Mechanical properties and failure mechanisms of Zr-based amorphous alloys with various element compositions under different strain rates

This study investigated the mechanical properties, glass-forming ability (GFA), and microstructural evolution of ZrCuAlNi amorphous alloys with various niobium (Nb) and silver (Ag) additions under static (1 × 10–1 – 1 × 10–3 s−1) and dynamic (3 × 103 – 5 × 103 s−1) loading conditions. The x-ray diffraction (XRD) results revealed that all of the alloys had an amorphous structure, and the GFA improved with an increasing Ag content. The Zr56Cu24Al9Ni7Nb1Ag3 alloy showed the highest flow stress among the various alloys under both strain rate ranges. For all of the alloys, the fracture strain increased with an increasing Ag content. Moreover, the strain rate sensitivity increased with increasing strain rate. The scanning electron microscopy observations showed that the fracture surfaces had a dimple structure. As the Ag content increased, the dimple structure changed from a smooth to molten appearance. In addition, the dimple structure density increased with an increasing strain rate. The results present that the ductility of the ZrCuAlNi amorphous alloys could be improved by increasing the Ag element content.


Introduction
Metals are one of the earliest studied materials and are widely used in all manner of industrial and engineering applications. In recent decades, the goal of metal research has evolved from the preparation of single alloys to that of multicomponent composite alloys. Amorphous alloys, also known as metallic glasses, non-crystalline metals, or glassy metals, were first developed in the 1970 s [1]. Compared to pure metals, amorphous alloys have many unique and advantageous properties, including improved strength and toughness. Consequently, they are one of the most intensively researched metallic systems nowadays [2][3][4].
When crystalline materials are subjected to an external force, dislocations are produced, which lead to plastic deformation (i.e., dislocation slip). However, amorphous materials have only a short-range atomic structure [5], and their atoms are irregularly arranged. Thus, their structure is similar to that of a liquid, and hence they do not undergo dislocation slip or present any of the other defects exhibited by crystalline materials. Instead, they deform due to the emergence of local high-shear bands. As the load applied to the material increases, these shear bands multiply and cross one another, thereby blocking further deformation and fracturing. Moreover, since amorphous alloys do not have a long-range ordered atomic structure, they lack the lattice structure of general crystalline materials. Consequently, their x-ray diffraction (XRD) patterns reveal only wide diffraction peaks at low angles, which are very different from the sharp and intense peaks of metals and alloys with a periodic arrangement of crystalline atoms [6][7][8].
However, they have many favorable mechanical and electrical properties, including high mechanical strength, good wear resistance, higher hardness, superior magnetic properties, and enhanced corrosion resistance [9][10][11][12][13]. Consequently, they have attracted intensive research attention in recent years, and many Zr-, Fe-, Cu-, Ti-, Co-, Al-, Ni-, La-, and Pd-based alloys have been developed [14][15][16][17]. Amorphous alloys are now widely used in the electronics, biomedicine, aerospace, and engineering industries and are available in many sizes ranging from micron-sized thin strips to block-shaped rods with diameters of several millimeter.
Zr-based amorphous alloys have attracted particular attention in the recent literature due to their excellent mechanical hardness and corrosion resistance [18][19][20][21]. They have already been used to manufacture golf club heads, diving equipment, knives, watch surfaces, and tools, and there is growing interest in their potential use for medical applications. During these production process, materials will be surfaced the different deformation condition with the strain rate and temperature [22,23]. Hence, it needs to the strain rate efffect on the mechanical property of amorphous alloy. Compared with Al-, Fe-, Co-, Ni-, and Ti-based materials, Zr-based materials have better glass-forming ability (GFA). In addition, Zr-, Cu-, Pu-, Pd-, and Al-based alloys have superior ductility, a lower cost, and higher chemical stability. Accordingly, the present study selects ZrCuAlNi as the substrate material for the preparation of amorphous alloys with superior mechanical and thermal properties through the addition of carefully controlled quantities of Nb and Ag.

Material preparation and experimental procedures
Four Zr-based amorphous alloys were prepared with compositions of Zr 56 Cu 24 Al 9 Ni 7 Nb 4-x Ag x , where x = 0, 1, 2, 3 is the atomic percentage of Ag. For each alloy system, the materials were placed in a vacuum arc melting furnace filled with high-purity argon gas to prevent oxidation. The materials were melted and re-melted five times to ensure a uniform distribution of the elements and were then quickly pumped into a water-cooled copper mold for rapid cooling. The alloys were cast in the form of cylindrical rods with a diameter of 4 mm and length of 60-70 mm. Once the rods were fully cooled, they were cut into amorphous alloy specimens with a length of 4 mm using a slow cutting machine. (The dimension of the compression test was 4 × 4 mm, height and diameter).
The structures of the various alloys were examined using a multifunctional x-ray diffractometer (D8 DISCOVER, Bruker AXS). The internal phase transition temperatures of the alloys were then analyzed by a hightemperature differential scanning calorimeter (HT-DSC, Netzsch 404 F3) at heating rate of 10°C min -1 . Static compression tests were performed at strain rates in the range of 1 × 10 −1 -1 × 10 -3 s −1 on a universal testing machine (MTS 810). The dynamic impact response of the alloys was then examined using a split-Hopkinson pressure bar (SHPB) system at strain rates ranging from 3 × 10 3 to 5 × 10 3 s −1 . The surfaces of the fractured specimens were observed using scanning electron microscopy (SEM, HITACHI SU-5000), and the elemental compositions were confirmed using energy dispersive x-ray spectroscopy (EDX). Finally, the micro-hardness of the various specimens was measured using a Vickers hardness tester (JC1000A, JCHIA).

