Vacuum: Ali Hajitabar, Homam Naffakh-Moosavy
Vacuum: Ali Hajitabar, Homam Naffakh-Moosavy
Vacuum
journal homepage: www.elsevier.com/locate/vacuum
a r t i c l e i n f o a b s t r a c t
Article history: The effect of electron beam welding current changes on the microstructure and mechanical properties of
Received 9 January 2018 the Nb-based alloy has been investigated. The electron beam welding was applied with 4 different values
Received in revised form for electric current 20, 24, 30 and 35 mA on 3 mm thick plates. The aspects including different welding
24 January 2018
regions, geometry and depth of welding penetration, as well as the effect of heat input on the mechanical
Accepted 24 January 2018
properties are investigated. The mechanical properties including tensile and microhardness values of the
weld was measured. The results show that in a sample with a 30 mA welding current, the optimum
conditions for the depth of penetration, microstructure, mechanical properties and the geometry of the
Keywords:
Electron beam welding
weld are obtained. The welds showed a cellular structure, and intercellular dendrites in the central re-
Advanced alloy Nb-1Zr gion of the weld have been caused due to microsegregations created between the cells. In heat-affected
Welding current zone (HAZ), severe recrystallization and grain growth has occurred. Because of the high thermal con-
Microstructure ductivity of niobium, the HAZ size is relatively large. Based on the 3D Rosenthal's equation, the
Mechanical properties recrystallization temperature of alloy was calculated as 713 C. It is observed that as G R increases, the
grain size in the central line of the weld decreases. The hardness profile shows that the hardness of the
weld zone and the HAZ is significantly less than that of the base metal due to elimination of work
hardening effect. The tensile strength of the weld for a sample with a current of 30 mA was 281 MPa,
which is 53% of the tensile strength of the base metal and the weld was broken in the HAZ.
© 2018 Elsevier Ltd. All rights reserved.
https://doi.org/10.1016/j.vacuum.2018.01.043
0042-207X/© 2018 Elsevier Ltd. All rights reserved.
A. Hajitabar, H. Naffakh-Moosavy / Vacuum 150 (2018) 196e202 197
aerospace and mechanical engineering [17]. The study of the 6 104 mbar has been used to produce bead-on-plate welds. The
physical and chemical properties of electron beam welding on re- EBW equipment is an assembled product which is made by Electron
fractory metals was carried out by Budkin [18]. He studied the Beam Application Development (EBAD) Company at 2014 a.m. in
weldability of these metals under different welding conditions and Iran. Samples were placed in a vacuum chamber after cleaning, and
studied the dependence of the thickness of the intermetallic then a vacuum of 104 mbar was applied.
compounds created on the welding strength [18]. Budkin also The parameters such as welding speed, current and voltage can
developed the physical-mathematical model of the electron beam affect the welding conditions. However, the previous studies
welding process for refractory metal welding to steel [19]. showed that welding current has more effect on the penetration of
Demyanov and Kaniukov considered the superconducting proper- weld, which is the most important aim of the present work. The
ties of ultra-pure electron beam welding [16]. They found that the applied welding current was in the range of 5e50 mA, at fixed
superconducting properties of the welded specimens were voltage and welding speed. This current range was selected on the
reduced, which is due to an increase in the concentration of im- basis of preliminary tests for start the work, and data extracted
purities in the melting area. M.J.Torkamany et al. also studied the from the related literature conducted on the stainless steels and
properties of laser welding of pure niobium with two conduction nickel-based superalloys. Finally, the applied welding currents
and keyhole modes. They also paid attention to a dissimilar welding which showed more meaningful and different results were pre-
of niobium to Ti-6Al-4V alloy [20]. sented in Table 2.
By reviewing the references in this field, it has been concluded According to Table 2, welding was carried out with four different
that limited research on the welding of niobium alloys have been values for current intensity of 20, 24, 30 and 35 mA on the samples.
conducted. In spite of Nb-1Zr alloy important applications in Other parameters such as voltage, filament current, focusing cur-
advanced industries such as aerospace, nuclear and military, very rent, welding speed and focusing current on the surface were
few articles and research reports have been published on its elec- considered constant. The electron beam focus was considered on
tron beam welding so far. In this study, it is tried to study the effect the surface due to the low thickness of the specimens. In high-
of electron beam welding current on microstructure, weld pene- thickness specimens, the electron beam focus is applied inside
tration and mechanical properties of Nb-1Zr alloy. the sample.
