2.metallurgical - ijmmse - modelling.pdf
TRANSCRIPT
-
8/10/2019 2.Metallurgical - IJMMSE - Modelling.pdf
1/24
[email protected] www.tjprc.org
MODELING AND EXPERIMENTAL STUDIES OF DIFFUSION BONDING OF INCONEL
600 TO PYROLYTIC GRAPHITE
EKRAM ATTA AL-AJAJ 1, AWFA ABDUL- RASSOL ABDULLAH 2 & AHMED ALI MOOSA 3 1Dept. of Physics, University of Baghdad, College of Science, Iraq
2Dept. of Applied Science, University of Technology, Australia3Dept. of Production and Metallurgy, University of Technology, Australia
ABSTRACT
Modeling and experimental studies of diffusion bonding of Inconel 600/Nickel/Pyrolytic Graphite is investigated
in this research. Modeling implies utilization of ANSYS package to predict axisymmetric thermoelastic finite element
analysis from the above materials. The purpose from introducing axisymmetric model are: to achieve more accurate resultsand less analysis time; to calculate thermal stresses induced across diffusion bonded joints. Investigating thermal stress
levels along the potential failure interface is extremely helpful; these residual stresses are mostly the deriving forces of
joint failure. Axisymmetric finite element analysis involves applying external pressure on the joint and temperature as
second main parameter.
Experimental study implies diffusion bonded joints of Inconel 600 to graphite using nickel was subjected to shear
test to assess the bond strength. Based on shear testing results, a critical interlayer thickness as well as temperature,
pressure and holding time give optimum diffusion bonding parameters. Furthermore the annealing of cold drawn
interlayers and its effect on joint strength were investigated. Modeling and experimental results show that diffusion bonded
joints of Inconel 600/ Nickel/ Pyrolytic graphite have optimum shear strength of 12.9 MPa at 850 C,10 MPa for 30 min
holding time using 0.15 mm nickel interlayer. Further studies of the joints were carried out using Metallography,
Fractography, X-ray diffraction and microhardness measurements. Metallographic examination and X-ray diffraction
demonstrate the formation of new phases and solid solutions
KEYWORDS: Diffusion Bonding, Inconel 600/ Nickel/ Pyrolytic Graphite
INTRODUCTION
Diffusion bonding is a joining process in which two nominally flat surfaces are held together at an elevated
temperature [typically above 0.6 T m (melting temperature) of the least refractory materials] for a period of time until a
bond is formed. This process is relatively simple when two identical materials are to be joined, but when joining dissimilar
materials, there are many potential complications. Diffusion bonding of metals has a long history; it was one of the first
joining technologies developed by man [1].
Over the last four decades, there has been a real need for special purpose structural materials, which can be loaded
for a long period of time, at a very high temperature [2]. The extreme environment in space presents both challenge and
opportunity for material scientists [3]. The base materials in this research is Inconel 600 and pyrolytic graphite; they are
used in combination due to their outstanding high temperature strength in severe environment.
International Journal of Metallurgical &Materials Science and Engineering (IJMMSE)ISSN(P): 2278-2516; ISSN(E): 2278-2524Vol. 4, Issue 5, Dec 2014, 11-34 TJPRC Pvt. Ltd.
-
8/10/2019 2.Metallurgical - IJMMSE - Modelling.pdf
2/24
-
8/10/2019 2.Metallurgical - IJMMSE - Modelling.pdf
3/24
Modeling and Experimental Studies of Diffusion Bonding of Inconel 600 to Pyrolytic Graphite 13
[email protected] www.tjprc.org
EXPERIMENTAL
Raw Materials
Materials used in this research are; Inconel 600 and Pyrolytic graphite as base materials, Nickel interlayer (foil).
Inconel 600 and Nickel Interlayer
The first base material in this research is Inconel 600 and was examined using X-ray fluorescence (EDS), the
measured concentration of alloying elements in Inconel 600 base material are 72% Ni, 16% Cr and 7.5 % Fe, which agree
well with reported data [11]. Samples of Inconel 600 alloy were cut into squares with dimensions, 14 mm x 14 mm x 5 mm
using hydraulic cutting machine.
One surface of the samples was wet ground using silicon carbide papers and then polished with 1 m diamond
paste and then cleaned by alcohol, and ultrasonically cleaned for 10 min using acetone as a medium. Prior to cleaning,
each sample was then subjected to diffusion bonding experiments as shown in Fig.2.
Nickel foil grade 270 was used in this research (source is Herpol Industrial Projects H.I.P) Nickel interlayer were
subjected to pickling (NaF, H 2SO 4 and distilled water) to remove oxide film.
Figure 2: Diffusion Bonded Joints before Diffusion Bonding Experiment
Pyrolytic Graphite
Pyrolytic graphite rod (220 mm length x 50 mm dia.), was cut into the desired dimensions. Pyrolytic graphite rod
was cut to the desired dimension (14x 14 x 4.5 mm), and (14 x 14 x 9 mm).. Two surfaces of the graphite samples were
wet grounded, silicon carbide papers.
The samples are polished with 0.3 m alumina suspension, and followed by cleaning with alcohol. Graphite
samples are subjected to fine polishing to achieve fine scale roughness, and to given corresponding small defect size [12].
