Microstructure and Mechanical Properties of Fiber Laser Welding of Aluminum Alloy with Beam Oscillation

Laser welding with beam oscillation is applied to join aluminum alloy plates in butt configuration. The effects of beam oscillating patterns on the quality of welds are compared and analyzed. The results indicate that beam oscillation can improve the weld formation and microstructure of butt joints. The circular oscillating weld has the features of fine grain and uniformly dispersed dendrites in the strengthening phase, and the porosity inhibitory effect of circular oscillation is the most obvious. In addition, beam oscillation has few effects on the tensile strength of welds, but exerts an influence on the elongation of welds.


Introduction
Aluminum alloys are an important engineering material in the automobile, rail, and aerospace industries as the overall performance meets the requirements of mechanical strength and low weight [1][2][3][4]. When aluminum alloy parts are used, a welding process is often needed to join the aluminum alloy parts with other components or structures. The weldability of an aluminum alloys is varied with different types of aluminum alloys. To some degree, the existence of magnesium, zinc, and silicon elements can improve the weldability of aluminum alloys due to the lower thermal conductivity and melting temperature [5]. The 5000-series aluminum alloys have better weldability than other types of aluminum alloys [5,6]. Various welding techniques are available [7][8][9]. In comparison with traditional welding techniques, laser beam welding has the advantages of a high energy density, a high welding speed, a deeper penetration depth, and a small heat-affected zone (HAZ) [10][11][12][13][14][15]. Therefore, this paper is especially intended to discuss a laser welding process that can meet the aforementioned requirements of a joining process.
Beam oscillation technology is proposed during electron beam welding [16], and then transplanted to laser beam welding [17,18]. Previously, laser beam oscillation was obtained by the mechanical swing of a welding head with low frequency and poor stability. However, a high-power galvanometer scanner was developed and widely used for laser beam oscillation due to its high frequency and high flexibility. Currently, laser welding with beam oscillation is applied widely in remote laser welding for its good weldability [19][20][21][22]. The primary cause of good weldability with an oscillating beam is the dilution degree of the melt pool. Therefore, laser beam oscillating welding is more popular in the industry. Zhang et al. [17] demonstrated that a good weld quality without porosity could be achieved with a weaving laser beam during the laser welding of carbon steel. Hao et al. [23] investigated the influences of beam oscillating parameters on the weld appearance during the laser welding of austenitic stainless steel. Wang et al. [24,25] found that beam oscillation stabilized the process and improved weld morphology. Furthermore, the elongation of welds was increased. Müller et al. [26] documented that beam oscillation strategies could improve the gap bridge ability in fillet welding. Hagenlocher et al. [27] revealed that the equiaxed dendritic grain was prone to be generated without the formation of hot cracks during the laser oscillation welding of aluminum. Kraetzsch et al. [28] argued that fiber laser welding with beam oscillation offered possibilities for crack-free welding of dissimilar Al-Cu metals, comparable to the electron beam. Kim et al. [29] observed that the solidification crack disappeared at a weaving frequency of 5 Hz during laser weave welding of a self-restraint tapered aluminum alloy. Berend et al. [30] found that the humping defect of aluminum alloy welds could be suppressed by high-frequency oscillation and the welding process could be stabilized under high-speed welding.
In summary, the research on laser beam oscillating welding is still at an initial stage [31]. Studies have indicated that laser oscillation welding has the advantages of process stability and good welding quality. However, the effects of beam oscillation on the microstructure and mechanical properties of aluminum alloy welds have seldom been studied [24,32]. The present study makes an attempt to compare and analyze the effects of no oscillation, transverse oscillation and circular oscillation on the performance of aluminum alloy butt welding through a welding test on a 4-mm-thick 5056 aluminum alloy butt joint with a fiber laser and a wobble laser welding head. The weld appearance, metallurgical, and mechanical properties of the butt-welded joints are analyzed comparatively. Figure 1 displays the setup of the experiments. A continuous-wave fiber laser (YLR-6000-CL; IPG Photonics Co., Oxford, MA, USA) was used as the laser source. The laser beam emitted from the end of the operation fiber was collimated and focused with a wobble laser welding head (FLW-D50-W). The laser beam emitted from the end of the optical fiber was collimated by a lens with a focal length of 200 mm and afterwards was transmitted by a galvanometer scanner unit and focused on the specimen surface by an f-theta focusing unit with a focal length of 300 mm. The spot size of the focused laser beam was approximately 0.225 mm.