The current manuscript provides a literature review on the
welding of Nb alloys in Introduction. The materials and electron 2.3. Metallographic examination
beam welding parameters used are presented in Experimental
procedures, which include the base metal, the process parameters, To study the effect of current on the microstructure and size
metallographic examination, mechanical properties and the cal- grading of different welding regions and changes in the geometry
culations. The results and discussion section includes the base of the weld, the welded samples were cut from the cross section
metal evaluation, the weld sections, the results of calculations, and containing all of the weld regions. The samples were prepared on
the mechanical properties. At the end, the most important results the basis of standard metallographic procedure. Samples were
are summarized as Conclusions. etched with a solution containing 4HF þ 3H2SO4 þ 1HNO3 þ 5H2O
at the optimum time of 45 s. The samples were first examined for
2. Experimental procedures the geometry and depth of penetration of the welds by the
OLYMPUS SZX2-ILLB Stereo Microscope, and analyzed for more
2.1. The base metal precision by optical microscope model BX51M manufactured by
OLYMPUS company and scanning electron microscope manufac-
In this study, a 3 mm sheet of advanced niobium alloy (Nb-1Zr) tured by Philips Model XL30.
was used. To obtain the exact chemical composition of the alloy, an
emission spectrometry method (EOS) was used; the results are 2.4. Mechanical properties
given in Table 1. XRD analysis was used to determine the phases in
this alloy, the XRD pattern is shown in Fig. 1. To do this, the Philips In order to investigate and compare the tensile properties of
X'Pert MPD software is used. electron beam welded Nb-1Zr alloy with base metal, the welded
In order to investigate the tensile properties of Nb-1Zr alloy, the specimens were prepared according to the ASTM E8/E8M standard,
sheets were cut by wire cut with smooth edges and prepared ac- which is shown in Fig. 2. A tensile load of 15 ton and single axis of
cording to ASTM.E8/E8M standard, then tested by SANTAM single- French MTS model at a strain rate of 0.001 S-1 was used. Also, the
axis tensioner, at strain rate 0.001 S-1, at room temperature. The microhardness of the different weld regions was measured by the
hardness of these two alloys was also measured by the Vickers Buehler LTD for all of the welded samples and compared to each
hardness test method. other.
The Nb-1Zr sheets were first cut into samples with dimensions 3.1. The base metal
of 50 mm 15 mm by cutting machine. Then, they were cleaned
by sandpaper, and they were placed in acetone solution for 5s. An According to the EOS analysis, it is observed that all the alloying
electron beam welding (EBW) equipment with a maximum voltage elements present in this material are less than 1 wt%. Referring to
of 85 kV, a maximum power of 20 kW and a vacuum level of the XRD pattern of this alloy in Fig. 1, it is seen that this material is
Table 1
Chemical composition of Nb-1Zr alloy.
Element H Ni N O C Fe Hf Mo Ta W Zr Nb
Wt.% 0.001 0.002 0.006 0.002 0.008 0.04 0.05 0.14 0.5 0.6 0.8 98.5
198 A. Hajitabar, H. Naffakh-Moosavy / Vacuum 150 (2018) 196e202
Table 2
Electron beam welding parameters for each sample.
N1 20 mA 85 KV 10 mm/s
N2 24 mA 85 KV 10 mm/s
N3 30 mA 85 KV 10 mm/s
N4 35 mA 85 KV 10 mm/s
metal, HAZ and base metal for different specimens are provided. As
Fig. 2. Dimensions of tension samples according to ASTM.E8/E8M standard.
also seen in Fig. 4 (a,c), the base metal in vicinity of HAZ exhibits an
elongated structure, which is a result of previous cold working of
essentially a single phase. A low shift of peaks of the graph to the the sheets. Even SEM (Fig. 3) could not precisely show the grain
right, that is, the increase of the 2q diffraction angle is seen in boundaries due to severe plastic deformation of structure, but the
comparison to angles of the pure niobium pattern. According to the elongated grains are observed.