-
8/10/2019 2.Metallurgical - IJMMSE - Modelling.pdf
4/24
14 Ekram Atta Al-Ajaj, Awfa Abdul- Rassol Abdullah & Ahmed Ali Moosa
Impact Factor (JCC): 2.9076 Index Copernicus Value (ICV): 3.0
The samples were ultrasonically cleaned for (10 min) using Hexane as a medium. Prior to cleaning, each sample was
subjected to diffusion bonding experiment as shown in Fig. 2.
Vacuum Diffusion Bonding System Construction
In this research work a vacuum diffusion bonding equipment was built to suit the proposed conditions and
specimen dimensions for diffusion bonding of Inconel 600 to graphite using nickel. Each unit of the utilized system is
described in details and their functions are also explained. The whole unit is shown in Fig. 3
Figure 3: Heating System Constriction before Assembly (Loading System One)
Finite Element Analysis (ANSYS)
Axisymmetric finite element method seems to be superior for checking the localized stress concentration indiffusion bonded joint. Therefore FEM was used, for a rough estimation of induced residual thermal stress in Pyrolytic
Graphite / Inconel 600 joints. In the present study, a finite element method was adopted to calculate the residual stresses
(principle stresses) induced in the joint. The thermal expansions of most metals are much higher than graphite. As a result,
cooling to room temperature must be slow enough to avoid the joint failure. Investigating stress levels along the potential
failure interfaces is extremely helpful; these residual stresses are the deriving forces of the joint failure [13].
-
8/10/2019 2.Metallurgical - IJMMSE - Modelling.pdf
5/24
Modeling and Experimental Studies of Diffusion Bonding of Inconel 600 to Pyrolytic Graphite 15
[email protected] www.tjprc.org
RESULTS AND DISCUSSIONS
ANSYS Modeling
The Finite Element Analysis of Diffusion Bonded Joints
In this investigation, a stimulated model for diffusion bonded joints of Inconel 600 / nickel / graphite using (0.1,0.15, 0.2, 0.3 and 1 mm) interlayer thickness was introduced. Thermal stresses induced in the diffusion bonded joints were
examined using ANSYS package. Axisymmetric finite element method using ANSYS package seems to be superior for
checking the localized stress concentration induced in the joint after cooling [14].
In the present, one postulate the bonding at interface by considering a reaction layer of 10 m thickness at
graphite / nickel and Inconel 600 / nickel interfaces and this assumption was applicable for previous study [15]. Figs. (4-7)
show the principle stress distribution through assembly when the applied thermal load is 850 C, assuming that the cooling
of joint from bonding temperature to 100 C for different nickel interlayer thickness (1, 0.3. 0.2, 0.1 mm).
Figure 4: Three Dimensions Principal Stress Distribution across the Diffusion Bonding Joint using 1mm
Nickel Interlayer
Figure 5: Three Dimensions Principal Stresses Distribution across the Diffusion Bonding Joint using 0.3 mmNickel Interlayer
-
8/10/2019 2.Metallurgical - IJMMSE - Modelling.pdf
6/24
16 Ekram Atta Al-Ajaj, Awfa Abdul- Rassol Abdullah & Ahmed Ali Moosa
Impact Factor (JCC): 2.9076 Index Copernicus Value (ICV): 3.0
Figure 6: Three Dimensions Principal Stresses Distribution across the Diffusion Bonding Joint using 0.2 mm
Nickel Interlayer
Figure 7: Three Dimensions Principal Stresses Distribution across the Diffusion Bonding Joint using 0.1 mm
Nickel Interlayer Thickness
It was noticed that the strain in the joint was not uniform through assembly, the maximum tensile stress
(maximum principle stress) appeared in the graphite near joining interface about 264 MPa when the nickel interlayer
thickness is 0.1 mm. The amount of stress is far beyond the shear strength of graphite. Therefore, it is thought that
cracking in graphite will easily initiate during cooling to room temperature.
The Effect of Interlayer Thickness on Induced Principle Stresses of Diffusion Bonded Joints
As mentioned earlier, it is common for some form of strain reliving to be used when joining ceramics to metals at
high temperature. Finite Element Analysis could be used to examine the effect of interlayer thicknesses on thermal stress
during cooling from bonding temperature [12]. Figs. (8-10) illustrate thermal stress distribution in the z-axis direction of
the joint surface obtained by annealing at 750 C, 850 C and 950 C using different nickel interlayer thicknesses (1, 0.3, 0.2,
-
8/10/2019 2.Metallurgical - IJMMSE - Modelling.pdf
7/24
Modeling and Experimental Studies of Diffusion Bonding of Inconel 600 to Pyrolytic Graphite 17
[email protected] www.tjprc.org
0.15 and 0.1 mm). Tensile thermal stress was induced in graphite, whereas compressive thermal stress was induced in
nickel and Inconel 600.
