Research Article
Microstructural evolution in friction self-piercing riveted aluminum alloy AA7075-T6 joints

https://doi.org/10.1016/j.jmst.2020.12.023Get rights and content

Highlights

  • Solid-state bonding of Al sheets in F-SPR is formed by continuous dynamic recrystallization (CDRX).

  • CDRX of Al outside the rivet is due to the sliding-to-sticking transition caused by rivet rotation.

  • CDRX inside rivet is due to the strain rate gradient and sticking zone created by the rotation of trapped Al.

  • η’ & η phases coarsen in HAZ & TMAZs and completely dissolve in FGZ, resulting in Al softening.

Abstract

Friction self-piercing riveting (F-SPR) is an emerging technique for low ductility materials joining, which creates a mechanical and solid-state hybrid joint with a semi-hollow rivet. The severe plastic deformation of work materials and localized elevated temperatures during the F-SPR process yield complex and heterogeneous microstructures. The cut-off action of the work materials by the rivet further complicates the material flow during joint formation. This study employed the F-SPR process to join AA7075-T6 aluminum alloy sheets and systematically investigated the microstructural evolutions using electron backscatter diffraction (EBSD) techniques. The results suggested that as the base material approached the rivet, grains were deformed and recrystallized, forming two distinct fine grain zones (FGZs) surrounding the rivet and in the rivet cavity, respectively. Solid-state bonding of aluminum sheets occurred in the FGZs. The formation of FGZ outside the rivet is due to dynamic recrystallization (DRX) triggered by the sliding-to-sticking transition at the rivet/sheet interface. The FGZ in the rivet cavity was caused by the rotation of the trapped aluminum, which created a sticking affected zone at the trapped aluminum/lower sheet interface and led to DRX. Strain rate gradient in the trapped aluminum drove the further expansion of the sticking affected zone and resulted in grain refinement in a larger span.

Introduction

The development of lightweight structures has recently been the focus of vehicle industries. Increasing the percentage of high specific strength alloys, e.g., aluminum alloys and magnesium alloys, is a trend toward realizing lightweighting target while maintaining structural stiffness and strength. To date, the most common solution to join these alloys is self-piercing riveting (SPR), in which a mechanical interlock is generated through deforming a semi-hollow rivet and the work materials [[1], [2], [3]]. SPR has been successfully applied to join 5xxx and 6xxx aluminum alloys in vehicle assembly. However, when riveting 7xxx series aluminum alloys, the low ductility nature of work materials caused an inevitable cracking issue [4]. A similar problem is also faced by magnesium alloys [5] and aluminum castings [6]. Several thermally assisted processes have been proposed, which preheat the work materials prior to SPR to inhibit cracking [7,8]. Although effective, their implementation in production is limited due to the high cost of auxiliary heating tools and the long process time.

Friction self-piercing riveting (F-SPR), which combines traditional SPR and friction stir heating, provides a solution to joining low ductility metals [9]. As illustrated in Fig. 1, a semi-hollow rivet rotates rapidly and meanwhile penetrates the workpieces. Therefore, frictional heat is generated and conducted to surrounding materials, which improves the formability of the work material and thus could inhibit cracking. Like the SPR process, a pip die is applied during F-SPR to support the work materials and facilitate the rivet shank spreading. In addition to crack-free joining, the introduction of frictional heat also results in solid-state bonding between workpieces. Both Li et al. [9] and Liu et al. [10] observed Al-Mg intermetallic in the F-SPR joints of aluminum alloy and magnesium alloy sheets. The F-SPR process was also employed to join carbon fiber reinforced plastic (CFRP) and magnesium alloy [11].

Solid-state bonding was verified to be an important strengthening factor of F-SPR joints. Ma et al. [12] joined AA7075-T6 by F-SPR and reported that the solid-state bonding between trapped aluminum in the rivet cavity and bottom aluminum sheet formed a metal “anchor”, which could significantly strengthen the joint. The F-SPR joints without solid-state bonding exhibit rivet pull-out failure mode in lap-shear tests, whereas the joints with solid-state bonding fail in the bottom aluminum sheet, resulting in a 13 % higher peak load and 20 % larger energy absorption. Another comparative study between F-SPR and SPR with AA5182-O [13] has shown that the solid-state bonding of F-SPR enhanced the stiffness of AA5182-O joints, which delayed fretting damage of aluminum sheet and improved the load amplitude of 106 fatigue life by 15 % compared to the SPR process. Solid-state bonding is a key strengthening feature of F-SPR joints, however, there is still a lack of microstructural understanding of its formation mechanism in the existing literature. Furthermore, it has been reported that the F-SPR joint of aluminum alloy AA6061-T6 and magnesium alloy AZ31B exhibited local softening in the aluminum surrounding rivet [14], indicating that the F-SPR process could change the aluminum microstructure. This also highlights the need to study the microstructure evolution during F-SPR so as to understand the process further and optimize joint performance.