Design of Experiments
Appl. Sci. 2019, 9, x FOR PEER REVIEW 2 of 11 et al. [23] investigated the influences of beam oscillating parameters on the weld appearance during the laser welding of austenitic stainless steel. Wang et al. [24,25] found that beam oscillation stabilized the process and improved weld morphology. Furthermore, the elongation of welds was increased. Müller et al. [26] documented that beam oscillation strategies could improve the gap bridge ability in fillet welding. Hagenlocher et al. [27] revealed that the equiaxed dendritic grain was prone to be generated without the formation of hot cracks during the laser oscillation welding of aluminum. Kraetzsch et al. [28] argued that fiber laser welding with beam oscillation offered possibilities for crack-free welding of dissimilar Al-Cu metals, comparable to the electron beam. Kim et al. [29] observed that the solidification crack disappeared at a weaving frequency of 5 Hz during laser weave welding of a self-restraint tapered aluminum alloy. Berend et al. [30] found that the humping defect of aluminum alloy welds could be suppressed by high-frequency oscillation and the welding process could be stabilized under high-speed welding.
In summary, the research on laser beam oscillating welding is still at an initial stage [31]. Studies have indicated that laser oscillation welding has the advantages of process stability and good welding quality. However, the effects of beam oscillation on the microstructure and mechanical properties of aluminum alloy welds have seldom been studied [24,32]. The present study makes an attempt to compare and analyze the effects of no oscillation, transverse oscillation and circular oscillation on the performance of aluminum alloy butt welding through a welding test on a 4-mm-thick 5056 aluminum alloy butt joint with a fiber laser and a wobble laser welding head. The weld appearance, metallurgical, and mechanical properties of the butt-welded joints are analyzed comparatively.  Four-millimeter-thick 5056 aluminum alloy plates were used as welding materials. Table 1 summarizes the chemical composition of the substrate. Table 2 gives the processing parameters of the laser welding experiments. Figure 2 is the schematic diagram of beam oscillating pattern. The oscillating patterns applied in our experiments are transversal and circular. The beam oscillation diameter/width ay is 1 mm, and the value of the degree of beam overlap ax is 0.3 mm. Four-millimeter-thick 5056 aluminum alloy plates were used as welding materials. Table 1 summarizes the chemical composition of the substrate. Table 2 gives the processing parameters of the laser welding experiments. Figure 2 is the schematic diagram of beam oscillating pattern. The oscillating patterns applied in our experiments are transversal and circular. The beam oscillation diameter/width a y is 1 mm, and the value of the degree of beam overlap a x is 0.3 mm.     The welded plates were cut into standard tensile samples by electrodischarge machining (EDM), as illustrated in Figure 3. The cross sections of the welded joints were also cut by EDM, and then polished with abrasive paper and polishing cloth. The finished cross sections of the welded joints were etched by Keller's reagent, and then observed by an optical microscope (OM) (Leica DM4000M, Solms, Germany). The composition of the joints was tested by scanning electron microscopy (SEM) (Tescan MIRA3 LMU, Brno, Czech) equipped with an energy-dispersive X-ray spectrometer (EDS). An X-ray diffraction (XRD) instrument (Rigaku CD/max2200VPC, Tokyo, Japan) was used to identify the strengthening phase in the fusion zone of the joint. The digital and intelligent micro-hardness tester (HVST-1000Z, Ningbo, China) was used to measure the Vickers microhardness. The tester has a load of 100 g force (gf) applied in a time of 20 s. Transverse tensile tests were carried out using a universal tensile testing machine (MTS CMT5105, Eden Prairie, USA) at room temperature with a crosshead speed of 1 mm/min. The fractured samples were cut and examined by SEM to reveal the fracture patterns. The welded plates were cut into standard tensile samples by electrodischarge machining (EDM), as illustrated in Figure 3. The cross sections of the welded joints were also cut by EDM, and then polished with abrasive paper and polishing cloth. The finished cross sections of the welded joints were etched by Keller's reagent, and then observed by an optical microscope (OM) (Leica DM4000M, Solms, Germany). The composition of the joints was tested by scanning electron microscopy (SEM) (Tescan MIRA3 LMU, Brno, Czech) equipped with an energy-dispersive X-ray spectrometer (EDS). An X-ray diffraction (XRD) instrument (Rigaku CD/max2200VPC, Tokyo, Japan) was used to identify the strengthening phase in the fusion zone of the joint. The digital and intelligent micro-hardness tester (HVST-1000Z, Ningbo, China) was used to measure the Vickers microhardness. The tester has a load of 100 g force (gf) applied in a time of 20 s. Transverse tensile tests were carried out using a universal tensile testing machine (MTS CMT5105, Eden Prairie, MN, USA) at room temperature with a crosshead speed of 1 mm/min. The fractured samples were cut and examined by SEM to reveal the fracture patterns.    