-8.00E+08
-7.00E+08
-6.00E+08
-5.00E+08
-4.00E+08
-3.00E+08
-2.00E+08
-1.00E+08
0.00E+00
1.00E+08
0 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 1.8 2 2.2
Distance from Graphite free surface, x 0.5 mm
P r i n c p
l e S t r e s s ,
P a
0.1 mm
0.15 mm
0.2 mm
0.3 mm
1 mm
Figure 8: Thermal Stress Distribution on the Surface of Graphite/Nickel/ Inconel 600 Joints at 750 C Diffusion
Bonding Temperature using Different Nickel Interlayer Thickness. The y-axis Represents Longitudinal Stress as
Calculated by Finite Element Method
The maximum compressive thermal stress increases with increasing joining temperature. The maximum tensile
thermal stress in graphite is located about (0.2mm) from the joining interface as shown in Fig. 10. This position remains
almost unchanged even when joining temperature is varied. The purpose of the present study is to examine the effects of
nickel
-8.00E+08
-6.00E+08
-4.00E+08
-2.00E+08
0.00E+00
2.00E+08
4.00E+08
0 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 1.8 2 2.2
Distance from Graphite free surface, x 0.5 mm
P r i n c p
l e S t r e s s ,
P a
0.1 mm
0.15 mm
0.2 mm
0.3 mm
1 mm
Figure 9: Thermal Stress Distribution on the Surface of Graphite Nickel / Inconel Joints at 850 C Diffusion
Bonding Temperature, the y-axis Represent Longitudinal Stress as Calculated using the Finite Element Method
interlayer thickness on residual stress and to find the critical thickness for applicable joint. It was noticed that the thinner
nickel interlayer almost reduces maximum stress drastically (1-0.15 mm). It is also, noticed that, for nickel interlayer
thinner than 0.15 mm, the residual stress increased markedly [16]. For nickel interlayer thickness 0.1 mm, the principle
stress also shows a high increase in graphite near the joining interface at bonding temperature 750 and above. The
-
8/10/2019 2.Metallurgical - IJMMSE - Modelling.pdf
8/24
18 Ekram Atta Al-Ajaj, Awfa Abdul- Rassol Abdullah & Ahmed Ali Moosa
Impact Factor (JCC): 2.9076 Index Copernicus Value (ICV): 3.0
maximum tensile stress (364 MPa) appeared near nickel / graphite interface at 950 C diffusion bonding temperature as
shown in Fig. 10.
-1.20E+09
-1.00E+09
-8.00E+08
-6.00E+08
-4.00E+08
-2.00E+08
0.00E+00
2.00E+08
4.00E+08
6.00E+08
0 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 1.8 2 2.2
Distance from Graphite free surface, x 0.5 mm
P r i n c p
l e S t r e s s ,
P a
0.1 mm
0.15 mm
0.2 mm
0.3 mm
1 mm
Figure 10: Thermal Stress Distribution on the Surface of Graphite/Nickel / Inconel 600 Joints at 950 C DiffusionBonding Temperature using Different Nickel Interlayer Thickness. The y-axis Represents Longitudinal Stress as
Calculated by Finite Element Method
The Effect of Bonding Temperature on Maximum Principle Stresses
The effect of annealing temperature on maximum residual stress induced in graphite for different interlayer
thicknesses were investigated by finite element analysis. Fig. 11 represents the relationship between maximum longitudinal
stress on the surface of graphite and joining temperature. It implies that the principle stresses increase linearly with
increasing annealing temperature for interlayer thickness 1mm. This behavior is similar to a previous study of diffusion
bonding of graphite to nickel [17]. This is likely due to a more pronounced difference between room temperature andheating temperature. As shown in Fig. 11, for 0.15 mm nickel interlayer, the principle stresses shows almost a slight
increase with increasing bonding temperature till it reaches 800 C, where the decrease in residual stresses was observed
(300 MPa), this is may be related to the decrease in elastic modulus with increasing bonding temperature. Therefore those
joints are expected to be stable at room temperature. The maximum principle stresses induced are minimum in graphite
interface at all joining temperature for 0.15 mm interlayer thickness.
Figure 11: The Relationship between Maximum Longitudinal Stress on the Surface of Graphite and Diffusion
Bonding Temperature at Different Nickel Interlayer Thickness
0.00E+00
1.00E+08
2.00E+08
3.00E+08
4.00E+08
5.00E+08
6.00E+08
7.00E+08
8.00E+08
9.00E+08
1.00E+09
650 750 850 950
Temperature
M a x
i m u m
P r i n c p
l e S t r e s s ,
P a
0.1 mm
0.15 mm
0.2 mm
0.3 mm
1 mm
C o
-
8/10/2019 2.Metallurgical - IJMMSE - Modelling.pdf
9/24
Modeling and Experimental Studies of Diffusion Bonding of Inconel 600 to Pyrolytic Graphite 19
[email protected] www.tjprc.org
EXPEREMENTAL WORK
Shear Test
Shear tests were preformed on Inconel 600 / nickel / pyrolytic graphite using nickel interlayer . The process
parameters were chosen according to elastic and thermal properties of the materials being bonded. These preliminary
bonding variables were predicted by finite element analysis. The prediction of certain optimum bonding parameters
simplified the assessment of the; bonding temperature and interlayer thickness effect on the bonded joints strength. An
optimum external pressure of 10 MPa at 850 C for 30 min duration and 0.15 mm nickel interlayer was obtained as shown
in Fig 12.
