1 Introduction

Al5083 has been the main material of choice for ship, automobile, transport equipment, aeronautics, and astronautics due to its excellent corrosion resistance and weldability [1]. Nonetheless, the hardness and wear resistance are very important in widening their application areas [2]. It is possible to obtain the high hardness and wear resistance of the surface layer and high toughness of the matrix via fabricating particles reinforced Al5083 surface composites [3]. In recent years, the fabricating technologies including stir casting, squeeze, and powder metallurgy lead to the formation of the brittle intermetallic compounds at the interface of molten aluminum and reinforce particles [4]. Thus, it is absolutely vital for the development of new technology which can be used for the Al5083 surface composites.

As a high-efficiency microstructural modification technology, friction stir processing (FSP) has been used for fabricating Al base composites [5,6,7]. In FSP process, grain refinement and microstructure uniform can be realized by severe plastic deformation [8]. This technology has been applied for eliminating the casting defects, modifying the microstructure, and processing the surface composites [9].

Recently, FSP technology has shown the advantages in fabricating Al5083 surface composites. As the most common reinforced phase, the addition of ceramic particles into the matrix enhances the mechanical properties of the Al5083 surface composites [10]. Mirjavadi et al. investigated the effect of the processing passes on the microstructure, microhardness, tensile, and wear properties of Al5083/ZrO2 and Al5083/TiO2 nanocomposites [11, 12]. The results showed that increasing process passes improved the microhardness and tensile properties of the composites and the wear rate of the composites significantly decreased after multi-pass friction stir processing. Yuvaraj et al. investigated the microhardness of Al5083/B4C surface composites via FSP which was 52.2% higher compared with the matrix of Al5083 [13]. The increase of microhardness was due to the presence of nano size B4C particles which contributed to producing ultrafine grain size. Soleymani et al. fabricated hybrid composites via dispersing SiC and MoS2 into Al5083, and the results showed that better wear resistance for hybrid was obtained [14]. Amra et al. developed Al5083/CeO2/SiC hybrid composites by FSP technology [15]. The microstructure and wear resistance of composites were studied, and the results showed that the hybrid composites exhibited the lowest coefficient and the highest wear resistance, and the Al5083/CeO2 composites revealed the highest hardness. The improvement of the wear resistance was attributed to the solid lubrication effect of the CeO2. Carbon nano-tubes (CNTs) and graphite were also used to produce the Al5083 hybrid composite. Hosseini et al. reported that high hardness and tensile strength were obtained for hybrid composites of Al5083 when the volume ratio of CNTs and CeO2 was 75:25 [16]. Khan et al. produced Al5083 hybrid composites containing CNTs and B4C by FSP [17]. The hardness and tensile strength of the composites increased by 20% and 41%. Mostafapour Asl and Khandani studied the role of the hybrid ratio of the Graphitep/Al2O3p in mechanical and wear properties of Al5083 composites [18]. The results showed that the hardness and wear rate of the composites had two stages with the variation of hybrid ratio. Kurt investigated the effect of hybrid ratio on the tensile strength of Al5083 hybrid composites containing graphite and ZrO2 [19]. The maximum tensile strength of the composite was obtained when the volume fractions of graphite and ZrO2 were 10% and 5%, respectively. As alternative to the brittle ceramic particles, the metallic particles which were stronger than Al were also used for enhancing Al5083 alloy. García-Bernal et al. studied the hot deformation behavior of FSPed Al5083 alloy with different Mn contents, and the results revealed that adding Mn in the range of 0.49–1.00% had a positive effect on microstructure refinement of Al5083 alloy processed via FSP [20, 21]. W and Ni particles incorporated in Al5083 alloy via FSP technology to produce metal particles reinforced composites were studied by Bauri et al. [22,23,24,25]. The results showed that the composite exhibited both high strength and ductility. Huang et al. added Ti particles into Al5083 alloy via submerged friction stir processing (SFSP), and it both inhibited grain growth and a boosted the recrystallization [26, 27]. Yang et al. incorporated the high-entropy alloy (HEA) particles (AlCoCrFeNi) into Al 5083 matrix via underwater friction stir processing, and the results showed that HEAs particles increased the microhardness and wear resistance [28, 29]. Jain et al. manufactured Al5083-CNTs/SiC composites by FSP, and the results showed that in situ Al4C3 intermetallic compound was formed [30].

