• 沒有找到結果。

Infrared brazing of Mo using the 70Au?22Ni?8Pd alloy

N/A
N/A
Protected

Academic year: 2021

Share "Infrared brazing of Mo using the 70Au?22Ni?8Pd alloy"

Copied!
7
0
0

加載中.... (立即查看全文)

全文

(1)

Infrared brazing of Mo using the 70Au–22Ni–8Pd alloy

D.W. Liaw

a

, R.K. Shiue

b,*

a

Department of Materials Science and Engineering, National Dong Hwa University, Hualien 974, Taiwan, ROC

b

Department of Materials Science and Engineering, National Taiwan University, Taipei 106, Taiwan, ROC Received 17 August 2004; accepted 28 October 2004

Abstract

Infrared brazing of Mo using the 70Au–22Ni–8Pd braze alloy is performed in the experiment. Both Au-rich and Ni-rich phases are observed in the infrared brazed joint at 1050C. The Au in the molten braze is not reacted with the Mo substrate, but the Ni in the braze preferentially reacts with the Mo. With increasing the brazing temperature and/or time, the interfacial reaction between the molten braze and Mo substrate is greatly enhanced, and the angular MoNi intermetallic compound is formed in the brazed joint, especially for the furnace brazed specimen due to its slow thermal history. The formation of interfacial MoNi intermetallic com-pound in the brazed joint is detrimental to its bonding strength. Cleavage dominated fracture is widely observed from the fracto-graph of furnace brazed specimen. Additionally, the application of rapid infrared brazing is a very effective process to inhibit excessive growth of the interfacial MoNi phase in the joint.

 2004 Elsevier Ltd. All rights reserved.

Keywords: Infrared brazing; Interface; Molybdenum; Metals and alloys; Microstructure

1. Introduction

Molybdenum is a type of refractory metals due to its melting point exceeding 2450C [1,2]. The use of the refractory metals is widely applied in the aircraft, de-fense, space and nuclear industries [2]. Developing appropriate joining process for metals is indispensable for structural applications. Brazing is accepted as one of methods in bonding assemblies made from refractory metals due to its limited effect on base-metal properties [3]. Refractory metals must be brazed at temperatures below those at which recrystallization occur for maxi-mum strength. Copper and silver-base braze alloys have been previously applied in brazing refractory metals for low-temperature service [3–6]. Additionally, it has also been reported that the single crystal molybdenum is suc-cessfully brazed by using Pd, Pd–20Ag and Mo–40Ru

for high temperature applications [7–9]. In contrast, the gold-based braze alloy is suitable for medium service temperatures. It is selected as the filler metal in brazing pure molybdenum.

Infrared brazing is featured with fast heating rate up to 3000C/min, which is much higher than that of tradi-tional furnace brazing[10–12]. It is an innovative braz-ing process due to its very rapid thermal history. This research is concentrated on vacuum brazing of molybde-num using the 70Au–22Ni–8Pd braze alloy. Both tradi-tional furnace brazing and infrared brazing are performed in the study. The wettability test, microstruc-tural evolution and shear strength of the brazed joint with various brazing conditions have been widely evalu-ated in the experiment.

2. Experimental procedures

The base metal used in the test was the pure molyb-denum rod. It was machined into the brazing specimen 0263-4368/$ - see front matter  2004 Elsevier Ltd. All rights reserved.

doi:10.1016/j.ijrmhm.2004.10.003

* Corresponding author Tel.: +886 2 27023069; fax: +886 2

23634562.

E-mail address:[email protected](R.K. Shiue).

(2)

with dimension of 10 mm· 10 mm · 3 mm. The sub-strate was polished with SiC papers up to 1200 grit, and subsequently cleaned using an ultrasonic bath with acetone as the fluid prior to the brazing experiment. Palniro-7 foil with 50 lm thick and 25 mm wide was purchased from Wesgo Metals Inc. Its nominal compo-sition in weight percent is 22% Ni, 8% Pd and Au bal-ance. Additionally, the chemical composition of the Palniro-7 braze alloy is in accordance with the BAu-6 braze alloy specified by the American Welding Society (AWS) [3]. Its solidus and liquidus temperatures are 1010 and 1045C, respectively. The recommended braz-ing range of 70Au–22Ni–8Pd alloy is between 1045 and 1120C.