0
5
10
15
650 750 850 950
Temperature ( C )
S h e a r S t r e n g t h
( M P a
Figure 12: Shear Stress Temperature Relationship for Inconel 600/Nickel/Graphite Joint Diffusion Bonded using
0.15 mm Nickel Interlayer at T C /10 MPa/ 30 min
Bonding was carried out in vacuum 2.6x10 -3 Pa. The effects of diffusion external pressure, temperature, holding
time and interlayer thickness on the bond strength were represented. Fig. 13 shows peak strength at the optimum diffusion
bonding condition. The optimum shear strength obtained is 12.9 MPa; the shear strength of bonded joint is approximately
equal to that of Pyrolytic Graphite (7-13 MPa) regardless of the joining conditions.
0
5
10
15
0 5 10 15 20 25 30 35 40
External Pressure (MPa)
S h e a r
S t r e n g
t h ( M P a
)
Figure 13: Shear Stress External Pressure Relationship for Inconel 600/nickel/Graphite Joint Diffusion Bondedusing 0.150 mm Nickel Interlayer at 850 C / P MPa /30 min
Nickel Interlayer with Thickness 0.125 mm
Figs.14-16 illustrate the relationship between diffusion bonding parameters. For 0.125 mm nickel interlayer,
diffusion bonded joints were made at temperature 800 C, for 60 min, but at different external pressure, as reported in Fig.
14, two regions were observed. In the first region, shear strength increases with increasing pressure to a certain limit 15
MPa, this can be explained that an increase in pressing load raise bond strength up to a certain value. Any further increase
in pressing load reduce it, the demonstration of this, and the depth of diffusion zone depend on pressing load, and pressing
load equally affects diffusion and self diffusion [2]. Diffusion bonded joints were made at temperature 700 C and for 180
-
8/10/2019 2.Metallurgical - IJMMSE - Modelling.pdf
10/24
20 Ekram Atta Al-Ajaj, Awfa Abdul- Rassol Abdullah & Ahmed Ali Moosa
Impact Factor (JCC): 2.9076 Index Copernicus Value (ICV): 3.0
min but at different external pressure they were tested to investigate the effect of external pressure on bond strength. The
results are presented in Fig. 15, where the increased bond strength is shown and no peak strength was observed, and this
was explained in terms of low activation energy to activate diffusion across interface and this is related to the lower
bonding temperature even at high external pressure.
0
5
10
15
0 5 10 15 20 25 30 35 40
External Pressure (MPa)
S h e a r
S t r e n g
t h ( M P a )
Figure 14: Shear Stress External Pressure Relationship for Inconel 600/Nickel/Graphite Joint Diffusion Bonded
using 0.125 mm Nickel Interlayer at 800 C / P MPa / 60 min
0
5
10
15
0 5 10 15 20 25 30 35 40
External Pressure (MPa)
S h e a r
S t r e n g
t h ( M P a )
Figure 15 Shear Stress External Pressure Relationship for Inconel 600/Nickel/ Graphite Joint Diffusion Bonded
using 0.125 mm Nickel Interlayer at 700 C / P MPa / 180 min
At 35 MPa, the shear strength is decreased. Results are presented in Fig. 16, to investigate the effect of holding
time on joint strength at temperature 750 C for external pressure 10 MPa, the increase in bond strength shows no peak
strength, the relationship between shear stress and holding time is almost constant.
0
5
1015
0 50 100 150 200
Time (min)
S h e a r
S t r e n g
t h ( M P a
Figure 16: Shear Stress Time Relationship for Inconel 600/Nickel/Graphite Joint Diffusion Bonded using 0.125
mm by Nickel Interlayer at 750 C / 10 MPa / t min
-
8/10/2019 2.Metallurgical - IJMMSE - Modelling.pdf
11/24
Modeling and Experimental Studies of Diffusion Bonding of Inconel 600 to Pyrolytic Graphite 21
[email protected] www.tjprc.org
As one can see from the above mentioned results, changing the diffusion bonding parameters did not affect the
joint strength markedly. This can be explained in view of residual stresses that induced at graphite interface during cooling
to room temperature for interlayer thickness thinner than 0.15 mm. At all bonding temperature, principle stresses (tensile
stress) were induced in graphite near joining interface and were propagated along graphite material, at 850 C and abovethe increase in bonding temperature causes marked increase in tensile stress in graphite (280-364 MPa), as reported in
Fig.9 and 10, this is confirmed with the finite element results. Fig. 17.d shows a graphite fractured specimen at low shear
strength at 850 C, 10 MPa and 30 min.
Figure 17: Shear Test Fractured Diffusion Bonded Samples: a) at High Shear Strength, b) at Moderate Shear
Strength, c) at Low Shear Strength and d) Failure of Graphite Specimen at Almost High Thermal Tensile
Stresses using thin Nickel Interlayer
Nickel Interlayer with Thickness 0.15 mm
Diffusion bonded joints were made at external pressure 10 MPa, 30 min holding time, but at different bondingtemperature, the results are illustrated in Fig. 12. Fig. 13 shows peak strength at diffusion bonding condition, 850 C, 10
MPa and 30 min holding time.
Two regions were found, in the first region, the shear stress increases with increasing external pressure up to 10
MPa this is due to the formation of intimate contact and the activation of diffusion process along interface. In the second
region, the shear strength starts to decrease with increasing external pressure after 10 MPa. This can be explained by the
damaging of graphite in the vicinity of the joint and by providing the path for crack propagation as shown in Fig. 18.
