Trans. Nonferrous Met. Soc. China 22(2012) 596-603

Development of Al-12Si-xTi system active ternary filler metals for Al metal matrix composites

ZHANG Gui-feng^{1, 2}, SU Wei¹, ZHANG Jian-xun¹, A. SUZUMURA³

1. State Key Laboratory for Mechanical Behavior of Materials, Xi’an Jiaotong University, Xi’an 710049, China;

2. State Key Laboratory of Advanced Welding and Joining, Harbin Institute of Technology, Harbin 150001, China;

3. Department of Mechanical and Aerospace System Engineering, Tokyo Institute of Technology, Japan

Received 4 March 2011; accepted 26 June 2011

Abstract: To improve the wettability of Al metal matrix composites (Al-MMCs) by common filler metals, Al-12Si-xTi (x=0.1, 0.5, 1, 3.0; mass fraction, %) system active ternary filler metals were prepared. It was demonstrated that although the added Ti existed within Ti(Al_1-xSi_x)₃ (0≤x≤0.15) phase, the shear strength and shear fracture surface of the developed Al-12Si-xTi brazes were quite similar to those of traditional Al-12Si braze due to the presence of similar microstructure of Al-Si eutectic microstructure with large volume fraction. So, small Ti addition (~1%) did not make the active brazes brittle and hard compared with the conventional Al-12Si braze. The measured melting range of each Al-12Si-xTi foil was very similar, i.e., 580-590 °C, because the composition was close to that of eutectic. For wettability improvement, with increasing Ti content, the interfacial gap between the Al₂O₃ reinforcement and filler metal (R/M) could be eliminated, and the amount of the remainder of the active fillers on the composite substrate decreased after sessile drop test at 610 °C for 30 min. So, the wettability improvement became easy to observe repeatedly with increasing Ti content. Additionally, the amount and size of Ti(AlSi)₃ phase were sensitive to the Ti content (before brazing) and Si content (after brazing).

Key words: Al metal matrix composites; brazing; transient liquid phase bonding; wettability; filler metal

1 Introduction

Compared with fusion welding processes, both transient liquid phase (TLP) diffusion bonding and brazing do not involve the melting of the base metals. So, the problems of reinforcement-metal (R/M) reactions, reinforcement destruction, etc, can be reduced [1]. Currently, TLP bonding and brazing for Al metal matrix composites (Al-MMCs) are generally accepted as commercial practicable methods instead of fusion welding processes. The principal factor governing the interfacial microstructure and properties of TLP bonded or brazed joints is the alloying system selection and corresponding composition design of interlayer or brazing filler metal, which determines the removal of oxide film, particle segregation and wettability of Al-MMCs by filler metals. In previous reports, Al-12Si brazing filler metal [2] and Zn-Al eutectic solder [3-5] have been used in the studies of brazing of Al-MMCs, while several pure materials such as Cu foil [6-9], Ag foil [10] and Si powders [11] were used as interlayer in TLP bonding of Al-MMCs or monolithic Al to effectively remove oxide film on Al matrix surface by undermining mechanism [4, 12].

As improved interlayers, Al-Cu alloy interlayer [13] and Cu/Al/Cu sandwich interlayer [14], Cu/Ni/Cu [15] sandwich interlayer, and Al-Si-Ti-SiC mixed powders [16] were used to reduce particle segregation, to reduce bonding temperature, and to get a bond seam with reinforcement, respectively. On the other hand, as improved process for TLP bonding, such as temperature gradient TLP bonding (TG-TLP) [17], TLP bonding followed by isostatic pressing [18], mechanical action assisted TLP bonding (introducing ultrasonic vibration [19], electric [20] or impact [21]), semi-solid state TLP bonding [22,23], and quasi-melting [24,25] were attempted to make the profile of bond line flexural, to improve the interface densification with a minimal deformation, or to effectively remove tenacious oxide film on Al matrix surface.