The welded plates were cut into standard tensile samples by electrodischarge machining (EDM), as illustrated in Figure 3. The cross sections of the welded joints were also cut by EDM, and then polished with abrasive paper and polishing cloth. The finished cross sections of the welded joints were etched by Keller's reagent, and then observed by an optical microscope (OM) (Leica DM4000M, Solms, Germany). The composition of the joints was tested by scanning electron microscopy (SEM) (Tescan MIRA3 LMU, Brno, Czech) equipped with an energy-dispersive X-ray spectrometer (EDS). An X-ray diffraction (XRD) instrument (Rigaku CD/max2200VPC, Tokyo, Japan) was used to identify the strengthening phase in the fusion zone of the joint. The digital and intelligent micro-hardness tester (HVST-1000Z, Ningbo, China) was used to measure the Vickers microhardness. The tester has a load of 100 g force (gf) applied in a time of 20 s. Transverse tensile tests were carried out using a universal tensile testing machine (MTS CMT5105, Eden Prairie, USA) at room temperature with a crosshead speed of 1 mm/min. The fractured samples were cut and examined by SEM to reveal the fracture patterns.  Figure 4 shows the macroscopic morphologies of the butt-welded joints. As seen in Figure 4, the formation of the top and root surfaces of the weld can be affected by beam oscillation. Specifically, the top and root surfaces of the weld are discontinuous, unsmooth, and narrow in width without oscillation. For the weld formed by transverse oscillation, the top and root surfaces turn out to be smoother and still have a narrow width. However, the circular oscillating weld shows great improvement in the smoothness and consistency of its top and root surfaces and has a wider width. In addition, the specimens are completely welded in any case, but, as shown in Figure 4, root sagging or undercut is obvious on the top and root surfaces of the weld without oscillation, but insignificant root sagging or undercut is noted under transverse oscillation, and less or no root sagging or undercut is observed in circular oscillation.   Figure 4 shows the macroscopic morphologies of the butt-welded joints. As seen in Figure 4, the formation of the top and root surfaces of the weld can be affected by beam oscillation. Specifically, the top and root surfaces of the weld are discontinuous, unsmooth, and narrow in width without oscillation. For the weld formed by transverse oscillation, the top and root surfaces turn out to be smoother and still have a narrow width. However, the circular oscillating weld shows great improvement in the smoothness and consistency of its top and root surfaces and has a wider width. In addition, the specimens are completely welded in any case, but, as shown in Figure 4, root sagging or undercut is obvious on the top and root surfaces of the weld without oscillation, but insignificant root sagging or undercut is noted under transverse oscillation, and less or no root sagging or undercut is observed in circular oscillation.  Figure 5 demonstrates the optical micrographs of the butt joints. It is shown in Figure 5a that, in the fusion zone without oscillation, Mg3Al2 dendrites in the strengthening phase are accumulated and distributed inconsistently. The XRD analysis suggests that the Mg3Al2 phase is generated in the fusion zone of the welds, as shown in Figure 6. This can be explained by the fast cooling rate and the high degree of supercooling during the laser welding process. In contrast, Mg3Al2 dendrites tend to disperse in the fusion zone using transverse oscillation, as shown in Figure 5b. This is because there is a strong stirring effect in the welded keyhole, which forces the dendritic growth to disperse [32]. The fusion zone with circular oscillation is the best for the dispersion of Mg3Al2 dendrites, being very uniform and almost equal to that of the base material, as illustrated in Figure 5c. Therefore, we can conclude that the microstructure of the weld is significantly improved and the mechanical properties  Figure 5 demonstrates the optical micrographs of the butt joints. It is shown in Figure 5a that, in the fusion zone without oscillation, Mg 3 Al 2 dendrites in the strengthening phase are accumulated and distributed