Optimum shear strength obtained is 12.9 MPa, bonded joints were fractured at almost high shear strength. This
implies that strong diffusion bonding requires both adequate joining pressing load to crash asperities and sufficiently highbonding temperature to activate atomic diffusion, see Fig. 17.a
-
8/10/2019 2.Metallurgical - IJMMSE - Modelling.pdf
12/24
22 Ekram Atta Al-Ajaj, Awfa Abdul- Rassol Abdullah & Ahmed Ali Moosa
Impact Factor (JCC): 2.9076 Index Copernicus Value (ICV): 3.0
Figure 18: Optical Micrograph of Cross Section of Inconel 600 /Nickel/Graphite Joint Bonded at 850 C, 25 MPa
and 30 Min using 0.15 mm Nickel Interlayer. Crack Initiation at interface and Propagation in Graphite Specimen
The corresponding predicted finite element results confirm the experimental results, where Fig. 11 illustrates the
lowest principle stress value that induced in graphite. Diffusion bonded joints were made at 5 MPa and 180 min at different
bonding temperatures. Fig. 19 illustrates the results; it shows an increase in bond strength with increasing temperature.
0
5
10
15
650 750 850 950
Temperature ( C )
S h e a r
S t r e n g
t h ( M P a
Figure 19: Shear Stress Temperature Relationship for Inconel 600/Nickel/Graphite Joint Diffusion Bonded using
0.15 mm Nickel Interlayer at T C / 5 MPa /180 min
The highest value of shear strength were obtained at; 900 C, 5 MPa, and 180 min. The highest shear strength
obtained is 6.73 MPa, the bonded joints were fractured at a moderate shear strength value, and this was due to insufficient
pressing load to achieve intimate contact between graphite and nickel interface. Diffusion bonded joints were obtained at
temperature 800 C, 5 MPa but at different holding time. Fig. 20 illustrates shear tested samples, where the increase in bond
strength with increasing holding time was observed.
The highest value of shear strength 6.1 MPa were obtained at holding time 180 min for bonded joint at 800 C, 5
MPa, at different holding time. This can be explained by the removal of defects, where at the instant of actual contact the
matching surface contained ultramicroscopic discontinuities and inclusions, were the removal of defects required some
more time [18,19]. Thus, the shear strength of the joint depends not only on joining temperature and pressure, but also on
keeping time.
-
8/10/2019 2.Metallurgical - IJMMSE - Modelling.pdf
13/24
Modeling and Experimental Studies of Diffusion Bonding of Inconel 600 to Pyrolytic Graphite 23
[email protected] www.tjprc.org
0
5
10
15
0 50 100 150 200
Time (min)
S h e a r S
t r e n g
t h ( M P a )
Figure 20: Shear Stress Time Relationship for Inconel 600/Nickel/Graphite Joint Diffusion Bonded using 0.15 mm
Nickel Interlayer at 800 C / 5 MPa / t min
It is clear that 0.15 mm nickel interlayer is the best choice for reliable bonded joint for Graphite / Nickel / Inconel
600 which utilized the lowest value of principle stress that induced in graphite.
This principle stress is the major reason for reducing the joint strength, as shown in Figs. 4-7 where residual stressanalysis for 0.15 mm nickel interlayer at different bonding temperatures shows two regions, the first region indicates a
slight increase of residual stresses with increasing temperature up to 800 C, the second regions indicate decrease of
residual stresses with increasing temperature from 800 C and above. This behavior reduces the chance for initiating crack
along interface during cooling and yields a stable joint [13].
Nickel Interlayer with Thickness 0.2 mm 0.5 mm
For nickel interlayers thicknesses; 0.2 and 0.4 mm the results show no bonded joints at different bonding
conditions, the residual stress analysis confirms this observation as shown in Fig. 11. It can be seen that nickel interlayer
thickness thicker than 0.15 mm gives maximum principle stress of (400-900 MPa) which is higher than the fracture stressof graphite (120-220MPa). For 0.3 and 0.5 mm nickel interlayer, the increase in shear strength with increasing bonding
temperature was observed, the highest shear strength obtained was 4.7 MPa at, 800 C, 25 MPa and 30 min holding time
using 0.5 mm, as shown in Fig. 21, this can be explained in view of relaxation of thermal stress by plastic deformation,
where the maximum tensile thermal is released by plastic deformation of nickel interlayer.
0
5
10
15
650 750 850 950
Temperature ( C )
S h e a r S t r e n g t h
( M P a
Figure 21: Shear Stress Temperature Relationship for inconel 600/Nickel/Graphite Joint Diffusion Bonded using
0.5 mm Nickel Interlayer at T C / 25 MPa / 30 min
Fractographic Examination Results using Optical Microscope
Fracture surface of the shear tested joint at different diffusion bonding conditions were examined. Topography at
optimum diffusion bonding condition; 850 C, 10 MPa and 30 min using 0.15 mm nickel interlayer are shown in Fig. 22,the samples were fractured at almost high shear strength. It shows the fractured nickel side with graphite fragments. The
-
8/10/2019 2.Metallurgical - IJMMSE - Modelling.pdf
14/24
24 Ekram Atta Al-Ajaj, Awfa Abdul- Rassol Abdullah & Ahmed Ali Moosa
Impact Factor (JCC): 2.9076 Index Copernicus Value (ICV): 3.0
black area is graphite while white area is nickel; the fractured type mode (1) was the crack initiated at interface and then
propagated through graphite material.