However, the popularly used Al-12Si filler metal [2], Al-Si-Mg filler metal [26] and Cu interlayer [27] showed poor wettability at R/M micro-interfaces, especially for Al₂O₃ reinforcing particulates, resulting in interfacial gap [2] and interfacial debonding fracture mode at R/M micro-interfaces [27, 28]. Recently, a novel process of active-TLP bonding (A-TLP) was proposed by ZHANG et al [27] to improve the wettability at R/M micro-interfaces, in which the interlayer must contain both melting point depressant (MPD) and active element for ceramic reinforcement (e.g., Ti). Despite numerous efforts made to overcome a specific problem, a generally accepted alloying system selection route and corresponding composition design route for the interlayer or brazing filler metal have not been established.

It should be pointed out that a binary interlayer consisting of MPD and active filler metal (e.g., Ti) can not reduce particle segregation and simultaneously can not improve wettability at R/M interfaces. Therefore, in the present work, Al-12Si-xTi (x=0.1, 0.5, 1.0, 3.0; mass fraction, %) system active ternary filler metals were prepared to reduce particle segregation and simultaneously to improve the wettability of particulate ceramic reinforcement by traditional Al-12Si filler. Their composition design route, melting range, microstructures and properties (including microhardness and shear property), and wettability were described and compared to optimize alloying system selection and corresponding composition design.

2 Experimental

As mentioned above, according to the composition design route of interlayer for A-TLP bonding of Al-MMCs [26, 27], the active ternary filler metal of Al-Si-Ti system was developed, in which Al was added to reduce particle segregation, Si was selected as MPD to remove the oxide film on Al matrix of Al-MMC by undermining, and Ti was added to improve the wettability for Al-MMC through introducing the reactions at filler metal/matrix (M/M) and/or at R/M interface.

To improve the reactive wettability between the active filler metal and Al₂O₃ ceramic reinforcement at TLP bonding temperature (generally about 600 °C), the added Ti should be in molten state, but not in solid state, because the added Ti atoms in liquid state will have high activity and distribute uniformly. However, the high melting point of Ti (1668 °C) and the absence of eutectic reaction in Al-Ti binary system imply that the added Ti may be refractory as a “melting point raiser” for Al matrix. Therefore, the Ti content should be selected carefully to make the added Ti atoms molten and uniform during the A-TLP bonding.

Based on the liquidus projection of Al-Si-Ti ternary phase diagram at Al-rich corner (see Fig. 1) [29,30], in which a Al-Si-Ti ternary eutectic reaction exists for the composition of Al-12Si-1Ti at 573 °C (described as: L?(Al)+(Si)+τ₁), a series of Al-12Si-xTi (x=0.1, 0.5, 1.0, 3.0) system active ternary filler metals with a composition very close to the eutectic composition of Al-12Si-1Ti were selected to completely (or partially) make the added Ti atoms molten and uniform in molten interlayer during A-TLP bonding. So, the reactive wetting both at M/M and R/M micro-interfaces may be expected. The used braze ingot was obtained by melting three kinds of commercially available blocks of pure Al, Al-20Si and Al-5Ti master alloys in flowing argon with high frequency induction heating. The resultant melt was kept at 830 °C for 30 min, and then cooled in situ to room temperature. And their foils were prepared by re-melting the in-situ solidified ingot and then rapid cooling.

Fig. 1 Liquidus projection of Al-Si-Ti ternary phase diagram at Al-rich corner showing the presence of ternary eutectic reaction [29, 30] (p: Peritectic reaction; e: Eutectic reaction; u: Transition reaction)

As described in detail in a previous studies [10], the used Al-MMC produced by pressure casting method had pure aluminum matrix (to simplify element analysis) and short fiber reinforcement consisting of Al₂O₃ (85%, mass fraction) and SiO₂ (15%). For convenience, the used Al MMC was marked by Al₂O_3sf/Al in this paper. The average diameter and average length of the short fiber reinforcement were 3.6 μm and 140 μm, respectively, and the volume fraction of the short fiber reinforcement in the Al₂O_3sf/Al composite was 30% [10]. The as-received Al₂O_3sf/Al composite block was cut into pieces of 15 mm×15 mm×2 mm. Before each test, the piece surface was polished with 600 emery paper, and then ultrasonically degreased in acetone.

