Microstructures in interface region and mechanical behaviours of friction stir lap Al6060 to Ti–6Al–4V welds

Abstract

There has been uncertainty about the best pin position for friction stir lap welding (FSLW) of Al-to-Ti so that weld samples can achieve the highest attainable load during tensile-shear testing. In this study, tests were conducted to evaluate how much tensile-shear load an Al6060 to Ti–6Al–4V (6060-Ti64) FSLW sample of a set width (F_m/w_s) can support before fracture depends on the pin positioning related microstructures in the weld interface region. Microstructures differ depending on whether or not the tool pin penetrates the lapping interface. It has been found that F_m/w_s values of the present macro-defect free weld samples vary quite significantly but in general are significantly higher than those reported in the literature. When the penetration depth is zero a thin Al6060–Ti64 interface layer forms and this layer does not grow beyond 250 nm. It will be shown that the thin interfacial layer can support a high tensile-shear load and thus the adjacent Al6060 material shears to fracture. When the pin penetrates during FSLW and thus the commonly observed mix stir zone (MSZ) forms, values of F_m/w_s are lower than that of zero pin penetration welds but remain quite high. This can be explained by cracks needing to propagate through the tough and high strength α(Ti) regions in MSZ, requiring a high load to cause fracture.

Introduction

Joining of dissimilar metals is a necessity in many industries. However, in general, high quality joints using fusion welding are difficult to obtain. This difficulty, as recently explained by Kah et al. [1] in their assessment of the current trends in welding of dissimilar metals, is due to the excessive growth of interface brittle intermetallics. Thus, joining processes during which the growth of the intermetallics is slow, such as solid state joining and brazing, are attractive for dissimilar metals joining. Dissimilar metals welding of an Al alloy to a Ti alloy (Al-to-Ti) is industrially important, particularly for the aerospace industry. Thus, friction stir welding (FSW) which is a solid state process has been studied in the last few years for Al-to-Ti welding in both butt joint [2], [3], [4], [5], [6], [7] and lap joint geometry [8], [9], [10], [11], [12].

Earlier work by Dressler et al. [2] demonstrated how good quality Al-to-Ti FSW butt welds could be made. In their work, the pin was off set to the Al (AA2024-T3) side but the path of the pin also included a very small (<1 mm) portion of the Ti (Ti–6Al–4V, named as Ti64) butting side. A small Ti–Al mixed interface zone formed and fracture took place in this zone during tensile testing. The fracture strength of the joint was 348 MPa which is a high value considering that the maximum strength of AA2024 to Ti64 joints by a welding–brazing process was 158 MPa [13]. Welding–brazing is a dissimilar metals joining method which has recently been quite intensively studied particularly for Ti-to-Al joining [13], [14], [15], [16], [17], [18], [19], [20].

Along with the early study on butt joint FSW, friction stir lap welding (FSLW) of Al-to-Ti (Al–Si die cast alloy to Ti) was conducted by Chen and Nakata [8]. In this work, using force control, the different forward speed values under the same down force resulted in different microstructural features in the joint interface region. In general, the interface region of dissimilar metals FSL welds is a mixed stir zone (MSZ), which is a highly irregular structure of mixed metallic and intermetallic layers [20], [21]. The zone is the mechanical consequence of the tool pin having penetrated to the bottom plate and also is the consequence of the formation of intermetallics. The interface can also be a thin layer without MSZ, if the penetration depth of the pin (d_P) can be precisely set to zero [21]. Lack of weld and voids can also be present. Joint strengths of welds with different interface features differ accordingly.

FSL welds are commonly tested using the tensile-shear test. The force at fracture over the width of the specimen (F_m/w) is often used to indicate how strong a joint is. In Chen and Nakata׳s work [8], (F_m/w)_Max=470 N/mm (9390 N/20 mm). Using force control, the value of d_P could not be pre-set. Micrographs of the (F_m/w)_Max joint display Ti particles which appear lighter on the Al–Si side, indicating d_P>0. However, the absence of a clear MSZ also suggested that d_P was very close to zero. The photo of the fractured sample of the (F_m/w)_Max joint also indicated little penetration. Chen and Nakata [8] provided images of another joint showing MSZ with large voids (>10 μm in size) and for this joint F_m/w=375 N/mm. As there were large voids, the effect of MSZ on F_m/w could not be evaluated.

Two later studies on FSLW of Al-to-Ti [9], [10] show that F_m/w values were less than 225 N/mm, for the case of a significant pin penetration. The considerably lower F_m/w values, in comparison to the values shown by Chen and Nakata, may mean a high amount of defects (microvoids or microcracks) in the interface region. For their case of no penetration, however, F_m/w values were also low at ~200 N/mm [10]. The reason for this is not clear as the controlling of pin penetration was not intended in the study and their photos of cross sections do not reveal whether or not a complete joint had actually been established. If d_P⪡0, insufficient joining or no joining can be expected.

In the recent Al-to-Ti FSLW experiments (by Krutzlinger et al. [12]) using position control, the (F_m/w)_Max value (for A6082 to Ti64) equal to ~480 N/mm has again been obtained. In their study, position control set d_P values to range from −0.4 mm to −0.1 mm, meaning that there should not be pin penetration in their experiments. However, their X-ray computer tomogram (for d_P value aimed at −0.1 mm) and micrographs (for d_P value aimed at −0.2 mm) clearly show the presence of MSZ. Thus, although they have not specifically explained the observation, tool pin did penetrate in their experiments of −0.1 and −0.2 mm d_P setting. This means that when a very small |d_P| value is used, as can be reasonably expected, whether the pin penetrates or not during FSLW experiments can be highly uncertain.

As a comparison, Al-to-Ti (Al7075 to Ti64) joining for which the maximum value can be expressed to be F_m/w=450 N/mm was achieved through transient liquid phase bonding (TLPB) [22]. A comparison can be further made to FSLW of Al-to-steel which has recently been studied extensively [21], [23], [24]. For all defect-free but MSZ-containing welds, F_m/w values were close to 310 N/mm. This was in spite of the significant differences in various parameters (size and shape of tool, thicknesses of joining plates, rotation/forward speeds, and size of test sample) used in the different experiments. On the other hand, when d_P≈0 and a thin and continued intermetallic layer forms without the MSZ, the F_m/w value was considerably higher at 435 N/mm [21]. Thus, values of F_m/w at 470–480 N/mm may be regarded being moderately high but not very high for lap Al-to-Ti welds.

An objective of this work is to determine whether, using the condition of position control aided by force monitoring so that some samples can be obtained with more certainty of d_P≈0, F_m/w values higher than 470–480 N/mm are possible for FSL Al-to-Ti welds. As has also been indicated, the microstructural nature of the joining interface for achieving high strength values in FSL Al-to-Ti welds has not been sufficiently clear. Thus the other objective of the present study is to explain the effect on F_m/w of microstructures due to pin penetration (and non-penetration) in Al-to-Ti interface region formed during FSLW.