inconsistently. The XRD analysis suggests that the Mg 3 Al 2 phase is generated in the fusion zone of the welds, as shown in Figure 6. This can be explained by the fast cooling rate and the high degree of supercooling during the laser welding process. In contrast, Mg 3 Al 2 dendrites tend to disperse in the fusion zone using transverse oscillation, as shown in Figure 5b. This is because there is a strong stirring effect in the welded keyhole, which forces the dendritic growth to disperse [32]. The fusion zone with circular oscillation is the best for the dispersion of Mg 3 Al 2 dendrites, being very uniform and almost equal to that of the base material, as illustrated in Figure 5c. Therefore, we can conclude that the microstructure of the weld is significantly improved and the mechanical properties seem to be better with beam oscillation. As seen in Figure 5d, the grains at the heat-affected zone of the welds with beam oscillation grow to equiaxed dendrites more easily without columnar dendrites, which generally appear in laser welding without beam oscillation. This can be attributed to the periodic cycles of keyhole expansion and shrinkage in laser welding with beam oscillation [32].

Microstructure
Appl. Sci. 2019, 9, x FOR PEER REVIEW 5 of 11 seem to be better with beam oscillation. As seen in Figure 5d, the grains at the heat-affected zone of the welds with beam oscillation grow to equiaxed dendrites more easily without columnar dendrites, which generally appear in laser welding without beam oscillation. This can be attributed to the periodic cycles of keyhole expansion and shrinkage in laser welding with beam oscillation [32].   Appl. Sci. 2019, 9, x FOR PEER REVIEW 5 of 11 seem to be better with beam oscillation. As seen in Figure 5d, the grains at the heat-affected zone of the welds with beam oscillation grow to equiaxed dendrites more easily without columnar dendrites, which generally appear in laser welding without beam oscillation. This can be attributed to the periodic cycles of keyhole expansion and shrinkage in laser welding with beam oscillation [32].    Figure 7 is the cross-sectional hardness distribution of welds in different beam oscillation modes. The averages of three measurements is shown as one point in Figure 7. It is determined from Figure 7 that no significant difference is identified between the hardness of welds in the three beam oscillation modes, and also no difference between the hardness in different regions' base material (BM), fusion zone (FZ), and heat-affected zone (HAZ). In particular, the average microhardness of the fusion zone of the circular oscillating weld is max. 69.50 HV vs. min. 68.71 HV, compared to the weld without oscillation. Moreover, the change in microhardness change in the fusion zone of the weld is greater with oscillation than with circular or transverse oscillation. This confirms that beam oscillation has a beneficial effect on the uniform distribution of the fusion zone microstructure. In addition, the microhardness value of the annealed base material before welding is 66.83 HV. In summary, the microhardness of the weld is well matched to the metallographic microstructure, and the hardness of the weld and the heat-affected zone are both higher than those of the base material. An explanation for this is that the heat input is concentrated, the heating and cooling times are extremely short, and the weld structure formed is very small during laser welding, thus increasing the hardness. In other words, the circular oscillating weld has the finest and most dispersed dendrites and thus the largest microhardness, which confirms the Hall-Petch relationship. Figure 7 is the cross-sectional hardness distribution of welds in different beam oscillation modes. The averages of three measurements is shown as one point in Figure 7. It is determined from Figure  7 that no significant difference is identified between the hardness of welds in the three beam oscillation modes, and also no difference between the hardness in different regions' base material (BM), fusion zone (FZ), and heat-affected zone (HAZ). In particular, the average microhardness of the fusion zone of the circular oscillating weld is max. 69.50 HV vs. min. 68.71 HV, compared to the weld without oscillation. Moreover, the change in microhardness change in the fusion zone of the weld is greater with oscillation than with circular or transverse oscillation. This confirms that beam oscillation has a beneficial effect on the uniform distribution of the fusion zone microstructure. In addition, the microhardness value of the annealed base material before welding is 66.83 HV. In summary, the microhardness of the weld is well matched to the metallographic microstructure, and the hardness of the weld and