The microstructures and properties of the ingot filler metals (including shear strength and microhardness for different phases) were examined. Especially, the existing form and morphology of Ti in filler metals and the effect of Ti addition on the microstructure and property of filler metals were examined by back-scattered electron (BSE) images and by energy dispersive X-ray (EDX) analysis. The melting range of each Al-12Si-xTi (x=0, 0.1, 0.5, 1.0, 3.0) foil was measured with differential scanning calorimetry (DSC). According to the measured melting range, the wettability of acceptable foils on Al₂O_3sf/Al composite was tested by sessile drop method.

3 Results and discussion

3.1 Microstructure of Al-Si-Ti system filler metals and existing form of active element Ti

Figure 2 shows the BSE images of the cross-section of in situ solidified ingots of Al-12Si, Al-12Si-0.1Ti, Al-12Si-0.5Ti and Al-12Si-1Ti, respectively. The microstructures of them are quite similar, mainly consist of α(Al)-Si eutectic. However, with increasing Ti content from 0.1% to 1%, a flake-like white phase is present considerably, which must contain Ti with an atomic number of 22, much higher than that of Al (13) or Si (14). So, the microstructure of Al-12Si-xTi ingot consists of α(Al)-Si eutectic (similar to Fig. 2(a)) and a flake-like phase containing Ti, and the added Ti should be almost within the precipitated phase. This result is consistent with the reports by ZEREN and KARAKULAK [31, 32].

To identify the kind of the white precipitated phase containing Ti and to further investigate the existing form of active element Ti, the composition of the white precipitated phase containing Ti was tested. EDX spot analysis (see Figs. 2(c) and (d)) results show that the light flake-like white phase has a composition of 30% (mole fraction) Ti, 58% Al and 12% Si, which could be identified as a ternary compound of Ti₃Al₅Si. So, the flake-like white precipitated phase belongs to a Al-rich phase of Ti(Al_1-xSi_x)₃ (0≤x≤0.15), commonly written as Ti(AlSi)₃ or t₃ [32], in which up to 15% Al can be replaced by silicon in TiAl₃ lattice structure, resulting in various chemical compositions and a range of lattice parameters. It can also be seen from Fig. 2 that the size of the present Ti(AlSi)₃ phase is larger than that of Si phase, and both the number and size of the precipitated Ti(AlSi)₃ phase increase with increasing Ti content. For Al-12Si-1Ti, the size of Ti(AlSi)₃ phase is 100-200 μm in length and 10-20 μm in width within the cross-section (see Fig. 2(d)). So, large size of the Ti(AlSi)₃ phase is not beneficial to the ductility of the braze and leads to nonuniform distribution of Ti (especially for small content of about 1%) at each micro-area. So, the filler metal should be used in foil obtained by rapid cooling process to reduce precipitation and to improve the distribution of Ti in solid filler metal.

To further investigate the mechanical property of the Ti(Al_1-xSi_x)₃ phase (0≤x≤0.15), the Vickers microhardness testing was carried out on distinct phases including Al, Si, and Ti(Al_1-xSi_x)₃ in the Al-12Si-1Ti ingot at a load of 24.5 mN for a dwell time of 10 s (HV_25g). Figure 3 shows the indentation on distinct phases in Al-12Si-1Ti ingot and corresponding Vickers microhardness, respectively. As expected, the Ti(Al_1-xSi_x)₃ phase is much harder (HV_25g 363) than Al substrate (HV_25g 56). On the other hand, micro crack is observed only on Si phase, showing that the hard Ti(Al_1-xSi_x)₃ phase is not brittle as Si phase. In fact, the Ti(AlSi)₃ phase may be a promising reinforcement for Al-Si-xTi cast alloy, which is used as a wear-resistant material [32]. Moreover, it is interesting that some Si is in contact with the Ti(Al_1-xSi_x)₃ phase, suggesting that some Si particles possibly nucleate at Ti(AlSi)₃.