Dynamic wetting angle tests were employed using a conventional vacuum furnace with various temperatures for 0–2400 s, respectively [4–6]. The braze alloy with near spherical shape used in the wetting angle test was prepared from 70Au–22Ni–8Pd foil by vacuum arc remelting (VAR) with the operation voltage of 60 V and 130–150 A. The weight of each spherical ball was kept at 0.12 g [4–6]. The spherical ball was placed on the molybdenum substrate, and images of the molten braze on the substrate were recorded simultaneously using an Olympus C-5050 digital camera during the wet-ting angle measurement.

Both furnace and infrared brazing were performed in the experiment with a vacuum of 5· 10 3Pa at various brazing conditions. The heating rate of conventional vacuum furnace was 0.5C/s, and the heating rate of infrared furnace was kept at 10C/s throughout the experiment. Additionally, all brazed specimens were pre-heated at 600C for 600 s before the brazing tempera-ture was attained. Table 1 summarizes all brazing conditions applied in the test.

The cross-section of the brazed specimen was exam-ined using a Hitachi 3500H scanning electron micro-scope (SEM) with the accelerating voltage of 20 kV. Chemical analyses were performed using energy-disper-sive X-ray spectrometer (EDS) with an operation volt-age of 20 kV and minimum spot size of 1 lm. Shear tests were employed using a Shimadzu AG-10 universal testing machine with a constant crosshead speed of 0.5 mm/min compressed the brazed specimens [10–12]. Failure analyses of the brazed specimen after shear tests were inspected using an SEM.

3. Results and discussion

Fig. 1 displays the dynamic wetting angle

measure-ments of 70Au–22Ni–8Pd alloy on the molybdenum substrate at 1070C and 1100 C for 0–2400 seconds, respectively. According to the figure, the molten braze alloy can completely wet the Mo substrate at 1070C in 300 seconds. Additionally, increasing the test temper-ature has little effect in further improving wettability of the molten braze on the substrate.

Fig. 2 displays the SEM secondary electron image

(SEI), backscattered electron image (BEI) and chemical analysis results of the furnace brazed Mo/70Au–22Ni– 8Pd/Mo specimen at 1050C for 1200 s. According to the EDS analysis results, the furnace brazed joint pri-marily consists of Au-rich and Ni-rich matrix and angu-lar MoNi intermetallic compound. Additionally, the location of huge angular MoNi intermetallics (marked by point C in Fig. 2) is close to the interface between the braze alloy and Mo substrate. The Au-rich phase (marked by B) is alloyed with Ni and Pd as marked by

B in Fig. 2, and there is no Mo content detected in the

EDS chemical analysis. In contrast, the Ni-rich phase (marked by A) is alloyed with Au, Pd and minor Mo. The Au–Ni–Pd ternary alloy phase diagram is cited here in order to unveil the microstructural evolution of the brazed joint.

Fig. 3(a) shows the liquidus projection of Au–Ni–Pd

ternary alloy phase diagrams, and Fig. 3(b) is the iso-thermal section of Au–Ni–Pd at 300C[13]. Both Fig. 3(a) and (b) are in atomic percent. The chemical compo-sition of the braze alloy in atomic percent is 44.1% Au, 46.5% Ni and 9.4% Pd as marked by point B in Fig. 3(b). According to the isothermal section of Au–Ni–Pd at 300C, the braze alloy tends to separate into two phases, the Au-rich and Ni-rich phases[13].Fig. 4shows the 10 wt% Pd vertical section of Au–Ni–Pd pseudo-binary phase diagram [13]. The atom percent Pd varied

Table 1

Summary of brazing variables used in the experiment

Composition of braze alloys (wt%) 70Ag–22Ni–8Pd Furnace brazing temperature (C) 1050

Furnace brazing time (s) 1200

Infrared brazing temperature (C) 1050, 1100 Infrared brazing time (s) 90, 180, 300

Wetting Angle (degree)

140 120 100 80 60 40 20 0 0 300 600 900 1200 1500 180021002400 1070 °C 1100 °C Time (sec) 2700