the heat-affected zone are both higher than those of the base material. An explanation for this is that the heat input is concentrated, the heating and cooling times are extremely short, and the weld structure formed is very small during laser welding, thus increasing the hardness. In other words, the circular oscillating weld has the finest and most dispersed dendrites and thus the largest microhardness, which confirms the Hall-Petch relationship. The transverse tensile test represents an effort to assess the tensile strength of the butt joints. The stress-strain curves of welds are displayed in Figure 8. The tensile strength of joints welded with or without beam oscillation is 212-215 MPa, which lends support to the idea that the beam oscillation and weld surface conditions exert only minor influences on the weld tensile strength. However, the beam oscillation is an important predictor of weld ductility. The circular oscillating weld has the biggest strain of about 6.0%, 17.6% higher than that of the weld without oscillation (5.1%). The elongation of a transversely oscillating weld is measured at about 5.8%. According to [24], this is due to the increased proportion of equiaxed grains in the weld under circular oscillation.   Figure 9a,d, the specimen without oscillation starts to fracture from the undercut, and ends at the weld fusion zone after tension. A large pore is seen in the fracture, the presence of which reduces the force area and thus increases the tensile strength of the weld. The cause of the pores in the welds is the instability of the keyhole [33]. Porosity is a typical effect during the laser welding of aluminum. A similar facture pattern is established after tension in a specimen welded with transverse oscillation, compared to one without oscillation, as shown in Figure 9b. However, for the circular oscillating specimen, the fracture starts from the heat-affected zone of the weld, and ends at the weld fusion zone, as indicated in Figure  9c. In addition, no obvious pore is observed in the fractures of the welds with transverse oscillation and circular oscillation, as evidenced in Figure 9e,f. This sheds light on the fact that the undercut The transverse tensile test represents an effort to assess the tensile strength of the butt joints. The stress-strain curves of welds are displayed in Figure 8. The tensile strength of joints welded with or without beam oscillation is 212-215 MPa, which lends support to the idea that the beam oscillation and weld surface conditions exert only minor influences on the weld tensile strength. However, the beam oscillation is an important predictor of weld ductility. The circular oscillating weld has the biggest strain of about 6.0%, 17.6% higher than that of the weld without oscillation (5.1%). The elongation of a transversely oscillating weld is measured at about 5.8%. According to [24], this is due to the increased proportion of equiaxed grains in the weld under circular oscillation. The transverse tensile test represents an effort to assess the tensile strength of the butt joints. The stress-strain curves of welds are displayed in Figure 8. The tensile strength of joints welded with or without beam oscillation is 212-215 MPa, which lends support to the idea that the beam oscillation and weld surface conditions exert only minor influences on the weld tensile strength. However, the beam oscillation is an important predictor of weld ductility. The circular oscillating weld has the biggest strain of about 6.0%, 17.6% higher than that of the weld without oscillation (5.1%). The elongation of a transversely oscillating weld is measured at about 5.8%. According to [24], this is due to the increased proportion of equiaxed grains in the weld under circular oscillation.  Figure 9 is a graph showing the fracture effect of a tensile specimen. As indicated in Figure 9a,d, the specimen without oscillation starts to fracture from the undercut, and ends at the weld fusion zone after tension. A large pore is seen in the fracture, the presence of which reduces the force area and thus increases the tensile strength of the weld. The cause of the pores in the welds is the instability of the keyhole [33]. Porosity is a typical effect during the laser welding of aluminum. A similar facture pattern is established after tension in a specimen welded with transverse oscillation, compared to one without oscillation, as shown in Figure 9b. However, for the circular oscillating specimen, the fracture starts from the heat-affected zone of the weld, and ends at the weld fusion zone, as indicated in Figure  9c. In addition, no obvious pore is observed in the fractures of the welds with transverse oscillation and circular oscillation, as evidenced in Figure 9e,f. This sheds light on the