Fig. 2 BSE images of in situ solidified filler metal ingots: (a) Al-12Si; (b) Al-12Si-0.1Ti; (c) Al-12Si-0.5Ti; (d) Al-12Si-1Ti

Fig. 3 Micro-indentation test on distinct phases in Al-12Si-1Ti ingot (a) and their microhardness (24.5 mN, 10 s) (b)

3.2 Shear strength and fracture surface of Al-12Si-xTi ingots

Figure 4 shows the shear strength values of machined ingot samples of Al-12Si and Al-12Si-xTi (x=0.1, 0.5, 1.0) with the same size (cross section: 5 mm×5 mm; length: 15 mm). The shear strength values of Al-12Si-xTi (x=0.1, 0.5, 1) samples are quite similar to each other, and are also quite similar to that of the Al-12Si sample. The measured shear strength, with small scatter, is about 100 MPa. Figure 5 shows the corresponding fracture surfaces of Al-12Si and Al-12Si-xTi (x=0.1, 0.5, 1.0) samples after shear test. Like Al-12Si sample, significant plastic deformation marks can be observed on the shear fracture surface of each developed Al-12Si-xTi samples.

Fig. 4 Shear strength of Al-12Si and Al-12Si-xTi (x=0.1, 0.5, 1.0) filler metals ingots

Fig. 5 Shear fracture surfaces of Al-12Si (a), Al-12Si-0.1Ti (b), Al-12Si-0.5Ti (c) and Al-12Si-1Ti (d) ingots

Based on the fact that both shear strength and shear fracture surface of each Al-12Si-xTi (x=0.1, 0.5, 1) are quite similar to those of the Al-12Si ingot, it can be deduced that small Ti addition (~1%) does not make the novel brazes brittle and hard compared with the conventional Al-12Si braze. The similar shear strength and similar shear fracture surface should be a result of similar microstructure. Additionally, it should be noted that although small addition of Ti does not affect shear strength and shear fracture surface of the braze ingots significantly, and the amount of Ti(AlSi)₃ phase is sensitive to the Ti content.

3.3 Melting temperature range of braze foils

To make the distribution of small added Ti uniform via reducing the gravity segregation and to make the pre-placement of the developed fillers convenient, the Al-12Si-xTi (x=0, 0.1, 0.5, 1.0, 3.0) master alloys were remelted and then rapidly cooled to fabricate braze foils. All of them showed good ability to form brazing foil during rapid cooling process.

Figure 6 shows the DSC curves of conventional Al-12Si foil and the developed Al-12Si-xTi foils (x=0.1, 0.5, 1.0, 3.0). Like Al-12Si foil, the developed Al-12Si-xTi (x=0.1, 0.5, 1, 3) braze foils also exhibit only one endothermic peak, showing that they should be near ternary eutectic filler metal. It can be seen that the solidus and liquidus temperatures of the developed Al-12Si-xTi filler metal are about 580 °C and 590 °C, respectively. Compared with the melting point of Al-12Si foil, the solidus and liquidus temperatures do not increase when Ti is up to 1%. When the Ti content increases even to 3%, no significant increase in melting range is observed, showing that the effect of Ti content on melting point for Al-Si-Ti ternary is not as sensitive as that for Al-Ti binary system, in which the melting point of Al-Ti binary alloy sharply increases with small Ti addition. Therefore, when bonding temperature is above 600 °C, the added Ti should be in liquid state, and thus be present uniformly at whole joint interface and further can easily react with oxide film on Al matrix surface and/or ceramic reinforcement at whole joint interface.

