Enhancing Fracture Toughness, Strength and Ductility of Zr_58.75Cu_21.15Fe_4.7Al_9.4Nb₆ Bulk Metallic Glass via Ultrasound Excitation Technique

Abstract

The inherent brittleness and limited toughness of bulk metallic glasses (BMGs) remain critical challenges for their application as structural engineering materials. In this study, ultrasonic excitation was applied to Zr_58.75Cu_21.15Fe_4.7Al_9.4Nb₆ BMG with the aim of enhancing its mechanical performance. The results reveal that ultrasonic treatment significantly increases the fracture toughness by approximately 28% and induces a pronounced plastic deformation plateau following yielding. This improvement in both strength and ductility is attributed to the formation of nanoscale crystalline phases and ultrasound-induced phase separation within the amorphous matrix, which collectively promote shear band multiplication and inhibit crack propagation.

1. Introduction

Bulk metallic glasses (BMGs) exhibit a range of remarkable properties that make them attractive for various applications, including a high elastic strain limit, high strength and hardness, and excellent corrosion resistance [1,2,3,4,5]. However, their lack of an inherent microstructure often results in severe strain localization within narrow shear bands. Consequently, BMGs typically present limited macroscopic plasticity and low fracture toughness, sometimes fracturing without any measurable plastic strain at room temperature. These characteristics significantly restrict their use as structural materials.

For brittle rare-earth and ferrous-based metal glasses [6,7], fracture toughness values are typically just above 1 MPa m^0.5. The notch toughness of the first commercial bulk metallic glass, Vitreloy 1 (Zr_41.2Ti_13.8Cu_12.5Ni₁₀Be_22.5) [8], ranges from 55 to 59 MPa m^0.5 [9], which is significantly lower than that of ductile metals such as low-carbon steels (>200 MPa m^0.5) [10]. Enhancing the ductility of BMGs requires the activation of multiple shear bands with reduced shear band spacing, thereby mitigating the rapid propagation of individual shear bands and improving overall toughness.

Recent studies have identified structural heterogeneities in BMGs, characterized by the presence of soft and hard regions [11,12,13]. The soft regions exhibit a viscoelastic flow feature, facilitating deformation and shear band nucleation. Based on those deformation mechanisms, various novel approaches have been explored depending on the mode of external force [14,15,16,17]. For example, thermal cycling [14,15] can rejuvenate the structure of BMGs to a heterogeneous state, resulting in the improvement of the plasticity of BMGs. Ion radiation [16] can induce the appearance of vacancy-like defects in the BMG structure, which can also enhance its plasticity. Laser shock peening [17] can introduce compressive residual stresses and the formation of localized shear bands, enhancing the ductility of BMGs. However, these methods typically compromise material strength while improving ductility.

Ultrasound excitation represents another potential method for modifying the atomic structure of BMGs through forced atomic and molecular vibrations. Recent studies have demonstrated that ultrasonic vibration near the glass transition temperature T_g accelerates the crystallization in Pd₄₀Ni₄₀P₂₀ BMG [18,19], while ultrasonic nanocrystal surface modification has been shown to induce microstructural inhomogeneity in Vit1b BMG [20]. However, whether ultrasonic excitation can enhance the fracture toughness of BMGs remains an open question.

In this study, we applied intensive 20 kHz ultrasound excitation to a Zr_58.75Cu_21.15Fe_4.7Al_9.4Nb₆ BMG and observed a significant improvement in its structure toughness. A transmission electron microscope (TEM) was employed to analyze the structural changes before and after ultrasonic treatment, as well as the shear band structure after fracture. The underlying mechanisms through which ultrasound excitation enhances the fracture toughness of BMGs are discussed. These findings provide valuable insights for the development of more ductile BMGs.

2. Experimental

Alloy ingots with a composition of Zr_58.75Cu_21.15Fe_4.7Al_9.4Nb₆ were produced through the process of arc melting. Elemental metals of exceptional purity (≥99.5% by weight) were handled within a titanium-gettered argon environment. To guarantee uniformity in the composition of the alloy, each ingot underwent a minimum of four remelting processes. Subsequently, the alloy was suction-cast into a copper mold, yielding plates measuring 3 mm by 6.5 mm by 40 mm.

