going far from the base metals. The microdilution of Nb is the minimum at the root of the weld leading to formation of
EP
less brittle phases. As a result, no cracks are observed at the root. The
AC C
microdilution of Nb at the top of weld just near to Nb-based alloy increases which leads to formation of more brittle phases and microcracks. A narrow zone of 200µm from the weld surface showed microcracks.
3. In the weld, the hardness profile increases from the 321 steel to Nb-1Zr alloy due to increase in niobium content resulting from increase of Fe2Nb and FeNb in weld. 4. In the root of weld, the intensity of beam is enough only for melting a thin layer of Nb alloy due to high melting temperature and high heat conductivity of niobium
ACCEPTED MANUSCRIPT alloy. It leads to less microdilution of Nb to the weld, and consequently less Fe-Nb phase to form. 5. In the as-welded joint, the tensile strength of tested samples was 170 MPa which is 62% of the autogenously welded base metal Nb-1Zr alloy.
RI PT
6. Applying 200µm thickness machining on the surface of weld can give a completely free of crack welds that show strength more than the Nb-alloy base.
SC
7. Finally, it is concluded that a sound Nb-1Zr alloy to 321 stainless steel weld can be achieved by 50-50 offset and Nb microdilution less than 0.6. No interlayer and/or
obtained. Acknowledgment
M AN U
transition piece are required, and appropriate mechanical properties can be
The authors would like to thank Mr Ahmad Reza Mohammad and Mr Seyed Abolfazl
References
TE D
Hosseini for their assistance.
1- Mishra, Brajendra; “Review of Extraction, Processing, Properties & Applications of
EP
Reactive metals”, 2013, john Wiley & Sons.
AC C
2- Jianshu Zheng; Xinmei Hou; “Isothermal oxidation mechanism of a newly developed Nb–Ti–V–Cr–Al–W–Mo–Hf alloy at 800–1200 °C”, Refractory Metals and Hard Materials, 2016, 322-329 .
3- Barry O’Brien; Jon Stinson; “Development of a new niobium-based alloy for vascular stent Applications”, The Mechanical Behavior of Biomedical Materials, 2008, 303-312. 4- Q. Zheng, X. Feng, Y. Shen, et al., Dissimilar friction stir welding of 6061 Al to 316 Stainless steel using Zn as a filler metal, J. Alloys Compd. 686 (2016) 693–701. 5- S. Chen, L. Li, Y. Chen, et al., Joining mechanism of Ti/Al dissimilar alloys during laser
ACCEPTED MANUSCRIPT welding-brazing process, J. Alloys Compd. 509 (3) (2011) 891–898. 6- S. Guo, Q. Zhou, J. Kong, et al., Effect of beam offset on the characteristics of copper/ 304 stainless steel electron beam welding, Vacuum 128 (2016) 205–212. 7- T. Soysal, S. Kou, D. Tat, et al., Macrosegregation in dissimilar-metal fusion welding, Acta
RI PT
Mater. 110 (2016) 149–160.
8- W. Guo, G. You, G. Yuan, et al., Microstructure and mechanical properties of dissimilar
SC
Inertia friction welding of 7A04 aluminum alloy to AZ31 magnesium alloy, J. Alloys Compd. 695 (2017) 3267–3277.
M AN U
9- M.J. Torkamany, F. Malek Ghaini, R. Poursalehi, Dissimilar pulsed Nd:YAG laser Welding of pure niobium to Ti6Al4V, Mater. Des. 53 (2014) 915–920.
10- P. Mastanaiah, G.M. Reddy, K.S. Prasad, C.V.S. Murthy, An investigation on microstructures and mechanical properties of explosive cladded C103 niobium alloy
TE D
over C263 nimonic alloy, J. Mater. Process. Technol. 214 (2014) 2316–2324. 11- Anawa EM, Olabi AG. Control of welding residual stress for dissimilar laser welded materials. J Mater Process Technol 2008;204:22–33.
EP
12- M.S. Węglowski, S. Błacha, A. Phillips, Electron beam welding-techniques and Trends —
AC C
review, Vacuum 130 (2016) 72–92. 13- Study on strengthening mechanism of Ti/Cu electron beam welding. Shun Guo, Qi Zhou, Yong Peng. Materials and Design. 2017. 14- A. Kumar, P. Ganesh, et al., A new vacuum brazing route for niobium-316 L stainless steel transition joints for superconducting RF cavities, J. Mater. Eng. Perform. 24 (2015) 952–963. 15- ASM Hand book, volume 3, alloy phase diagrams.