Fig. 1. The dynamic wetting angle measurements of 70Au–22Ni–8Pd alloy on molybdenum substrate at 1070C and 1100 C for 0–2400 s. 92 D.W. Liaw, R.K. Shiue / International Journal of Refractory Metals & Hard Materials 23 (2005) 91–97

(3)

from one end of the section to the other does maintain a constant weight percentage due to different atomic num-bers of Au, Ni and Pd in the ternary alloy system. For example, 10 wt% Pd can be obtained from the composi-tion of 82.9 at% Au and 17.1 at% Pd at the left end of

Fig. 4. Similarly, 10 wt% Pd can also be obtained from

the composition of 94.2 at% Ni and 5.8 at% Pd at the right end of Fig. 4. In other words, a straight line in weight percent will become a curve in atomic percent for the ternary alloy phase diagram. According to the

Fig. 4, the Au-rich phase can dissolve Pd and Ni, and

the Ni-rich phase dissolves Pd and Au. It is expected that both Au-rich and Ni-rich phases are expected in the brazed joint. Based onFig. 2(b) and (d), the brazed joint mainly consists of the Au-rich phase, and minor needle like Ni-rich phase separated from Au-rich matrix is also observed. It is consistent with the ternary Au–Ni– Pd phase diagram.

Based on the binary alloy phase diagram, the Au–Mo phase diagram is a simple eutectic with no intermediate phase[14]. The Au-rich matrix has the maximum solid solution of 1.25 at% Mo at the eutectic temperature of 1054C, and the Mo merely has the maximum solubility of 0.4 at% Au[14]. In contrast, the Ni can dissolve Mo

up to 17 at% at 700C. Accordingly, the dissolution of Mo substrate into the Ni-rich melt is much more prom-inent than that of Au-rich melt. It is concluded that wet-ting of the Mo substrate by the molten braze is primarily attributed to the molten braze containing Ni. Addition-ally, the presence of interfacial angular MoNi inter-metallic compound demonstrates strong metallurgical reaction between the molten braze and Mo substrate as displayed in Fig. 2. The formation of interfacial MoNi phase significantly decreases the amount of Ni content from the molten braze alloy. Although the Ni content of the braze alloy is as high as 46.5 at%, the Ni-rich phase in the brazed joint is greatly decreased after brazing.

Fig. 5shows SEM images and EDS chemical analysis

results of infrared brazed joints with various brazing conditions. The microstructure of the infrared brazed joint at 1050C (Fig. 5(a)–(c)) is very different from that of furnace brazed specimen (Fig. 2). The infrared brazed joint is primarily consists of the Au-rich matrix and interfacial Ni-rich as shown in Fig. 5(a)–(c). Addition-ally, the interfacial Ni-rich phase alloyed with Au, Pd and Mo is observed as marked by A in Fig. 5(c). It is also noted that there is no intermetallic phase found in Fig. 2. The SEM: (a), (c) SEIs; (b), (d) BEIs and EDS chemical analysis results of furnace brazed joints at 1050C for 1200 s.

(4)

the figure. As described earlier, the existence of Ni-rich melt is beneficial to wettability of the molten braze on the substrate, and the Au-rich melt has little help in wet-ting of the Mo substrate. It is in accordance to the experimental observations.

The microstructure of the infrared brazed joint at 1100C for 180 s is very similar to that of infrared brazed one at 1050C. However, the microstructure of the infrared brazed joint at 1100C for 300 s is close to that of furnace brazed specimen. In addition to the Au-rich and Ni-rich phases, the interfacial MoNi inter-metallic compound is extensively observed in Fig. 5(f). Strong metallurgical reactions between the braze and substrate are widely proceeded. Accordingly, its

micro-structure is similar to the furnace brazed specimen as compared between Fig. 5(f) andFig. 2.