Similar to other friction stir related joining processes, e.g., friction stir welding (FSW), the F-SPR also has a thermo-mechanical nature. The generated frictional heat and the stirring of the rivet would affect the microstructures as well as the hardness of materials, which in turn determines the mechanical performance of the joint. Since the microstructure evolution of F-SPR has not been reported in the literature, related studies of FSW or its variants were reviewed in the following.

The FSW joints feature several typical affected zones, including heat affected zone (HAZ), thermomechanical affected zone (TMAZ), and stir zone (SZ). For FSWed heat-treatable aluminum alloy joints, the affected zones are more or less softened compared to the base material (BM). Shen et al. [15] reported the hardness reduction in FSWed AA7075-T6 aluminum alloy joints with the minimum hardness value at the HAZ/TMAZ boundary, which was attributed to the coarsening of Mg2Zn and Al2CuMg precipitates in the affected zones. Reimann et al. [16] reported that the SZ and HAZ of refill friction stir spot welded (FSSWed) AA7075-T651 joints presented 36 % and 42 % lower hardness compared to the BM, respectively. Tao et al. [17] found the SZ hardness in AA6061-T6 FSW joints, although lower compared to the BM, was higher than that in the HAZ, which was further explained as a result of the increased dislocation density and grain refinement induced by the intense stir in SZ. Wang et al. [18] explained the grain refinement mechanism in the SZ of AA6061-T6 FSW joints as the formation of low angle grain boundaries and a gradual increase in the boundary misorientations during hot deformation, i.e., the so-called continuous dynamic recrystallization (CDRX). Zeng et al. [19] explained the dynamic recrystallization process of 6061-T6 FSW joint as the progressive lattice rotation due to the pinning of precipitation particles. Hu et al. [20] pointed out that the strengthening phases precipitated at grain boundaries in the SZ of FSWed 2219-T6 aluminum alloy joint effectively inhibited the growth of recrystallized grains and contributed to the increment of SZ hardness compared to the HAZ and TMAZ. Min et al. [21] observed the occurrence of recrystallization within the SZ of a friction stir blind riveted AA6111-T4 sheet, which decreased the original grain size from 24.9 μm to around 1 μm in the SZ.

It is noteworthy that although the grain refinement and increased dislocation density enhance the local materials in FSWed heat-treatable aluminum alloys, hardness loss caused by the coarsening of precipitates can hardly be compensated, which results in relatively low joint efficiency. In contrast, the FSWed joints of non-heat treatable aluminum alloys, e.g., refill FSSWed AA5042-O [22] and FSWed AA5754-H22 [23] present increased hardness in the affected zones due to grain refinement and work hardening, and therefore a relatively high joint efficiency, as listed in Table 1.

F-SPR features a semi-hollow rivet, which creates diverse material flow patterns in the rivet cavity and outside of the rivet. Also, the semi-hollow rivet cuts through top workpiece and traps part of the work material within its cavity, creating a unique joint profile compared to that of the FSW process. The objective of the current work is to investigate microstructure evolution and to characterize affected zones for F-SPR joints. In the following, the F-SPR experiments and material characterization methods were presented firstly. Then, the macro profiles, microstructures, and hardness distributions in different locations of the joints were investigated, followed by the partition of affected zones. Finally, the formation mechanisms of individual affected zones and solid-state bonding were discussed.

Section snippets

Materials and experimental procedures

Two layers of 2.0-mm-thick AA7075-T6 aluminum alloy sheets were joined by F-SPR. The semi-hollow rivet was made by a cold heading process from bar-shaped 35CrMo steel. The average hardness of the as-received AA7075-T6 sheets and the as-fabricated rivets were measured to be 169 HV and 255 HV, respectively. Table 2, Table 3list the nominal chemical composition and mechanical properties of the base materials.