It is well known that the different expansion coefficient and elastic modulus between ceramic particles and Al5083 matrix cause the weak interfacial bonding, and single metal particles cannot improve the performance of composites. Thus, seeking new reinforced phase is urgent. As a breakthrough of conventional alloys, high-entropy alloys has high hardness and strength due to the lattice distortion effect and sluggish diffusion effect resulting from different atomic sizes of principal elements [31,32,33]. Now, some technologies, such as laser melt injection and laser additive synthesis, have been used for producing the Al/HEA composites [34, 35]. However, there are severe interface reaction between Al matrix and HEA particles. Now, fabricating high-entropy alloy reinforced aluminum matrix composites by FSP are scarce.

In this paper, FeCoNiCrAl high-entropy alloy was used as the reinforced phase, and Al5083 alloy was used as the matrix. The Al5083 composites were produced via FSP method. The effects of the processing pass on the microstructure, microhardness, wear resistance, and corrosion resistance have been studied.

2 Materials and methods

Commercially Al5083 alloy sheets (200 × 200 × 5 mm3) were used as the matrix in this study, and the chemical composition of the Al5083 is shown in Table 1. The FeCoNiCrAl high-entropy alloy with the mean size of 16 μm was used as the reinforced phase, as shown in Fig. 1.

Table 1 The chemical composition of Al5083
Full size table
Fig. 1
figure 1

SEM image of FeCoNiCrAl high-entropy alloy particles

Full size image

The fabricated process is shown in Fig. 2. First, the groove, with 2 mm in width and 1.5 mm in depth, was cut in order to place the FeCoNiCrAl high-entropy alloy particles, as shown in Fig. 2a. Second, the HEA particles was put into the groove, as shown in Fig. 2b. Third, the HEA particles were compacted and sealed into the groove with a pinless FSP tool, as shown in Fig. 2c. Finally, the FeCoNiCrAl high-entropy alloy reinforced Al5083 composites were produced via FSP technology, as shown in Fig. 2d. The rotation speed, traveling speed, and tool angle were set at 1200 rpm, 45 mm/min, and 0°, respectively. The processing passes were 1, 3, and 5. The FSP tool was processed using H13 steel, and the schematic view of the size and configuration of the tool is shown in Fig. 3.

Fig. 2
figure 2

Schematic diagrams for fabricating composites via FSP

Full size image
Fig. 3
figure 3

Graph of the FSP tool: (a) diagram, (b) photo

Full size image

The samples were etched in two steps after grinding and polishing. First, the samples were etched by the Keller reagent (95 ml of H2O, 2.5 ml of HNO3, 1.5 ml of HCl, 1 ml of HF) for 15 s. Second, the samples were etched using the concentrated HCl for 5 s in order to make the corrosion morphology more distinct. The microstructure of composites was observed by stereomicroscopy, Zeiss-Supra 55, and S4800 scanning electron microscope (SEM) coupled with an energy-dispersive X-ray spectroscope (EDS). D8 Advance X-ray diffraction (XRD) was used for analyzing the phase of the composites.

The microhardness was measured via HV-1000B Vickers microhardness tester, and the load and dwell time were 50 g and 15 s, respectively. The wear resistance of the composites was measured by BRUKER UMT-2 friction abrasion testing machine according to ASTM G99 using GCr15 steel with 40 mm/s line speed at a load of 5 N for 15 min. The computational equation of the wear rates was:

$$ K=\frac{V}{Fm} $$
(1)

where F is the load, m is the sliding distance, V is the wear volume, and K is the wear rates.

3 Results and discussion

Figure 4 shows the typical cross-section image of the 5-pass FSPed composites. The different characteristic zones, including stir zone (SZ), thermo-mechanically affected zone (TMAZ), heat affected zone (HAZ), and the base materials (BM), were observed. In the process of FSP, the recrystallized microstructure with small equiaxed grains was produced in the SZ because of the deformation and heat which came from the friction and stirring of the FSP tool [12]. HAZ and TMAZ were the transition zones between BM and SZ. The deformation occurred, but the recrystallization did not occur in TMAZ. HAZ was only subjected to the heat cycle.