Table 2 displays shear strengths of various brazed

specimens. Based on the table, all infrared brazed joints are fractured at the Mo substrate, so there is no accurate shear strength available for these infrared brazed joints. The shear strength of infrared brazed specimen shown in

Table 2 reflects the shear strength of Mo substrate

in-stead of the brazed joint. The fluctuation of experimental data for the infrared brazed specimen is probably caused by the inherent brittleness and/or defects of the Mo sub-strate at room temperature. In general, the shear strength of infrared brazed joint is higher than that of the furnace brazed one. Fig. 6shows the SEM BEI cross-sectional image and SEI fractograph of the infrared brazed joint at 1050C for 180 s after the shear test. The braze alloy is free of crack as indicated by the arrow inFig. 6(a). It is clear that cracks initiate and propagate along the Mo substrate, and quasi-cleavage fracture is widely observed in the fractograph as illustrated inFig. 6(b).

Only the furnace brazed specimen is fractured along the brazed joint, and its average shear strength is 110.4 MPa. Fig. 7 shows the SEM BEI cross-sectional image and SEI fractograph of the furnace brazed joint at 1050C for 1200 s after shear tests. Based on Fig. 7(a), the furnace brazed joint is primarily fractured along the interfacial angular MoNi intermetallics. Cleavage dominated fracture is extensively observed in the fractured surface as illustrated inFig. 7(b). The huge angular MoNi phase with cracks is further identified in the EDS chemical analysis as indicated by arrows in the figure. Consequently, the formation of interfacial MoNi intermetallic compound in the brazed joint is detrimen-Fig. 3. Au–Ni–Pd ternary alloy phase diagrams in atomic percent: (a)

liquidus projection and (b) isothermal section at 300C[13].

Fig. 4. 10 wt% Pd vertical section of Au–Ni–Pd pseudobinary phase diagram[13].

(5)

tal to its bonding strength. Additionally, the application of rapidly infrared brazing is a very effective way to

inhibit excessive growth of the interfacial MoNi phase in the brazed joint.

Fig. 5. The SEM images and EDS chemical analysis results of infrared brazed joints: (a) BEI, 1050 for 90 s, (b) BEI, 1050C for 180 s; (c) BEI, 1050C for 300 s; (d) BEI, 1100 C for 180 s; (e) SEI and (f) BEI, 1100 C for 300 s.

Table 2

Shear strengths of various brazed specimens

Brazing type Temperature (C) Time (s) Shear strength (MPa) Average Shear strength (MPa) Standard deviation (MPa)

Furnace 1050 1200 119.2 110.4 8.9

1200 101.5

Infrared 1050 180 511 (fracture of the Mo substrate) Infrared 1050 300 303 (fracture of the Mo substrate) Infrared 1100 180 469 (fracture of the Mo substrate) Infrared 1100 300 390 (fracture of the Mo substrate)

(6)

4. Summary

Microstructural evolution of the brazed Mo joint using the 70Au–22Ni–8Pd braze alloy has been accessed in this study. The molten braze can effectively wet the Mo substrate above 1070C. Both Au-rich and Ni-rich phases are observed for the infrared brazed joint at 1050C. The Au-rich liquid is not reacted with the Mo substrate. In contrast, the Ni-rich liquid preferen-tially reacts with the Mo. With increasing the brazing temperature and/or time, the interfacial reaction be-tween the molten braze and Mo substrate is significantly increased, and the huge angular MoNi intermetallic compound is formed in the brazed joint, especially for the furnace brazed specimen due to its slow thermal his-tory. In general, the shear strength of infrared brazed joint is higher than that of the furnace brazed one. All infrared brazed joints are fractured at the Mo sub-strate, so there is no shear strength available for these infrared brazed specimens. Only the furnace brazed specimen is fractured along the brazed joint, and its average shear strength is 110.4 MPa. The formation of interfacial MoNi intermetallic compound in the brazed joint is detrimental to its bonding strength. Cleavage dominated fracture is widely observed in fractographs

of the furnace brazed specimen. Additionally, the appli-cation of rapid infrared brazing is a very effective way to inhibit excessive growth of the MoNi phase in the brazed joint.

Acknowledgment

The authors gratefully acknowledge the financial sup-port of this research by the National Science Council (NSC), Republic of China under NSC grant 93-2216-E-002-028.

References

[1] Smith WF. Structure and properties of engineering alloys. 2nd ed. New York: McGraw-Hill Inc; 1993.

[2] Davis JR. 9th ed. Metals handbook, vol. 2. Materials Park: ASM International; 1990.

[3] Schwartz M. Brazing: for the engineering technologist. Materials Park: ASM International; 1995.