fact that the undercut  Figure 9 is a graph showing the fracture effect of a tensile specimen. As indicated in Figure 9a,d, the specimen without oscillation starts to fracture from the undercut, and ends at the weld fusion zone after tension. A large pore is seen in the fracture, the presence of which reduces the force area and thus increases the tensile strength of the weld. The cause of the pores in the welds is the instability of the keyhole [33]. Porosity is a typical effect during the laser welding of aluminum. A similar facture pattern is established after tension in a specimen welded with transverse oscillation, compared to one without oscillation, as shown in Figure 9b. However, for the circular oscillating specimen, the fracture starts from the heat-affected zone of the weld, and ends at the weld fusion zone, as indicated in Figure 9c. In addition, no obvious pore is observed in the fractures of the welds with transverse oscillation and circular oscillation, as evidenced in Figure 9e,f. This sheds light on the fact that the undercut defect of the surface weld is a weak part of the welded joint, and can be improved by the Appl. Sci. 2019, 9, 5096 8 of 11 application of circular oscillation. It should be noted that all the fractures of the specimen are 45 • from the direction of the tensile stress applied, which is the ductile shear fracture [27]. Appl. Sci. 2019, 9, x FOR PEER REVIEW 8 of 11 defect of the surface weld is a weak part of the welded joint, and can be improved by the application of circular oscillation. It should be noted that all the fractures of the specimen are 45° from the direction of the tensile stress applied, which is the ductile shear fracture [27]. In order to explore the causes of the variation of tensile strength and plasticity of butt joints, the fracture surfaces of the tensile samples were observed by SEM, as shown in Figure 10. The morphology of the joint welded without beam oscillation exhibited a fracture with characteristic pores and dimples, as shown in Figure 9d and 10a. This indicated that the test specimen failed as a ductile fracture with the feature of dimple/microvoid coalescence. The pores worsened the stress concentration in the weld during tensile testing, and resulted in the reduced tensile strength and toughness of the weld. The transversely oscillating specimen also showed a ductile fracture, which conforms a dimple/microvoid coalescence pattern in which the fracture has many dimples but few pores, as displayed in Figure 10b,c. In the circularly oscillating specimen, numerous cavities were seen on the fractured blade, as shown in Figure 9f. This was ascribed to the vast reinforced phase particles that were pulled out from the weld bead during tensile testing. It is evident that the appearance of reinforced phase particles led to the higher dispersion strengthening of the weld bead. The formation of dense small dimples with a higher depth-diameter ratio confirmed the hypothesis that the plasticity of a joint welded with circular oscillation was good, which was consistent with the tensile stress-stain curve and the macroscopic necking phenomenon of the tensile samples. In order to explore the causes of the variation of tensile strength and plasticity of butt joints, the fracture surfaces of the tensile samples were observed by SEM, as shown in Figure 10. The morphology of the joint welded without beam oscillation exhibited a fracture with characteristic pores and dimples, as shown in Figure 9d and 10a. This indicated that the test specimen failed as a ductile fracture with the feature of dimple/microvoid coalescence. The pores worsened the stress concentration in the weld during tensile testing, and resulted in the reduced tensile strength and toughness of the weld. The transversely oscillating specimen also showed a ductile fracture, which conforms a dimple/microvoid coalescence pattern in which the fracture has many dimples but few pores, as displayed in Figure 10b,c. In the circularly oscillating specimen, numerous cavities were seen on the fractured blade, as shown in Figure 9f. This was ascribed to the vast reinforced phase particles that were pulled out from the weld bead during tensile testing. It is evident that the appearance of reinforced phase particles led to the higher dispersion strengthening of the weld bead. The formation of dense small dimples with a higher depth-diameter ratio confirmed the hypothesis that the plasticity of a joint welded with circular oscillation was good, which was consistent with the tensile stress-stain curve and the macroscopic necking phenomenon of the tensile samples. defect of the surface weld is a weak part of the welded joint, and can be improved by the application of circular oscillation. It should be noted