Fig. 6 DSC curves of Al-12Si foil and Al-12Si-xTi foils (x=0.1, 0.5, 1.0, 3.0)

3.4 Wettability test

Figure 7 shows the interfacial microstructure (BSE images) of the cross section of the wettability testing samples for Al-12Si foil and Al-12Si-xTi foils (x=0.1, 0.5, 1.0) after sessile drop test at 610 °C for 30 min. No particulate segregation occurs for Al-12Si. For conventional Al-12Si foil, a continuous layer slightly thinner than initial thickness of the foil (100 μm) remains on the surface of the composite substrate due to limited penetration of the molten braze into the composite; moreover, although the interfacial gap at M/M micro-interfaces disappears, the interfacial gaps at R/M micro-interfaces could not be eliminated, especially at the areas with numerous short alumina fibers. Generally speaking, the Al₂O_3sf/Al composite will show poorer wettability because the pure Al matrix does not contain active Mg element and the Al₂O₃ reinforcement is more stable than SiC in contact with molten Al.

In contrast, in most cases, the interfacial gaps at R/M micro-interfaces could be eliminated finally when small Ti is added, even less than 1%, showing that the developed Al-12Si-xTi active braze could react with Al₂O₃ reinforcement to some extent compared with traditional Al-12Si braze; moreover, the amount of the remainder of the active braze foil decreases with increasing Ti content, showing that with increasing Ti content, the penetration of the Si atoms in the molten brazes into the composite substrate could occur earlier. However, since the Si atoms in the molten braze foil penetrate into the composite substrate earlier with increasing Ti content, resulting in significant decrease in Si content in the remainder of braze foil, the Ti(AlSi)₃ phase is present significantly on the surface of remainder of braze foil with increasing Ti content.

For example, Al-12Si-1Ti braze foil completely penetrates into the Al₂O_3sf/Al composite, except several light flake-like phases of Ti(AlSi)₃ having a composition of 28% Ti, 57% Al and 15% Si. The significant penetration and resultant increase in Ti(AlSi)₃ area fraction show that the developed Al-12Si-1Ti filler metal foil could react with the Al₂O_3sf/Al composite earlier than conventional Al-12Si braze foil, thus earlier creating some penetration paths without oxide film barrier for Si and then accelerating subsequent undermining. Moreover, bare Al₂O₃ reinforcements are too hard to observe on the whole surface (with the overall width of 5 mm) after sessile drop test. The results show that the Al₂O₃ reinforcements on initial surface could be covered by the Al-12Si-1Ti filler metal without interfacial gaps. Therefore, the wettability between R/M micro-interface can be improved by Ti addition to traditional Al-12Si filler metal.

Fig. 7 Interfacial microstructures (BSE images) of cross section of wettability testing samples after sessile drop test at 610 °C for 30 min: (a) Al-12Si; (b) Al-12Si-0.1Ti; (c) Al-12Si-0.5Ti; (d) Al-12Si-1Ti

It should be noticed that although Ti hardly diffuses into Al matrix of the Al₂O_3sf/Al composite and appears only on the upper surface in the form of flake-like Ti(AlSi)₃ phase, the maximum size of the Ti(AlSi)₃ phase become to be about 25 μm in length and 5 μm in width after wettability test. Especially, the length of the Ti(AlSi)₃ phase decreases from 100-200 μm (in initial solid ingot) to 20-30 μm due to the decease in Si content. Moreover, the flake-like Ti(AlSi)₃ phases are found to be parallel to the surface of the composite substrate after sessile drop test. If the arranging direction, size and volume fraction of the Ti(AlSi)₃ phase with high hardness (HV_25g 363) could be controlled properly, the Ti(AlSi)₃ phase may be used as a reinforcement to enhance the joint seam.