The F-SPR joints were made using a specially designed machine as given in [31]. A

Macrostructure

Fig. 3(a) gives a typical F-SPR joint’s cross-section profile made under 3600 rpm-2.0 mm/s. As shown, the rivet cut through the upper sheet and displaced the lower sheet towards the die cavity. The aluminum of the upper sheet that had been cut off was trapped in the rivet cavity. The joint was sliced along the middle plane of the upper and lower sheets, respectively, refer to Fig. 3(b)–(c). As shown in Fig. 3(d) and (f), large gaps were observed at the interface between the outer rivet wall and

Microstructure evolution during the F-SPR process

In an F-SPR joint, the grain structures outside the rivet are similar to those of an FSW joint with FGZ/SZ, TMAZ, and HAZ extending from the weld/joint center towards BM. However, the F-SPR joint has unique microstructures in the rivet cavity exhibiting coarse grains in the upper portion and fine grains in the lower portion. Besides, solid-state bonding occurs in the FGZs, forming a key strengthening feature of F-SPR. To understand the formation process of the FGZs and solid-state bonding, the

Conclusions

This paper investigates the microstructural evolution of friction self-piercing riveted AA7075-T6 aluminum alloy joints. The unique affected zones in the F-SPR joints were identified for the first time. The corresponding formation mechanisms for the fine grain zones and their influence to the solid-state bonding were clarified. The following conclusions can be drawn:

  • (1)

    The F-SPR process introduces intense plastic deformation and grain structure variations of the aluminum alloy sheets, creating an

Declaration of Competing Interest

The authors report no declarations of interest.

Acknowledgments

The authors would like to acknowledge the financial support of the National Natural Science Foundation of China (Grant Nos. 52025058 and U1764251), the National Key Research and Development Program of China (Grant No. 2016YFB0101606-08) and Shanghai Jiao Tong University. A part of this study was also financially supported by Project to Create Research and Educational Hubs for Innovative Manufacturing in Asia, Joining and Welding Research Institute, Osaka University.

References (55)

  • K. Mori et al.

    J. Mater. Process. Technol.

    (2012)
  • X. Zhang et al.

    J. Mater. Process. Technol.

    (2019)
  • H. Jiang et al.

    Mater. Des.

    (2019)
  • H. Duan et al.

    Mater. Des.

    (2014)
  • Y. Durandet et al.

    Mater. Des.

    (2010)
  • X. Liu et al.

    J. Mater. Process. Technol.

    (2016)
  • Y. Ma et al.

    J. Mater. Process. Technol.

    (2020)
  • M. Reimann et al.

    Mater. Des.

    (2017)
  • W. Tao et al.

    Mater. Sci. Eng. A

    (2016)
  • B. Wang et al.

    Mater. Des.

    (2015)
  • X. Zeng et al.

    J. Mater. Sci. Technol.

    (2019)
  • Y. Hu et al.

    Mater. Sci. Eng. A

    (2018)
  • M.D. Tier et al.

    J. Mater. Process. Technol.

    (2013)
  • X. Liu et al.

    J. Mater. Sci. Technol.

    (2018)
  • A.H. Baghdadi et al.

    Mater. Sci. Eng. A

    (2019)
  • W. Xu et al.

    J. Mater. Sci. Technol.

    (2018)
  • F. Liu et al.

    Mater. Des.

    (2017)
  • H. Miura et al.

    Scr. Mater.

    (2004)
  • J. Zhang et al.

    J. Mater. Sci. Technol.

    (2018)
  • H. Liu et al.

    Mater. Sci. Eng. A

    (2018)
  • R.S. Mishra et al.

    Mater. Sci. Eng. R

    (2005)
  • H. Ji et al.

    Mater. Des.

    (2020)
  • Z. Zhang et al.

    J. Mater. Sci. Technol.

    (2020)
  • J.Q. Su et al.

    Acta Mater.

    (2003)
  • M. Ahmed et al.

    J. Mater. Process. Technol.

    (2017)
  • T.R. McNelley et al.

    Scr. Mater.

    (2008)
  • Y. Hu et al.

    J. Manuf. Processes

    (2020)
  • Cited by (0)

    View full text