that all the fractures of the specimen are 45° from the direction of the tensile stress applied, which is the ductile shear fracture [27]. In order to explore the causes of the variation of tensile strength and plasticity of butt joints, the fracture surfaces of the tensile samples were observed by SEM, as shown in Figure 10. The morphology of the joint welded without beam oscillation exhibited a fracture with characteristic pores and dimples, as shown in Figure 9d and 10a. This indicated that the test specimen failed as a ductile fracture with the feature of dimple/microvoid coalescence. The pores worsened the stress concentration in the weld during tensile testing, and resulted in the reduced tensile strength and toughness of the weld. The transversely oscillating specimen also showed a ductile fracture, which conforms a dimple/microvoid coalescence pattern in which the fracture has many dimples but few pores, as displayed in Figure 10b,c. In the circularly oscillating specimen, numerous cavities were seen on the fractured blade, as shown in Figure 9f. This was ascribed to the vast reinforced phase particles that were pulled out from the weld bead during tensile testing. It is evident that the appearance of reinforced phase particles led to the higher dispersion strengthening of the weld bead. The formation of dense small dimples with a higher depth-diameter ratio confirmed the hypothesis that the plasticity of a joint welded with circular oscillation was good, which was consistent with the tensile stress-stain curve and the macroscopic necking phenomenon of the tensile samples.

Conclusions
Motivated by the scarcity of research on the welded joint of a fiber laser welding aluminum alloy with beam oscillation, we conducted a laser welding experiment on 5056 aluminum alloy specimens, and analyzed the appearance, microstructure, hardness, and strength properties of welded joints. The following conclusions were made after the experiment.
(1) Beam oscillation causes great improvements in the weld formation of laser butt welding of aluminum alloy. The surface of the weld with circular oscillation is well formed and has good consistency without spatter, undercut, and root sagging. (2) Beam oscillation results in a significant improvement of the weld microstructure of laser butt welding of aluminum alloy. The circular oscillating weld has the features of fine grain and uniformly dispersed dendrites in the strengthening phase, and the largest microhardness is observed in the fusion zone. (3) Beam oscillation is significantly inhibitory to the porosity of the weld, and the inhibitory effect is even stronger with circular oscillation than with transverse oscillation. There is no pore in the tensile fracture of the circular oscillating weld, but a small number of pores can be seen in the corresponding position of a transversely oscillating weld, and many more pores with a large volume are noted without oscillation. Beam oscillation has few effects on the tensile strength of the weld, but exerts a great influence on the elongation of the weld, which is evidenced by the fact that the largest elongation of the weld is associated with circular oscillation.

Conclusions
Motivated by the scarcity of research on the welded joint of a fiber laser welding aluminum alloy with beam oscillation, we conducted a laser welding experiment on 5056 aluminum alloy specimens, and analyzed the appearance, microstructure, hardness, and strength properties of welded joints. The following conclusions were made after the experiment.
(1) Beam oscillation causes great improvements in the weld formation of laser butt welding of aluminum alloy. The surface of the weld with circular oscillation is well formed and has good consistency without spatter, undercut, and root sagging. (2) Beam oscillation results in a significant improvement of the weld microstructure of laser butt welding of aluminum alloy. The circular oscillating weld has the features of fine grain and uniformly dispersed dendrites in the strengthening phase, and the largest microhardness is observed in the fusion zone. (3) Beam oscillation is significantly inhibitory to the porosity of the weld, and the inhibitory effect is even stronger with circular oscillation than with transverse oscillation. There is no pore in the tensile fracture of the circular oscillating weld, but a small number of pores can be seen in the corresponding position of a transversely oscillating weld, and many more pores with a large volume are noted without oscillation. Beam oscillation has few effects on the tensile strength of the weld, but exerts a great influence on the elongation of the weld, which is evidenced by the fact that the largest elongation of the weld is associated with circular oscillation.