A tailor-made ultrasonic apparatus, featuring a KNbO₃ piezoelectric transducer with a resonant frequency of 20 kHz and a cylindrical emitter measuring 20 mm in diameter and 500 mm in length, was utilized to administer ultrasonic stimulation. The as-cast plate was affixed to a supporting platform using adhesive, positioned directly beneath the emitter. The emitter was lowered via a precision displacement controller until its end face made tangential contact with the plate surface. Longitudinal ultrasound waves, possessing an amplitude near 15 micrometers, were administered for periods of 15 min and 30 min.

Samples intended for the evaluation of fracture resistance were precisely fabricated from both the processed and unprocessed panels. These specimens were characterized by dimensions with a thickness of 3 mm, a width of 6 mm, and a distance between supports of 24 mm. The six surfaces of the sample were successively ground with 400, 800, 1200 and 2000 grit sandpapers, and then polished to a mirror finish with diamond grinding paste of 5 μm particle size. They were meticulously smoothed and buffed to achieve a reflective surface before assessment. A continuous notch was created by means of wire-electrode discharge machining, featuring a root radius close to 300 micrometers and a notch approximately 0.4 times the specimen width. A cycle of pre-cracking to induce fatigue was applied using a WDW-10E (Shanghai, China) universal testing apparatus under ambient conditions. The cumulative length of the crack, combining both the notch and the pre-existing fatigue-induced crack, was maintained between 0.45 and 0.55 times the width (W).

The fracture toughness of the material was assessed through the application of three-point bend tests utilizing the WDW-10E apparatus at ambient temperature. The thin-foil specimens intended for the transmission electron microscope (TEM, Tecnai G2 F20, FEI, Hillsboro, OR, USA) were crafted from samples that had undergone a mechanical polish (with thicknesses of 30–40 micrometers) by employing the method of twin-jet electropolishing (Buehler, Lake Bluff, IL, USA); the volume ratio of the double spray electrolytic polishing solution is 5:85:5, consisting of perchloric acid, methanol and deionized water.

The amorphous nature of the samples was verified by X-ray diffraction (XRD, Bruker D8 ADVANCE diffractometer in Billerica, MA, USA) with Cu Kα radiation. Crystallization kinetics were investigated via differential scanning calorimetry (DSC, Netzsch STA 449C, Sürburg, Germany) under high-purity argon atmosphere, employing a constant heating rate of 5 K/min. Microstructural observations of lateral and fracture surfaces were examined using field-emission scanning electron microscopy (SEM, LEO1430VP, Zeiss, Jena, Germany). For atomic-scale structural analysis, high-resolution transmission electron microscopy (HRTEM, Tecnai G2 F20, FEI, Hillsboro, OR, USA), coupled with an energy-dispersive X-ray spectroscopy (EDX) system for concurrent elemental mapping.

3. Results

3.1. Characteristic Structure of Both As-Cast and Ultrasonically Processed Specimens

Figure 1 presents X-ray diffraction patterns for the Zr_58.75Cu_21.15Fe_4.7Al_9.4Nb₆ plate samples in the as-cast state (U-0) and after ultrasonic treatment for 15 min (U-15) and 30 min (U-30). Every specimen displays a wide diffuse peak indicative of a non-crystalline framework, lacking any distinct peaks associated with crystalline structures. This substantiates the fact that the amorphous state is preserved in both the non-treated and the ultrasonically processed specimens.

3.2. Thermal Characterization

The differential scanning calorimetry (DSC) traces for the U-0, U-15, and U-30 samples, as depicted in Figure 2, reveal that at a rise of 5 kelvins per minute, each sample presents a clear glass transition, progresses through a sharp superliquid region, and undergoes crystallization. It is important to highlight that the starting point of the glass transition temperature, T_g, is 649 K, and the initial crystallization, T_x, is 706 K, which are almost identical across the samples. The width of the undercooled liquid-phase region, T_x-T_g, is 57 K, which is not significantly different from the width of the cold liquid-phase region in most Zr-based BMGs [21], indicating that it has a high glass-forming ability. This suggests that the ultrasonic exposure has a negligible impact on the thermal stability of the alloy.