Fig. 4
figure 4

Macroscopic overview of the cross-section of 5-pass FSPed composites

Full size image

The microstructure of the fabricated composites in the SZ using different processing passes is shown in Fig. 5a–c, and the white particles were HEA particles from the EDS analysis (Fig. 5d). It indicated that the processing passes had a significant effect on the distributions of the HEA particles. In 1 processing pass, the distribution of the HEA particles was not uniform, and the HEA particles breaking was also uneven. Meanwhile, the tunnel defects can be found as the processing pass was 1 (Fig. 5a). SEM micrograph of the 3-pass FSPed composites (Fig. 5b) shows relatively uniform dispersion of the HEA particles, and the tunnel defects disappeared. When the processing pass was 5, the HEA particles were further broken and more evenly distributed in the matrix, as shown in Fig. 5c. During the FSP process, the matrix and the HEA particles were sheared, squeezed, and broken due to the heat cycle, mechanical mixing, and severe plastic deformation [14]. For 1 pass, the heat input was insufficient, which resulted in the weak materials flow. These factors caused the formation of cavity at the end of the stir pin. Increasing processing pass increased the heat input and the stir time. The Al matrix recurred plastic flow with the rotation of the stir tool when increasing the processing pass, which increased the amount of the plasticity deformation and decreased plastic deformation resistance. The HEA particles were wrapped by plastic Al matrix. With the increase of the plastic flow, the aggregation of the HEA particles was eliminated, which made the HEA particles distribute more evenly in the Al matrix. A further comparison between Fig. 5 a, b, and c showed that the increase of the processing pass made the particle degree of the HEA particles smaller. This was due to the fact that the diameter of the original HEA particles was about 16 μm, which meant that they were further break due to the friction heating, mechanical mixing, and severe plastic deformation generated during FSP.

Fig. 5
figure 5

SEM images of the fabricated composites in different processing pass: (a) 1 pass, (b) 3 pass, (c) 5 pass, and (d) EDS spectrum of the red box in Fig. 5a

Full size image

Although 5-pass processing already made the HEA particles uniformly dispersed and refined in SZ, the nonuniform phenomenon of the HEA particles still existed in the edge of friction stir zone, as shown in Fig. 6. This was because the metal plastic flow of the edge of friction stir zone was weaker than that in the SZ.

Fig. 6
figure 6

(a) SEM image of the edge of friction stir zone. (b) EDS elemental mapping

Full size image

It was found that the HEA particles were embedded in the Al matrix and combined with the Al matrix well when further analyzing the microstructure of the composites, as shown in Fig. 7 a and b. In addition, it can be seen that the interface between the HEA particles and the Al matrix was clear without serious interfacial reaction. The EDS map (Fig. 6) also revealed that there was no reaction layer around the HEA particles at the interface. And the EDS line scans of the HEA particles and Al matrix across the interface did not show new phase, as shown in Fig. 8. The FSP was a solid and fast processing technology, and the contact time between the HEA particles and Al matrix was short, which could effectively control the generation of the interface.

Fig. 7
figure 7

(a) SEM images of composites via 5-pass processing. (b) HEA particle and Al matrix interface

Full size image
Fig. 8
figure 8

EDS line scan across a HEA particles: (a) SEM, (b) EDS

Full size image

The XRD pattern of the composites could give a further explanation. Figure 9 shows the XRD pattern of the Al/FeCoNiCrAl composite in different process pass, which showed only Al and HEA peaks without intermetallic phase peak in both the composites. There was no diffusion of the HEA particles into the Al5083 alloy seen from the lack of evidence for the reaction products in Fig. 9. The HEA was mainly Fe, Co, Ni, and Cr, which were difficult to dissolve due to their high dissolution temperatures. The thermally stable FeCoNiCrAl particles also existed, and the melting temperature of FeCoNiCrAl was 1350 °C which was much higher than the peak temperature reached by FSP. As a result, the HEA particles remained unchanged independent on the number of passes.

Fig. 9
figure 9

XRD of the specimen FSPed HEA particles 5 passes

Full size image

The variation in the microhardness of the Al5083 base materials, without and with particles FSPed specimens, is shown from Fig. 10. The average microhardness of the BM was 78 HV. As can be seen in Fig. 10, the microhardness of the sample without HEA particles by 1-pass FSP increased due to the grain refinement strengthening according to the Hall-Petch relationship. By adding the HEA particles, the microhardness of the composites was further increased compared with the BM and FSPed samples without the HEA particles due to the Orowan strengthening (the dispersion and obstruct dislocation movement of the HEA particles). In general, the hardness of the particles reinforced aluminum matrix composites owing to the refinement of the grain, the dispersion of reinforcing particles, and the quench hardening effect due to difference in thermal contraction coefficient of Al alloy and reinforcing particles [13]. The heat input, dislocation density, HEA particles, and grain size influenced the microhardness of the composites. Meanwhile, the increase of the microhardness resulted in the high hardness of the HEA particles and the resistance to plastic deformation during indentation [12]. It can also be seen from Fig. 10 that the maximum microhardness of 158 HV was obtained with 5 passes. Increasing the processing pass could enhance the distribution of the HEA particles, thus improving the dispersion strengthening effect.