4 Conclusions

1) The added Ti existed within a flake-like precipitated phase of Ti(AlSi)₃, which had a length of 100-200 μm (much longer than eutectic Si phase) and a microhardness of HV_25g 363 (much higher than Al phase of HV_25g 56). The microstructure of Al-12Si-xTi (x=0.1, 0.5, 1.0) ingot consisted mainly of α(Al)-Si eutectic with Ti(AlSi)₃ phase, and both the number and size of the precipitated Ti(AlSi)₃ phase increased with increasing Ti content.

2) Compared with the traditional filler metal of Al-12Si, although Ti(AlSi)₃ was present, the melting temperature range (580-590 °C), shear strength (~100 MPa) and ductile shear fracture surface of each Al-12Si-xTi (x=0.1, 0.5, 1.0) braze ingots were quite similar to that of the Al-12Si ingot. The results showed that small Ti addition did not increase the melting point or make the novel brazes brittle. These similar characteristics should be a result of similar microstructure.

3) Compared with the traditional Al-12Si braze, in most cases, the interfacial gaps at R/M micro-interfaces could be eliminated finally when small Ti was added, even less than 1%, showing that the developed Al-12Si-xTi active braze ingots could react with Al₂O₃ reinforcement to some extent; moreover, the amount of residual filler decreased with increasing Ti content, showing that the penetration of the Si atoms in the molten brazes into the composite substrate could occur earlier with increasing Ti content.

4) Because the Si atoms in the molten braze foil penetrated into the composite substrate earlier with increasing Ti content, resulting in significant decrease in Si content in the remainder of braze foil, the number of Ti(AlSi)₃ phase increased after sessile drop test at 610 °C for 30 min, while the size of Ti(AlSi)₃ phase decreased. It should be noted that although small addition of Ti did not affect shear strength and shear fracture surface of the braze ingots significantly, the amount and size of Ti(AlSi)₃ phase were sensitive to the Ti content (before brazing) and Si content (after brazing).

References

[1] ELLIS M B D. Joining of aluminum based metal matrix composites [J]. International Materials Reviews, 1996, 41(2): 41-58.

[2] WENG W P, CHUANG T H. Interfacial characteristic for brazing of aluminum matrix composites with Al-12Si filler metals [J]. Metallurgical and Materials Transactions A, 1997, 28(12): 2673-2682.

[3] YAN J C, XU H B, SHI L, WANG X L, YANG S Q. Vibration assisted brazing of SiC_p/A356 composites: Microstructure and mechanical behaviour [J]. Science and Technology of Welding and Joining, 2008, 13(8): 760-764.

[4] XU Z W, YAN J C, CHENG W, YANG S Q. Substrate oxide undermining by a Zn–Al alloy during wetting of alumina reinforced 6061 Al matrix composite [J]. Materials Chemistry and Physics, 2008, 112: 831-837.

[5] URENA A, GIL L, ESCRICHE E, de GOMEZ S J M, ESCALERA M D. High temperature soldering of SiC particulate aluminum matrix composites (series 2000) using Zn-Al filler alloys [J]. Science and Technology of Welding and Joining, 2001, 6(1): 1-11.

[6] ZHAI Y, NORTH T H. Counteracting particulate segregation during transient liquid-phase bonding of MMC–MMC and Al₂O₃–MMC joints [J]. Journal of Materials Science, 1997, 32: 5571-5575.

[7] MAITY J, PAL T K, MAITI R. Transient liquid phase diffusion bonding of 6061-15 wt% SiCp in argon environment [J]. Journal of Materials Processing Technology, 2009, 209: 3568-3580.

[8] ZHANG H T, CAO J, LU H. Reactive brazing of aluminium to aluminium-based composite reinforced with alumina borate whiskers with Cu interlayer [J]. Vacuum, 2010, 84: 474-477.

[9] WU M F, YU C, PU J. Study on microstructures and grain boundary penetration behaviours in contact reactive brazing joints of 6063 Al alloy [J]. Materials Science and Technology, 2008, 24(12): 1422-1426.