Results and discussion
3.1. XRD patterns and DSC thermal stability analysis Figure 1 presents the XRD analysis results for the Zr 56 Cu 24 Al 9 Ni 7 Nb 4-x Ag x (x = 0, 1, 2, 3 at%) alloys. All of the patterns show a broad diffraction peak without any sharp Bragg peaks. Hence, it is inferred that all of the alloys have an amorphous structure. As the level of Ag addition increases, the diffusion peaks broaden and their intensity reduces. Thus, it is further inferred that the addition of Ag improves the GFA of the ZrCuAlNi amorphous alloy.
The thermal stability properties of the Zr 56 Cu 24 A l9 Ni 7 Nb 4-x Ag x (x = 0, 1, 2, 3 at%) alloys were analyzed by DSC. Briefly, specimens with weights of 5-50 mg were heated from 303 K to 1273 K in an N 2 environment at a rate of 10 K mm −1 , and the DSC curves were obtained by observing the change in heat flow (see figure 2). The glass transition temperature T g , crystallization temperature T x , melting temperature T m , and liquid phase temperature T l , of the alloys were obtained from the DSC curves using the differential scanning carlorimetry method. The corresponding results are presented in table 1. The supercooled liquid phase region, ΔT x , simplified glass transition temperature T rg , and γ value were calculated using equations (1)-(3), respectively. The calculated values are presented in table 1 for each alloy. It is seen that the addition of Ag results in a melting temperature, T m , and liquid phase temperature, T l , significantly lower than the melting point of the original Zrbased amorphous alloy [24,25].
Among all the alloys, the Zr 56 Cu 24 Al 9 Ni 7 Nb 3 Ag 3 alloy has the widest supercooled liquid phase region, i.e., ΔT x = 62 K. Moreover, the γ coefficient value is also higher than those of the alloy with 3 at% Ag. An inference analysis based on the experimental DSC values showed that the Zr 56 Cu 24 Al 9 Ni 7 Nb 1 Ag 3 alloy had the best GFA. The DSC results are thus consistent with the XRD analysis results presented in figure 1, which show that the diffraction peak intensity of the 3 at% Ag alloy is the lowest among all the alloys.     Figures 3(a)-(d) present the static and dynamic stress-strain curves of the Zr 56 Cu 24 Al 9 Ni 7 Nb 4-x Ag x alloys with 0, 1, 2 and 3 at% Ag addition, respectively. Under static strain rates, the ultimate stress and ductility decrease slightly with an increasing strain rate. Furthermore, under dynamic loading, the ultimate stress increases with increasing strain rate and the content of Ag. However, for the same strain, the flow stress increases with an increasing strain rate. In other words, the ductility of the amorphous alloys increases at lower strain rates. For both strain rate ranges, the ultimate stress and strain both increase with an increasing Ag content. In other words, the replacement of Nb with Ag yields an effective improvement in both the strength and the ductility of the Zr 56 Cu 24 Al 9 Ni 7 Nb 4-x Ag x alloys.

Strain rate sensitivity coefficient
The stress-strain curves in figures 3(a)-(d) show that, under the same strain, the stress increases with an increasing strain rate for all of the specimens. To further understand the relationship between the flow stress and the strain under low and high strain rate conditions, the variation of the flow stress with the strain rate was investigated for fixed strains of ε = 0.01 and 0.03, respectively. The corresponding results are presented in figures 4(a) and (b) for the static and dynamic loading conditions, respectively. It is seen that, for a constant strain, the flow stress increases approximately linearly with an increasing strain rate for all of the alloys and all of the strain rates. Moreover, the rate at which the flow stress increases with an increasing strain rate is slightly higher at a strain of 0.03 than at a strain of 0.01. The effect of the strain rate on the flow stress was quantified by the strain rate sensitivity coefficient, defined as [26,27]. where σ 1 and σ 2 are the strain values under two different strain rates and  e 1 and  e 2 are the corresponding strain rate. In other words, for materials with a larger β value, the stress is more sensitive to the strain rate. Moreover, for the same strain rate range, the difference in stress values is comparatively larger. Figures 5(a) and (b) show the variation of the strain rate sensitivity with the true strain for the four amorphous Zr alloys under static and dynamic loading conditions, respectively. For all of the alloys, the strain rate sensitivity increases with increasing strain. Moreover, the strain rate sensitivity reduces slightly with an increasing Ag content. However, while a higher Nb content results in a greater strain rate sensitivity, it also leads to a lower ductility and strength (see figure 3). Conversely, a higher Ag content reduces the strain rate sensitivity slightly, but improves the ductility and failure strength. It dues to the addition of Ag can enhance the GFA and stablize the amorphous phase. This result corresponds with reports of similar compositions containing various contents of Ag [28,29].