Figure 22: Fracture Surface (nickel interface) of Inconel 600/nickel/Graphite Joint Bonded at 850 C, 10 MPa and
30 min using 0.15 mm Nickel Interlayer, The Specimen were Fractured at Almost High Shear Strength (mode 1)
The formation of bond zone was assisted by the presence of nickel on the graphite fractured side. The diffusion
bonded joints at bonding condition 900 C, 5 MPa, and 180 min using 0.15 mm interlayer is shown in Fig. 23, the white
area is nickel, while the black area is graphite particles bonded with nickel matrix [20,21]. The samples were fractured at
almost moderate shear strength (fracture mode 2) due to low pressing load that causes discontinuity through graphite /
nickel interface.
Figure 23: Fracture Surface (nickel interface) of Inconel 600/Nickel/Graphite Joint Bonded at 900 C, 5 MPa and180 min using 0.15 mm Nickel Interlayer, The Specimen were Fractured at Almost Moderate Shear Strength
(mode 2), the Increase in Grain Size with Increasing Holding Time
Fig. 24 shows a photomicrograph of nickel / graphite bonded region (left) and non- bonded region (right) at800 C, 10 MPa and 30 min using 0.15 mm nickel interlayer, the joint was fractured at moderate shear strength (fracture
mode 2) and this can be explained as that the plastic deformation of nickel interlayer was insufficient to enlarge the contact
zones between the faying surfaces [19].
-
8/10/2019 2.Metallurgical - IJMMSE - Modelling.pdf
15/24
Modeling and Experimental Studies of Diffusion Bonding of Inconel 600 to Pyrolytic Graphite 25
[email protected] www.tjprc.org
Figure 24: Photomicrograph of Nickel/Graphite Previously Bonded Region (left) at 800 C, 10 MPa and 30 min
using 0.15 mm Nickel Interlayer and Non-Bonded Region (Right), Fracture Mode 2
Metallographic Examination
Figs. 25-32 show the microstructure of the longitudinal polished cross section of the bonded joints after etching at
different bonding conditions. According to the phase diagram of nickel-carbon system, the equilibrium phases at room
temperature are nickel and graphite.
Figure 25: Optical Micrograph of Cross Section of Inconel 600/Nickel/Graphite Joint Bonded at 850 C, 10 MPa and
30 min using 0.15 mm Nickel Interlayer. Diffusion of Carbon through Nickel at Graphite/Nickel Interface
Figure 26: Optical Micrograph of Cross Section of Inconel 600 /Nickel/Graphite Joint Bonded at 800 C, 10 MPa
and 30 min using 0.15 mm Nickel Interlayer
-
8/10/2019 2.Metallurgical - IJMMSE - Modelling.pdf
16/24
-
8/10/2019 2.Metallurgical - IJMMSE - Modelling.pdf
17/24
Modeling and Experimental Studies of Diffusion Bonding of Inconel 600 to Pyrolytic Graphite 27
[email protected] www.tjprc.org
Figure 29: Optical Micrograph of Cross Section of inconel 600/Nickel/Graphite Joint Bonded at 800 C, 25 MPa and
30 min using 0.5 mm Nickel Interlayer. Recrystallization of Nickel at Nickel/ Graphite Interface
Fig. 30 shows recrystallized grains near nickel / Inconel 600 interface for diffusion bonding condition, 850 C, 20
MPa, and 180 min using 0.3 mm.
Figure 30: Optical Micrograph of Cross Section of Inconel 600 /Nickel/Graphite Joint Bonded at 850 C, 20 MPa
and 180 min using 0.3 mm Nickel Interlayer. Recrystallization at Inconel 600 /Nickel Interface
Fig. 31 and 32 show a micrographs of Inconel 600 / nickel / graphite joint interface bonded at 850 C, 25 MPa and
180 min using 0.3 mm interlayer thickness, but at different holding time (180 and 30 min). At holding time 180 min, a
scattered microvoids were appeared in Fig. 31, while large numbers of microvoids are seen at the Inconel 600 /nickel
interface at 30 min, see Fig. 32.
-
8/10/2019 2.Metallurgical - IJMMSE - Modelling.pdf
18/24
-
8/10/2019 2.Metallurgical - IJMMSE - Modelling.pdf
19/24
Modeling and Experimental Studies of Diffusion Bonding of Inconel 600 to Pyrolytic Graphite 29
[email protected] www.tjprc.org
Figure 33: Hardness versus Distance for Inconel 600/0.15 mm Nickel /Graphite JointBonded at 850 C/10 MPa/ 30 min
Figure 34: Hardness versus Distance for Inconel 600 / 0.15 mm Nickel / Graphite JointBonded at 700 C / 10 MPa / 30 min
Figure 35: Hardness versus Distance for Inconel 600 / 0.125 mm Nickel / Graphite JointBonded at 800 C / 10 MPa / 30 min
-
8/10/2019 2.Metallurgical - IJMMSE - Modelling.pdf
20/24
30 Ekram Atta Al-Ajaj, Awfa Abdul- Rassol Abdullah & Ahmed Ali Moosa
Impact Factor (JCC): 2.9076 Index Copernicus Value (ICV): 3.0
The change in Vickers hardness comes as a result of formation of supersaturated solution of carbon atoms in
nickel matrix, the increase was due to solid solution hardening. This explains the slight increase in hardness across the
interface, Fig. 33-35 show that the hardness decreases with increasing distance from graphite / nickel interface, a common
feature appear in all figures that the hardness of nickel matrix after diffusion bonding is less than the hardness of asreceived nickel material due to annealing process that companied the diffusion bonding process. At almost high bonding
temperature 850 C, 10 MPa external pressure, an increase in hardness at nickel / Inconel 600 interface Fig. 33 due to solid
solution hardening of Ni-Cr. The high solubility of chromium in nickel and high chromium contents of Inconel 600 lead to
appreciable concentration of chromium in solid solution hardening [23].