[10] SUZNMURA Akio, XING Yu-jin. Diffusion brazing of short Al₂O₃ fiber-reinforced aluminum composite [J]. Materials Transactions, JIM, 1996, 37(5): 1109-1115.

[11] HE Peng, FENG Ji-cai, QIAN Yi-yu, HUANG Zhen-feng, MAI Hen-hui. Resolve of fragile and defect problem for a new contact reaction brazing aluminum to stainless steel [J]. Materials Science and Technology, 2005, 13(1): 82-85. (in Chinese)

[12] GALE W F, BUTTS D A. Transient liquid phase bonding [J]. Science and Technology of Welding and Joining, 2004, 9(4): 283-301.

[13] LIU Wei-hong, SUN Da-qian, JIA Shu-sheng, QIU Xiao-ming. Transient liquid phase bonding of aluminium-based metal matrix composite with Al-Cu alloy interlayer [J]. Transactions of the China Welding Institution, 2003, 24(5): 13-16. (in Chinese)

[14] LIU Wei-hong, SUN Da-qian, SUN De-xin, JIA Shu-sheng. Transient liquid phase bonding of Al₂O_3p/6061 Al metal matrix composite [J]. Transactions of the China Welding Institution, 2007, 28(3): 57-60. (in Chinese)

[15] YAN J C, XU Z W, Wu G H, YANG S Q. Interface structure and mechanical performance of TLP bonded joints of Al₂O_3p/6061Al composites using Cu/Ni composite interlayers [J]. Scripta Materialia, 2004, 51: 147-150.

[16] HUANG J H, DONG Y L, WAN Y, ZHAO X K, ZHANH H. Investigation on reactive diffusion bonding of SiC_p/6063 MMC by using mixed powders as interlayers [J]. Journal of Materials Processing Technology, 2007, 190: 312-316.

[17] SHIRZADI A A, WALLACH E R. Temperature gradient transient liquid phase diffusion bonding: A new method for joining advanced materials [J]. Science and Technology of Welding and Joining, 1997, 2(3): 89-94

[18] SHIRZADI A A, WALLACH E R. New approaches for transient liquid phase diffusion bonding of aluminium based metal matrix composites [J]. Material Science and Technology, 1997, 13(2): 135-142.

[19] YAN Jiu-chun, XU Zhi-wu, SHI Lei, MA Xing, YANG Shi-qin. Ultrasonic assisted fabrication of particle reinforced bonds joining aluminum metal matrix composites [J]. Materials and Design, 2011, 32: 343-347.

[20] YU Z S, LI R F, QI K. Joining of SiC particle reinforced aluminium metal matrix composites by electromagnetic field aided brazing method [J]. Materials Science and Technology, 2010, 26(6): 695-698.

[21] GUO Wei, HUA Meng, HO J, KIN L. Study on liquid-phase-impact diffusion welding SiCp/ZL101 [J]. Composites Science and Technology, 2007, 67: 1041-1046.

[22] LIU Li-ming, NIU Ji-tai, TIAN Yang-hong, MENG Xiao-dong. Diffusion bonding mechanism and microstructure of welded joint of aluminium matrix composite Al₂O_3p/6061 [J]. Transactions of Nonferrous Metals Society of China, 1999, 9(4): 826-830.

[23] LI Jing-long, XIONG Jiang-tao, ZHANG Fu-sheng. Effect of thermal cycles on interface evolution of vacuum diffusion bonded aluminum alloy 2A14 [J]. Transaction of the China Welding Instruction, 2007, 28(3): 41-47. (in Chinese)

[24] NAGAOKA T, MORISADA Y, FUKUSUMIA M, TAKEMOTO T. Joint strength of aluminum ultrasonic soldered under liquidus temperature of Sn–Zn hypereutectic solder [J]. Journal of Materials Processing Technology, 2009, 209: 5054–5059.

[25] NAGAOKA T, MORISADA Y, FUKUSUMI M, TAKEMOTO T. Ultrasonic-assisted soldering of 5056 aluminum alloy using quasi-melting Zn-Sn alloy [J]. Metallurgical and Materials Transactions B, 2010, 41: 864-871.