Fig. 10
figure 10

Microhardness of Al5083 base alloy, without and with particles FSPed specimens

Full size image

Figure 11 illustrates the friction coefficient curve versus time and the friction coefficient under different processing conditions. Meanwhile, the average values of the friction coefficient regarding base material and FSPed samples are shown in Fig. 12. As can be seen in Fig. 11 and Fig. 12, the processing condition and HEA particles have significant effects on the trends of friction coefficient and average values. For the curves, the friction coefficient of base material and FSPed sample without HEA particles had great fluctuation. With the addition of HEA particles and the increase of the processing pass, the fluctuation of the friction coefficient curves declined. For friction coefficient, the average friction coefficient reduced from 0.5844 for the base material to 0.5491 for the FSPed sample without HEA particles due to the grain refining effect of the FSP technology. Adding the HEA particles further decreased the friction coefficient, and the friction coefficient of the 1-pass sample was 0.5011. The friction coefficient of the composites decreased with the increase of the processing passes. The minimum friction coefficient value of 0.4233 was obtained when the processing pass was 5.

Fig. 11
figure 11

Variation of friction coefficient with time in (a) base material, (b) without particles 1 pass, (c) with particles 1 pass, (d) with particles 3 pass, and (e) with particles 5 pass

Full size image
Fig. 12
figure 12

Friction coefficient for Al5083 base alloy and FSPed samples

Full size image

Figure 13 shows the wear rate of the base material and the FSPed samples. As can be seen in Fig. 13, the wear rate of the FSPed sample without HEA particles was lower than the Al5083 base alloy which was ascribed to the higher hardness (Fig. 10). In the composites, the wear rate was lower than the FSPed sample without HEA particles. This phenomenon was related to the load-bearing capacity of the HEA particles apart from the higher microhardness. Meanwhile, the wear rate of the composites showed decreasing trend with the increase of the processing pass. This can be ascribed to the better dispersion of the HEA particles in Al5083 alloy.

Fig. 13
figure 13

Wear rate for Al5083 base alloy and FSPed samples

Full size image

The SEM micrograph of the wear surface of Al5083 base material and FSPed composites is displayed in Fig. 14. As can be seen in Fig. 14a, the Al5083 base material showed the massive surface damage and high amount of materials flow and plastic deformation. Because of the lower microhardness of the base material, the abrasive ball could easily penetrate into the base material surface and caused a large surface damage which resulted in a severe plastic deformation and high amount of base material loss [12]. It indicates that the dominant wear mechanism was adhesive wear with barely visible scratches on the surface. Figure 14 b–d show the SEM image of the composites in different processing passes, which revealed a distinct difference from the Al5083 base alloy. The composites had a smooth surface which showed fine grooves and ploughing. In local areas, the cavities were formed by the locally adhesive wear because of the breaking of the micro-welds during sliding [12]. With the increase of the processing pass, the surface showed many shallow grooves, which implied that the wear mechanism was changed from the adhesive wear to the moderate abrasive wear. There are the following reasons for the variation in microstructure and wear mechanism. On one hand, the microhardness of the composites was higher than the base material, and the high microhardness could prevent large damage. On the other hand, the composites had fine microstructure, which increased the load area thereby decreased the friction and tension between the couples. Moreover, increasing the processing pass could improve the distributed HEA particles. The HEA particles could reduce the plastic deformation by inhibiting the dislocation motion.

Fig. 14
figure 14

SEM images of the wear surface of Al5083 (a) base materials, (b) with particles 1 pass, (c) with particles 3 pass, and (d) with particles 5 pass

Full size image

4 Conclusion

In this investigation, the FeCoNiCrAl high-entropy alloy reinforced Al5083 composite was produced via friction stir processing. The effect of the FSP passes on the microstructure, microhardness, and wear properties was studied. The conclusions are as follows:

  1. (1)

    FeCoNiCrAl high-entropy alloy particles were distributed homogeneously in Al5083 base alloy in their primitive state, which was just fragmentation due to the stirring of the tool.

  2. (2)

    Distribution of the FeCoNiCrAl high-entropy alloy particles was more homogeneous when increasing the processing pass; however, the big size particles were observed on the edge of friction stir zone.

  3. (3)

    The composite exhibited even better microhardness compared with the Al5083 base alloy and the FSPed samples without the FeCoNiCrAl high-entropy alloy particles.

  4. (4)

    Friction coefficient and wear rate of the composites were reduced with the increase of the processing pass. This was attributed to the increase of microhardness and the even distribution of the FeCoNiCrAl high-entropy alloy particles.