ACCEPTED MANUSCRIPT 16- Yu. V. Budkin (2011) Welding joints in dissimilar metals, Welding International, 25:7, 523-525. 17- D. Zhao, J. Yan, Y. Liu, Z. Ji, Interfacial structure and mechanical properties of hot-roll bonded joints between titanium alloy and stainless steel using niobium interlayer, Trans.
RI PT
Nonferrous Metals Soc. China 24 (2014) 2839–2844. 18- First Samples of Ti and Nb Tubes Explosion Welding Joint with Stainless Steel for ILC 1.8 K Cryomodule. B. M. Sabirov, J. A. Budagov, and G. D. Shirkov. Physics of Particles and
SC
Nuclei. 2013.
19- Laser welding of niobium to 410 steel with a nickel interlayer produced By electro
M AN U
spark deposition. Seyed Hamzeh Baghjari, Farshid Malek Ghaini. Materials and Design.
TE D
2016.
20- Effect of beam offset on microstructure and mechanical properties of dissimilar electron beam welded high temperature titanium alloys, V. Esfahani Yeganeh, Peijie Li.
EP
Materials and Design. 2017.
AC C
21- W. Liu, G. Chen, B. Zhang, et al., Investigation on process optimization of Cu/Ti electron beam welding, Trans. China Weld. Inst. 5 (2008) 022. 22- G.Q. Chen, B.G. Zhang, L. Wei, et al., Influence of electron-beam superposition welding on intermetallic layer of Cu/Ti joint, Trans. Nonferrous Metals Soc. China 22 (10) (2012) 2416–2420. 23- ASM Hand book, volume 2, properties and selection: nonferrous alloys and specialpurpose materials.
ACCEPTED MANUSCRIPT 24- H. Chen, G. Bi, B.Y. Lee, C.K. Cheng, Laser welding of CP Ti to stainless steel with different temporal pulse shapes, J. Mater. Process. Technol. 231 (2016) 58–65. 25- Grill R, Gnadenberger A. Niobium as mint metal: production–properties– processing. Int J Refract Metal Hard Mater 2006;24:275–82.
RI PT
26- Myneni GR, Umezawa H. Variation of mechanical properties of high RRR and reactor grade niobium with heat treatments. Mater Tech 2003;7–9:19–22.
SC
27- S.Q. Wang, J.H. Liu, D.L. Chen, Tensile and fatigue properties of electron beam welded dissimilar joints between Ti–6Al–4V and BT9 titanium alloys, Mater. Sci. Eng. A 584 (Nov.
AC C
EP
TE D
M AN U
2013) 47–56.
ACCEPTED MANUSCRIPT Figure captions Fig. 1. Binary phase diagram of Fe/Nb [15] Fig. 2. Schematic illustration of the Nb/321 welding set up. Fig. 3. Schematic drawing of the electron beam alignment (BOFs) (a) the welding with 50%
RI PT
of the beam diameter positioned on each side (SN3), (b) the welding with 75% of the beam positioned on the Nb-1Zr side (SN1 and SN2) and (c) the welding with 75% of the beam
SC
positioned on the 321 steel side (SN4). Fig 4. Weld profile of: (a) 50-50 (SN3) and (b) 75-25 (SN2).
M AN U
Fig. 5. XRD results of the dissimilar welding of SN3. (a) 50-50 (b) 75-25
Fig. 6. Cross section of dissimilar electron beam welding of SN3. (a) 321 stainless steel side; (b) Nb-1Zr alloy side.
Fig. 7. Microstructure of offset 75-25 on the 321 steel (SN2).
TE D
Fig. 8. EDS results of the points marked. (a) The SEM image of the SN3 weld. (b) The corresponding EDS analysis of the SN3 weld
Fig. 9. The SEM image and corresponding EDS analysis of the SN3 weld.
EP
Fig. 10. Hardness variations in upper and lower parts of the FZ for SN3 weld.
AC C
Fig. 11. Tensile test of dissimilar welded Nb-1Zr to 321 stainless steel Fig. 12. The sample of tensile test of dissimilar welded Nb-1Zr to 321 steel. The test sample broken from Nb side.
Fig. 13. The fractography and EDS analysis of tensile sample of dissimilar joint.
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
ACCEPTED MANUSCRIPT
Highlights
Dissimilar butt welding of Nb-1Zr to the 321 stainless steel is studied.
•
An electron beam welding equipment is used.
•
The offset and Nb microdilution effects on the weldability are studied.
•
The offset of 50-50 and 13mA, 1343V showed the best weldability.
•
A sound weld can be achieved by 50-50 offset and Nb microdilution less than 0.6.
•
No interlayer and/or transition piece are required.
AC C
EP
TE D
M AN U
SC
RI PT
•