3.3. Fracture Toughness Properties

The assessment of fracture toughness was performed by conducting three-point bending tests on fatigue pre-cracked samples. Figure 3 presents the related load-versus-displacement graphs. When examining the untreated cast specimen (U-0) relative to those subjected to ultrasonic treatment (U-15 and U-30), there is a noticeable enhancement in maximum loads, rising from 1.515 kN for the U-0 sample to 1.900 kN and 1.924 kN (U-15 and U-30, respectively). Additionally, the appearance of a distinct plastic plateau following yielding in the U-15 and U-30 samples suggests enhanced ductility due to ultrasonic excitation.

Computation of the peak fracture robustness with the open-type fracture toughness, K_Ic, was performed in adherence with the ASTM E399 standard [22], employing Equations (1) and (2).

$$K={\displaystyle \frac{PS}{B{W}^{1.5}}}f({\displaystyle \frac{a}{W}})$$

(1)

$$f({\displaystyle \frac{a}{W}})=3\sqrt{{\displaystyle \frac{a}{W}}}{\displaystyle \frac{1.99-(\frac{a}{W})(1-\frac{a}{W})[2.15-3.93\frac{a}{W}+2.7{(\frac{a}{W})}^2]}{2(1+2\frac{a}{W}){(1-\frac{a}{W})}^{1.5}}}$$

(2)

Here, P represents the maximum stress applied to the specimens, S refers to the distance between supports, and B is the measurement of thickness. The width of the sample and the length of the pre-crack are, respectively, denoted by W and a. The calculated K_max values are 69 MPa m^0.5 for U-0, 87 MPa m^0.5 for U-15 and 88 MPa m^0.5 for U-30, indicating an approximate 28% improvement in fracture toughness as a result of ultrasonic treatment.

3.4. Shear Bands and Fracture Morphologies

Figure 4 illustrates the side surfaces of the U-0 (Figure 4a) and U-30 (Figure 4b) specimens after bending. Shear bands are clearly observed around the notch area in all samples. However, in the U-30 sample, the shear bands (highlighted by green boxes) are more densely distributed and exhibit increased branching and interaction. Additionally, the size of the shear band area, marked by green circles, is significantly greater in U-30, with an approximate diameter of 0.7 mm, when contrasted with the roughly 0.3 mm diameter of U-0.

The fractographic analysis reveals that ultrasonically modified specimens exhibit characteristic meandering crack paths (as demarcated by yellow arrows in the representative SEM micrograph, Figure 4b), indicative of crack deflection mechanics through successive bifurcation events. This tortuous propagation morphology demonstrates the following: (1) enhanced extrinsic toughening via crack path optimization; (2) progressive energy dissipation through secondary crack branching; (3) residual stress-induced crack closure effects. These synergetic toughening mechanisms collectively elevate the fracture work compared to as-cast specimens, fundamentally altering the failure mode from brittle cleavage to ductile tearing.

3.5. Evolution of Microstructure Before and After Ultrasonic Treatment

Figure 5a and Figure 6a show low- and high-magnification TEM images of U-0 before the three-point bending test. It shows no contrast in the image, and the corresponding SAED pattern shown in the inset of Figure 5a indicates that there is only a diffusive halo, confirming the amorphous nature of this alloy. The high-resolution TEM (HRTEM) image also indicates that there is no nanocrystal (see the inset of Figure 6a); the image also confirms an amorphous structure for U-0. However, for U-30, as shown in Figure 5b, there are ~20 nm dark phases (highlighted by green circles) on the glassy matrix; as the fraction of them is less than 10%, the SAED pattern image only shows a diffuse halo and a small number of diffraction spots. However, the HRTEM image of U-30 (Figure 6b) shows that the dark phases are nanocrystals, which indicates that the alloy is partially crystallized via ultrasonic treatment, showing lattice fringes with equal spacing (highlighted by blue circles). Figure 7a shows the scanning TEM (STEM) image of U-30. From the elemental mapping of Zr, Cu, Fe, Al and Nb, as shown in Figure 7b–f, it can be found that the main element of the precipitates is Cu-rich, which indicates that the composition distribution is not uniform.

4. Discussion

As mentioned above, the ultrasonic-treated samples exhibited enhanced yield stress and a distinct plastic plateau following yielding, indicating improved plasticity compared to the as-cast sample. A thorough examination of the microscopic characteristics in both the U-0 and U-30 specimens, before and after deformation, was performed to clarify the underlying mechanisms that account for the noted enhancements in strength and ductility.