X - Ray Diffraction Analyses
Fig. 36 shows diffusion bonded joint at 900 C, 5 MPa and 180 min using 0.15 mm nickel interlayer ; were a
certain appearance of graphite and nickel peaks at nickel interface is observed. This means the existence of solid solution
of Ni C at diffusion bonding interface [92]. The same behavior were observed for Inconel 600 / nickel/ graphite shearfractured joint at diffusion boding condition; 800 C, 10 MPa and 30 min using 0.15 mm nickel interlayer, as shown in Fig.
37.
Figure 36: Diffraction Pattern from the Surface of Fractured Surface of Graphite /Nickel /Inconel 600 Bonded Joint
at 900 C / 5 MPa / 180 min using 0.15 mm Interlayer Thickness (Nickel Interface)
-
8/10/2019 2.Metallurgical - IJMMSE - Modelling.pdf
21/24
Modeling and Experimental Studies of Diffusion Bonding of Inconel 600 to Pyrolytic Graphite 31
[email protected] www.tjprc.org
Figure 37: Diffraction Pattern from the Surface of Fractured Surface of Graphite /Nickel /Inconel 600 Bonded Joint
at 800 C / 10 MPa / 30 min using 0.15 mm Interlayer Thickness (Nickel Interface)
Fig. 38 shows diffraction pattern of fractured shear tested joint at 850 C, 25 MPa and 30 min, using 0.3 mm
nickel interlayer. The XRD analysis shows no graphite existence at the graphite / nickel interface this is due to an increase
in external pressure which inhibits diffusion across interface and this can be explained as, there is no solid solubility of
carbon in nickel matrix which is believed to be the only available bonding mechanism in graphite / nickel couples. The
same peak intensity was observed for Inconel 600 / nickel / graphite diffraction pattern of fractured shear test joint at
750 C, 25, MPa and 30 min using 0.5 mm nickel interlayer. This is due to high residual stresses induced at nickel /
graphite interface, as shown in Fig. 39.
Diffraction Angle, 2 (degree)
Figure 38: Diffraction Pattern from the Surface of Fractured Surface of Graphite / Nickel / Inconel 600 Bonded
Joint at 850 C/ 25 MPa / 30 min using 0.3 mm Interlayer Thickness (Nickel Interface)
Diffraction Angle, 2 (degree)
Figure 39: Diffraction Pattern from the Surface of Fractured Surface of Graphite / Nickel / Inconel 600 Bonded
Joint at 750 C / 25 MPa / 30 min using 0.5 mm Interlayer Thickness (Nickel Interface)
CONCLUSIONS
-
8/10/2019 2.Metallurgical - IJMMSE - Modelling.pdf
22/24
32 Ekram Atta Al-Ajaj, Awfa Abdul- Rassol Abdullah & Ahmed Ali Moosa
Impact Factor (JCC): 2.9076 Index Copernicus Value (ICV): 3.0
This research has demonstrated the feasibility of diffusion bonding of Inconel 600/ nickel/ Pyrolytic graphite and
Inconel 600/304 st.st./ Pyrolytic graphite. Graphite was bonded to Inconel 600, using nickel and 304 st.st. interlayers in the
solid state under external pressure of 5-35 MPa in vacuum at various temperatures for different keeping times. The
objectives of this research were successfully met and had two stages;
The first stage was modeling axisymmetric finite element analysis
Study thermal stresses induced in Inconel 600 / nickel /graphite diffusion bonding joint during cooling to room
temperature.
Compressive thermal stress is induced on the surface of nickel and Inconel 600.
Maximum tensile thermal stress is induced on the surface of graphite, using 0.1 mm nickel interlayer at a distance
of 0.2 mm from the joining interface, and the stress increases with increasing bonding temperature.
The second stage, achieves diffusion bonding experimentally. This work demonstrated that shear testing,
metallographic study, microhardness test, X-ray diffraction and fractographic examination can reveal considerable
information which is helpful in describing the diffusion bonding of Inconel 600/ nickel / pyrolytic graphite.
Mechanical and physical properties of base materials under investigation were investigated. However, the
research reported has shown the following significant points about diffusion bonding of Inconel 600/nickel/ graphite:
1. Wide variation in physical properties (Young modulus and thermal expansion) of the base materials to be
diffusion bonded as well as interlayers requires sufficient control of heating cycle to achieve heat balance
between base materials.
Optimum shear strength of Inconel 600/ nickel / graphite diffusion bonded joints is 12.9 MPa at 850 C, 10 MPa
and 30 min holding time, using 0.15 mm nickel interlayer.