Bragg relation (Equation (1)), In all samples, there is no distinct boundary between the weld
metal and the base metal. The reason for this phenomenon is that
l ¼ 2dsinq (1) no filler metal is used and no difference in the composition of weld
metal and base metal is exist. Also, in all samples, the growth of
where l is the wavelength of the X-ray, d is the distance between solidification grains in the fusion zone is perpendicular to the
the atomic planes and q is the diffraction angle, with the increase of fusion boundary.
the Bragg angle, the distance between the crystal planes and, Since the growth direction of the fusion grains is perpendicular
consequently, the lattice parameter decreased. Therefore, it can be to the fusion line, the solidification structure of the weld metal is
interpreted that the alloying elements present in this material are formed as columnar grains. In the central line of the weld, equiaxed
dissolved to form a substitutional or interstitial (according to the grains are formed due to heat transfer in all directions.
type of alloying element) solid solution in the niobium phase, and In Fig. 4, the recrystallization and growth of the grains in HAZ is
reduce the size of the lattice parameter. observed for all specimens. In HAZ, the new grains have become
The microstructure of Nb-1Zr alloy sheet is shown in Fig. 3. As thicker as passing from the base metal to the fusion line. This grain
can be seen, the microstructure is consists of grains elongated in size gradient from the base metal to the molten metal can be seen
the rolling direction. The tensile strength and the hardness of alloy in Fig. 5a. In areas near to the fusion boundary, there is ample op-
were 526.5 MPa and 167 HV, respectively. These results show that portunity for the new grains to join to each other, and thus the
niobium has a relatively low hardness and tensile strength. coarse grain areas are formed. It is due to higher HAZ temperatures.
Although the room temperature tensile strength of the Nb-1Zr alloy In areas far from the fusion boundary, the conditions for the growth
is relatively low, the alloy keeps its strength at high temperatures. and movement of the boundaries are not easily feasible due to the
lower temperature; thus formation of smaller grains has occurred.
In Fig. 5b, the different regions of the formation of columnar grains
3.2. The weld section
and coaxial grains can be observed. On the border between the
fusion zone and the base metal, the grains grow in the form of
In Fig. 4, the macrostructure of weld regions including the weld
A. Hajitabar, H. Naffakh-Moosavy / Vacuum 150 (2018) 196e202 199
Fig. 4. Macrostructure of weld metal, HAZ and base metal: a) Sample N1 with a current of 20 mA, b) Sample N2 with a current of 24 mA, c) Sample N3 with a current of 30 m, and d)
Sample N4 with a current of 35 mA.
Fig. 5. Macrostructure of electron beam welding regions of sample N2 by optical microscope: a) Size grading of different welding regions, b) Columnar and axial grains in the weld.
Q ¼ h VI=S (2) Weld width and penetration depth for each sample were
measured by image J software, which are shown in Table 3. Ac-
where Q is the heat input, V is the welding voltage, I is the beam
cording to Table 3, it is found that N1 and N2 samples have no full
current, S is the welding speed and h is a constant. The amount of h
penetration due to lower welding current (less heat input), and
for electron beam welding is 0.9 [21].
200 A. Hajitabar, H. Naffakh-Moosavy / Vacuum 150 (2018) 196e202
Table 3
Geometric properties of welded Nb-1Zr alloy.
Sample number Current heat input (KJ/m) Welding width Welding depth Depth to width ratio
(mA) (mm) (mm)
only N3 and N4 have complete penetration. The ratio of the depth According to the three dimension Rosenthal's equation, the
to width of N3 sample is greater than the N4 sample. So, the heat thermal gradient G at the center line of the weld is obtained from
input of the N3 sample is more optimal than other samples. equation (4) [21]:
Fig. 7 shows the scanning electron microscope image of the
weld center. In this image, the cellular solidification within the 2pkðTL T0 Þ2
G¼ (4)
grains is observed. The intercellular regions show formation of very Q
fine dendrites, which can be caused by microsegregations created
between cell boundaries. where G is the thermal gradient, TL is the melting temperature, T0 is
The size of HAZ region for all samples was measured by the the initial temperature of the base metal, Q is the heat input and k is
image J software, and the average HAZ size for each sample is thermal conductivity.