Fracture feature analysis
In the static compression tests, all of the specimens fractured along the 45°shear plane ( figure 6(a)), forming staggered shear bands ( figure 6(b)). A similar failure tendency was also noted for the specimens tested under dynamic shear rates. In other words, for both strain rate ranges, the alloy specimens fractured predominantly as a result of brittle failure.
As shown in figure 7(a), the fracture surface of the Zr 56 Cu 24 Al 9 Ni 7 Nb 4 alloy with no Ag addition tested at a strain rate of 10 −3 s −1 has a dimple-like structure. As the strain rate increases to 10 -1 s −1 , the dimple structure becomes broader and flatter, and the dimple density decreases accordingly. As shown in figures 7(c)-(e), as the Ag content increases, the dimple structure changes from a broad and flat structure to a pulled morphology with a slightly increased dimple density. In other words, the alloy specimens become more ductile as the Ag content increases, as shown also in the stress-strain curves in figure 3.
As shown in figures 8(a)-(e), the fracture surfaces of the alloy samples impacted under dynamic strain rates show the same dimple-like characteristic as those tested under static conditions. However, the dimples generally have a larger size and a lower density. For the alloy with 0 at% Ag, the dimple structure again becomes broader and flatter as the strain rate increases (figures 8(a) and (b)). A similar tendency is observed for the alloys containing 1 at% and 2 at% Ag, respectively, under a high strain rate of 5 × 10 3 s −1 (figures 8(c) and (d)). However, for the alloy with 3% Ag, the dimples not only become flatter and broader, but a local melting effect also occurs ( figure 8(e)). In other words, it appears that the addition of 3 at% Ag not only increases the ductility of the alloy, but also reduces the melting point, and thus results in local melting of the fracture surface under  high strain rate conditions. Therefore, the Ag content not only enhances the plasticity strain but also increases the ductility. It dues to the Ag can form a higher bond with Zr, Cu, Al and Ni element, and increase number of shear bands [30].

Hardness analysis
The hardness of the various alloys was evaluated by micro-Vickers hardness tests performed under a load of 100 g and dwell time of 10 s. Table 2 shows the average Vickers hardness values of the four amorphous alloy, as evaluated over five times repeated measurements in each case. Furhtermore, in oder to ensure the data was preciseness. We indent the center and near center location of the specimen. It is seen that the hardness increases with an increasing Ag content and thus results in a higher ultimate stress, as shown in figure 3.

Conclusion
Zr 56 Cu 24 Al 9 Ni 7 Nb 4-x Ag x alloys with various Ag contents (x = 0, 1, 2, 3 at%) were prepared using a vacuum arc melting process. The alloys were then tested under both static (1 × 10 −1 -1 × 10 -3 s −1 ) and dynamic (3 × 10 3 -5 × 10 3 s −1 ) loading conditions. The XRD results showed that all of the alloys had an amorphous structure with a wide diffraction peak at θ = 40°. The intensity of the diffusion peak reduced with an increasing Ag content. Thus, the Zr 56 Cu 24 Al 9 Ni 7 Nb 1 Ag 3 alloy was inferred to have the highest GFA. The DSC thermal property analysis results showed that the Zr 56 Cu 24 Al 9 Ni 7 Nb 3 Ag 1 alloy had the highest ΔT x and γ values. By contrast, the Zr 56 Cu 24 Al 9 Ni 7 Nb 1 Ag 3 alloy had the highest T rg value and hence the best GFA.
For both strain rate ranges, the ultimate stress and ductility decreased slightly with an increasing strain rate. However, the ultimate stress and strain both increased with an increasing Ag content, indicating that a lower Nb content and higher Ag content improved the strength and ductility of the Zr 56 Cu 24 Al 9 Ni 7 Nb 4-x Ag x alloys. The SEM observations showed that the fracture surfaces contained dimple structures under both loading conditions. For both strain rate ranges, the dimple structures became broader and flatter as the Ag content increased. Furthermore, under a high strain rate of 5 × 10 3 s −1 , the fracture surface of the alloy with 3 at% Ag addition showed a local melting effect, which suggested that a higher Ag content reduced the melting point of the Zr 56 Cu 24 Al 9 Ni 7 Nb 4-x Ag x alloy system. Finally, the micro-Vickers hardness test results showed that the hardness of the alloys increased with an increasing Ag content and led to a higher ultimate stress as a result. It due to the Ag  can enhance the bond between the Zr and Cu. At the same time, the Ag element addition can improve the GFA and increase the thermal stability of amorphous alloy. Hence, the strength of Zr-based amorphous alloy increases due to the addition of Ag and the hardness also increases.