[26] ZHANG X P, QUAN G F, WEI W. Preliminary investigation on joining performance of SiCp-reinforced aluminium metal matrix composite (Al/SiC_p–MMC) by vacuum brazing [J]. Composites Part A: Applied Science and Manufacturing, 1999, 30(6): 823-827.

[27] ZHANG G F, ZHANG J X, PEI Y, LI S Y, CHAI D L. Joining of Al₂O_3p/Al composites by transient liquid phase (TLP) bonding and a novel process of active-transient liquid phase (A-TLP) bonding [J]. Materials Science and Engineering A, 2008, 488: 146-156.

[28] ZHANG Gui-feng, ZHANG Jian-xun, LI Si-yu, WEI Zhong-xin. Novel process of active transient liquid phase (A-TLP) bonding for aluminum matrix composite [J]. Journal of Xi’an Jiaotong University, 2009, 43(11): 71-74. (in Chinese)

[29] RAGHAVAN V. Al-Si-Ti (aluminum-silicon-titanium) [J]. Journal of Phase Equilibria and Diffusion, 2009, 30: 82-83.

[30] LIU S, WEITZER F, SCHUSTER J C, KRENDELSBERGER N, DU Y. On the reaction scheme and liquidus surface in the ternary system Al-Si-Ti [J]. International Journal of Materials Research, 2008, 99(7): 705-711.

[31] ZEREN M, KARAKULAK E. Influence of Ti addition on the microstructure and hardness properties of near-eutectic Al-Si alloys [J]. Journal of Alloys and Compounds, 2008, 450: 255-259.

[32] ZEREN M, KARAKULAK E. Microstructural characterization of Al-Si-xTi cast alloys [J]. Materials Science and Technology, 2009, 25(10): 1211-1214.

铝基复合材料用Al-12Si-xTi系三元活性钎料的制备

张贵锋^{1, 2}，苏伟¹，张建勋¹，A. SUZUMURA³

1. 西安交通大学 金属材料强度国家重点实验室，西安 710049；

2. 哈尔滨工业大学 先进焊接与连接国家重点实验室，哈尔滨 150001；

3. Department of Mechanical and Aerospace System Engineering, Tokyo Institute of Technology, Japan

摘  要：为了改善常用钎料对铝基复合材料的润湿性，制备了新型Al-12Si-xTi系三元活性钎料。研究表明，所添加的Ti以Ti(Al_1-xSi_x)₃(0≤x≤0.15)相存在，但由于钎料的主要组织均为Al-Si共晶组织，所以Al-12Si-xTi钎料与传统Al-12Si钎料具有相似的剪切强度和剪切断口。因此，加入少量Ti(~ 1%)不会使钎料脆化和硬化；而且Al-12Si-xTi钎料带的成分接近共晶成分，所以熔化区间非常相似，为580~590 °C。对于润湿性的改善，是由于随着钎料中Ti含量的增加，Al₂O₃增强相和钎料金属间的(R/M)界面间隙得到消除的原因。润湿实验后在铝基复合材料表面残留的钎料量也随着钎料含Ti量的增加而减少。因此，Ti对于润湿性的改善可以随着钎料含Ti量的增加而较为容易地被重复观察到。此外，Ti(AlSi)₃的量和尺寸对含Ti量(焊接前)和含Si量(焊接后)敏感。

关键词：铝基复合材料；钎焊；过渡液相扩散焊；润湿性；钎料

(Edited by YANG Hua)

Foundation item: Project (50875199) supported by the National Natural Science Foundation of China; Project supported by State Key Laboratory of Advanced Welding and Joining, China

Corresponding author: ZHANG Gui-feng; Tel: +86-29-82663115; E-mail: gfzhang@mail.xjtu.edu.cn

DOI: 10.1016/S1003-6326(11)61219-3