[4] Chan HY, Shiue RK. The study of brazing Ti–6Al–4V and TZM alloy using pure silver. J Mater Sci Lett 2003;22:1659–63. [5] Chan HY, Liaw DW, Shiue RK. Microstructural evolution of

brazing Ti–6Al–4V and TZM using silver-based braze alloy. Mater Lett 2004;58:1141–6.

Fig. 6. The SEM: (a) BEI cross-sectional image and (b) SEI fractograph of the infrared brazed Mo/70Au–22Ni–8Pd/Mo joint at 1050C for 180 s after the shear test.

Fig. 7. The SEM: (a) BEI cross-sectional image and (b) SEI fractograph of the furnace brazed Mo/70Au–22Ni–8Pd/Mo joint at 1050C for 1200 s after the shear test.

(7)

[6] Chan HY, Liaw DW, Shiue RK. The microstructural observation of brazing Ti–6Al–4V and TZM using the BAg-8 braze alloy. Int J Refract Met Hard Mater 2004;22:27–33.

[7] Hiraoka Y, Nishikawa S. Joining of single-crystalline molybde-num and carbon–ceramics by using palladium and palladium– silver alloy as brazing metal. Int J Refract Met Hard Mater 1996;14:311–7.

[8] Hiraoka Y. Brazing of single-crystalline molybdenum by using Pd–20%Ag alloy. Refract Met Hard Mater 1992;11:303–7. [9] Hidaka Y. High-temperature annealing embrittlement of

single-crystalline molybdenum joined by using Mo–40%Ru alloy. Hard Mater 1992;11:89–95.

[10] Shiue RK, Wu SK, Chen SY. Infrared brazing of TiAl interme-tallic using BAg-8 braze alloy. Acta Mater 2003;51:1991–2004. [11] Shiue RK, Wu SK, Hung CM. Infrared repair brazing of 403

stainless steel with a nickel based braze alloy. Metall Mater Trans 2002;33A:1765–73.

[12] Shiue RK, Wu SK, Chen SY. Infrared brazing of TiAl using Al-based braze alloys. Intermetallics 2003;11:661–71.

[13] Villars P, Prince A, Okamoto H. Handbook of ternary alloy phase diagrams. Materials Park: ASM International; 1995.

[14] Massalski TB. Binary alloy phase diagrams. Materials Park: ASM International; 1990.

數據

Fig. 2 displays the SEM secondary electron image (SEI), backscattered electron image (BEI) and chemical analysis results of the furnace brazed Mo/70Au–22Ni–
Fig. 5 shows SEM images and EDS chemical analysis results of infrared brazed joints with various brazing conditions
Fig. 3. Au–Ni–Pd ternary alloy phase diagrams in atomic percent: (a) liquidus projection and (b) isothermal section at 300 C [13].
Fig. 5. The SEM images and EDS chemical analysis results of infrared brazed joints: (a) BEI, 1050 for 90 s, (b) BEI, 1050 C for 180 s; (c) BEI, 1050 C for 300 s; (d) BEI, 1100 C for 180 s; (e) SEI and (f) BEI, 1100 C for 300 s.
+2

參考文獻

相關文件

We do it by reducing the first order system to a vectorial Schr¨ odinger type equation containing conductivity coefficient in matrix potential coefficient as in [3], [13] and use

Let f being a Morse function on a smooth compact manifold M (In his paper, the result can be generalized to non-compact cases in certain ways, but we assume the compactness

substance) is matter that has distinct properties and a composition that does not vary from sample

To stimulate creativity, smart learning, critical thinking and logical reasoning in students, drama and arts play a pivotal role in the..

 Promote project learning, mathematical modeling, and problem-based learning to strengthen the ability to integrate and apply knowledge and skills, and make. calculated

Wang, Solving pseudomonotone variational inequalities and pseudocon- vex optimization problems using the projection neural network, IEEE Transactions on Neural Networks 17

Define instead the imaginary.. potential, magnetic field, lattice…) Dirac-BdG Hamiltonian:. with small, and matrix

* School Survey 2017.. 1) Separate examination papers for the compulsory part of the two strands, with common questions set in Papers 1A & 1B for the common topics in