shown in Fig. 8. Because of high thermal conductivity of the Nb-1Zr The temperature gradient value in the center line of the weld
alloy, the size of the HAZ is considerably large [22]. As shown in was calculated for all welded samples, as shown in Table 5. At the
Fig. 8, the size of the HAZ increases as the welding heat input in the center line of weld, the growth rate R is assumed to be equal to the
samples increases. welding speed V, which is 10 mm/s for all of welding samples. In
Based on the three dimension Rosenthal's equation, the general, the G R value determines the coarseness and size of the
recrystallization temperature of this alloy can be obtained in elec- welding structure. So, with applying a higher G R, finer grains are
tron beam welding. The Rosenthal's equation is used. The three obtained. As shown in Table 5, the samples N1and N4 have the
dimension Rosenthal's equation is below [21]: maximum and minimum G R values, respectively. By comparing
the values of G R in the weld center line in Table 5, and the
2pðT T ÞKr VðR xÞ average grain size in the weld center in Fig. 9, it can be found that
¼ exp (3) N1 sample has the maximum G R value and also has the smallest
Q 2a
grain size among other samples. N4 sample with the smallest
where T is the initial temperature of the base metal, T is temper- amount of G R has the largest grain size among other welded
ature, Q is the heat input (Q ¼ h VI), k is the thermal conductivity of samples. According to the results, larger G R in the weld center
the base metal, a is The thermal permeability of the base metal, R is line leads to decrease in the grain size.
the radial distance of the welding center and V is the welding
speed. By solving the Rosenthal's equation using thermal properties 3.3. Mechanical properties of the weld
provided for the Nb-1Zr alloy in Table 4, the recrystallization
temperature of alloy is obtained. The recrystallization temperature Fig. 10 shows the profile of hardness variations. Weld hardness
has been calculated for electron beam welding to be 713 C. variations were measured by Vickers micro hardness using a load of
According to Fig. 9, the mean grain size in the center line of the 200 g and duration of 15 s in different regions of weld cross section
weld was measured and plotted versus the heat input applied for all for all of the welded samples. According to this diagram, it is
samples. Accordingly, for each sample, the diameters of eleven observed that the hardness of the weld region and the HAZ of the
grains in the central zone of weld were measured, and the average Nb-1Zr alloy are significantly less than the base metal. The base
size was calculated. It is seen that with increasing the heat input in metal is a worked (rolled) alloy with very fine and elongated grains.
the welded samples, the average grain size increases in the central As shown in Fig. 5, the grain size increases with a steep slope, from
line of the weld. the base metal to the HAZ and the weld. This sharp increase in the
grain size from the base metal to the weld zone caused by the in-
crease the temperature, leads to a significant reduction in the
hardness of the base metal to the HAZ and the weld. Recrystalli-
zation and formation of new dislocation-free grains in HAZ, and
elimination of strong cold worked structure at the fusion zone lead
to softer regions in comparison to work hardened base metal. So,
the least hardness is obtained at the weld center about 98 H V.
The tensile test was measured for N3 samples at room tem-
perature. The tensile strength of this welding sample was 281 MPa,
which is 53% of the tensile strength of the base metal (strength of
base metal is 526.5 MPa). This weld was broken from the HAZ re-
gion. Due to the severe growth of the grains in the HAZ region, the
strength of this area has decreased. Also, according to the hardness
profile in Fig. 8, the hardness of the base metal to the weld metal
decreases with a relatively large gradient, and then continues with
a gentle gradient. A significant change in the gradient in the HAZ,
due to the sharp increase in grain size in this area, has caused a
Fig. 7. Scanning electron microscopy from the Nb-1Zr alloy welding center. failure in the HAZ.
A. Hajitabar, H. Naffakh-Moosavy / Vacuum 150 (2018) 196e202 201
Fig. 8. Comparison of the average size of the HAZ and applied heat inputs for the welding samples.
Table 4
Parameters and thermal properties of Nb-1Zr alloy [23].
a(mm
2
Name of the alloy Initial temperature Melting point Thermal conductivity s ) Thermal permeability
T0 ( C) T L ( C) k (msJ C )
Fig. 9. Comparison of the average grain size in the central line of the welding and the applied heat inputs for the welding samples.
Table 5
Calculated values for thermal gradient G, growth rate R and G R in the weld center line welded samples.
Samples N1 N2 N3 N4
Fig. 10. Hardness profile in the sections of the weld, HAZ and base metal for all welded samples.