All joints fail at graphite /nickel diffusion bonded interfaces in shear testing.
The critical interlayer thickness that gives reliable bond strength is 0.15 mm for nickel interlayer.
New phases and solid solutions are formed in Inconel 600 / nickel and nickel /graphite interfaces.
The microhardness values of Inconel 600/nickel/ graphite diffusion bonded joints shows an increase at joint
interface due to the formation of solid solution hardening and precipitation of new phases; about 20 m at nickel /Inconel 600 interface is due to recrystallization. At nickel/ graphite interface, the increase in hardness was due to
formation of solid solution (Ni-C) at almost 40 m from interface.
REFERENCES
1. M.G. Nicholas ,Joining of Ceramics, Chapter 2 Chapman and Hull, London, UK, 1990.
2. N.F. Kazakov, Diffusion Bonding of Materials, Mir Publishers, Moscow, 1985.
3. S. Rawal, Metal Matrix Composites for Space Applications, Vol.4, No.53, 2001, PP 14-17.
-
8/10/2019 2.Metallurgical - IJMMSE - Modelling.pdf
23/24
Modeling and Experimental Studies of Diffusion Bonding of Inconel 600 to Pyrolytic Graphite 33
[email protected] www.tjprc.org
4. T. Nishada & H. Sueyoshi, Solid State Bonding of Graphite to Inconel 718, J. Japan Ins. Metals, Vol.65, No.4,
2001, PP 303-309.
5. H. I. Salah ,Evaluation of Graphite/ Graphite Brazing, M.Sc. Thesis, University of Technology, Dept. of
Production Engineering and Metallurgy, Baghdad, Iraq, 2005.
6. Y. Bienveuu, T. Massart, et al, The Metallurgy of Diffusion Bonding Cranfield, Ed. Pearce, 1987.
7. N. F. Kazakov, E. K. Soluyanova & B. F. Shibryaev, The Diffusion Bonding of High Porosity Materials, J. of
Welding Production, The Welding Institute, No.7, 1977, PP 14- 15.
8. B. Derby & E. R. Wallach, J. Mat. Sci., No.19, 1984, PP 3140-3148.
9. P. G. Partidge & C. M. Ward, Diffusion Bonding of Advanced Materials, J. Metals and Materials, Vol. 5, No.6,
1989, PP 334-339.
10. D. R. Askeland & P. P. Phule, The science and Engineering of Materials, 4th Ed., Thomson Learning, Academic
Center, 2003.
11. Handbook of Materials Selections, ED Myer Kutz, Pub. John Wiley & Sons, Inc, USA, 2002.
12. B. Derby G. A. Dbriggs & E. R. Wallach, Non- Distractive Testing and Acoustic Microscopy of Diffusion Bonds,
J. Mat. Sci., Chapmen and Hall Ltd. No.18, 1983, PP 2345-2353.
13. Levy, Thermal Residual Stresses in Ceramic to Metal Brazed Joints, J. Ceram. Soc, Vol. 72, No. 9, 1991, PP
2141- 2147.
14. E. R. Wallach and A. Hill, Modeling of Diffusion Bonding, in Diffusing Bonding (ed. R.. Pearce), Cranfield,UK, 1987.
15. Abdul-Rahman, Joining Sialon Stainless Steel and Sialon-Sialon, Ph.D. Thesis, University of Strethclyde,
Faculty of Engineering, Dept. of Metallurgy, Engineering Material, 1993.
16. S. H. Avner, Introduction to Physical Metallurgy, 2 nd ED, McGraw Hill, 1983.
17. T. Nishada and H. Sueyoshi, Solid State Bonding of Graphite to Nickel, J. Japan Ins. Metals, Vol.64, No.8, 2000,
PP 597- 603.
18.
A. A. Akber, Diffusion Bonding of Oxygen Free High conductivity Copper to Austenitic Stainless Steel, Ph. D.Thesis, University of Technology, Department of Production Eng. and Metallurgy, Baghdad, Iraq, 1996.
19. C. Beraud, M. Courbiere, C. Esnouf & D. Juve, D. Treheux , Study of Copper- Alumina Bonding, J. Mater. Sci.,
No.24, 1989, PP 4545-4554.
20. K. Landry, S. Kalogeropulou & N. Eustathopoulos, Wettability of Carbon by Aluminum and Aluminum Alloys,
Materials Science and Engineering A 254, 1998, PP 99-111.
21. S. W. Ip, R. Stridhar, J. M. Toguri, T. F. Stephenson & A. E. M. Warner, Wettability of Nickel Coated Graphite
by Aluminum, Materials Science and Engineering A244, 1998, PP 31-38.
-
8/10/2019 2.Metallurgical - IJMMSE - Modelling.pdf
24/24
34 Ekram Atta Al-Ajaj, Awfa Abdul- Rassol Abdullah & Ahmed Ali Moosa
22. Hansen, Metallurgy and Metallurgical Engineering Series, Constitution of Binary Alloy, McGraw-Hill Book
Company, 1958, PP 374.
23. A. K. Jena & M.C. Chaturvedi, The Role of Alloying Elements in the Design of Nickel Base Super Alloys, J.
Mat. Sci., No.19, 1984, PP 3121-3139.