Figure 5a displays a transmission electron microscopy (TEM) image at low magnification for the U-0 sample before it was deformed. The homogenous appearance in this image, coupled with the appearance of a blurred ring in the selected area electron diffraction (SAED) pattern, where the diameter of the SAED pattern was about 1 μm, corroborates the sample’s completely amorphous constitution. Moreover, as shown in Figure 6a, an examination with high-resolution transmission electron microscopy (HRTEM) substantiates this finding, as it shows the absence of any nanocrystalline structures.

Conversely, the U-30 specimen exhibits considerable alterations in its structure following exposure to ultrasonic processing. As shown in Figure 5b, small dark contrast features (~20 nm) appear within the amorphous matrix. The SAED pattern displays a diffuse halo and a small number of diffraction spots because the features occupy a minor volume fraction (less than 10%); HRTEM (Figure 6b) confirms that they are nanocrystals, as evidenced by the presence of lattice fringes in the inset. These findings suggest partial crystallization induced by ultrasonic excitation.

It is probable that the appearance of these nanocrystals can be ascribed to the inherent metastability of bulk metallic glasses (BMGs), which are prone to crystallizing when subjected to external factors like thermal exposure [23,24,25] or mechanical stress [5,13]. Ultrasound generates strong forced vibration, enhancing atomic mobility and diffusivity through mechanisms such as acoustic cavitation and streaming [26]. Recently, similar ultrasonic-induced crystallization phenomena have been reported in Pd-based BMGs by Ichitsubo et al. [18,19].

Therefore, it is reasonable to deduce that ultrasonic excitation promotes atomic rearrangement, allowing atoms to occupy lower-energy configurations. This enhanced mobility facilitates structural relaxation and eventual crystallization. Under such conditions, atomic jumps deviate from conventional diffusion behavior. The cumulative effect of such anomalous displacements perturbs the thermodynamic energy landscape, ultimately driving the nucleation of crystalline phases through localized energy minimization. According to Fan et al. [24], strength improvements occur when the crystalline volume fraction remains below ~30%, as also supported by other studies [23,25]. This study reveals that nanocrystals, which are more minute than typical shear bands (approximately 10 nanometers), are dispersed haphazardly throughout the material, obstructing the extension of shear bands and preserving the amorphous structure. Such an enhancement at the microscopic level contributes to a rise in both the resistance to breaking and yield strength.

Moreover, ultrasonic excitation likely promotes structural relaxation, driving atoms into more stable positions. This process increases packing density in formerly loose-packed regions, reducing local structural weaknesses and contributing to overall strength and ductility. When the size of the nano-particles is much smaller than that of the shear band and the nano-particles are separated by a glassy matrix, the deformation of the shear band is dominated by the glassy matrix, and the nanocrystals may inhibit the deformation of the shear bands [23]. Thus, it can be inferred that a small amount of precipitated nanocrystals leads to an increase in fracture stress and yield stress. The causes of the increase in fracture stress and yield stress may be as follows: Firstly, the high-strength nanocrystals serve as robust obstacles like nails that effectively restrict the propagation of shear bands and arrest crack advancement through microstructural pinning effects. In this work, 2–5 nm nanocrystals precipitated on the glassy matrix during the ultrasound excitation for U-30 (Figure 6b), and their size was much smaller than that of the shear band. The volume fraction of the crystalline part of U-30 was less than 10%, and the amorphous matrix was mainly retained. In this stress-concentrated configuration, these localized high-stress nanocrystals distributed along shear bands acted as effective obstacles, which not only obstructed further propagation of the primary shear bands but also delayed crack initiation and growth. Secondly, owing to the pronounced stress gradient at crystalline–amorphous interfaces, primary shear bands preferentially nucleate through stress-activated free volume coalescence, subsequently undergoing directional propagation governed by interfacial shear stress anisotropy. These secondary shear bands form hierarchical branching networks with wing-like primary counterparts, creating a three-dimensional stress redistribution architecture that enables (1) strain accommodation through coordinated shear band intersection and bifurcation; (2) energy dissipation via interface-mediated dislocation–STZ (shear transformation zone) interactions; and (3) crack blunting by shear band network rearrangement. This multiscale deformation synergy effectively delays strain localization while maintaining structural integrity under loading. Consequently, the partial crystallization, in tandem, improves both strength and the capacity of plastic deformation [27,28,29].

To deepen the insight into the impact of ultrasound, the elemental composition distribution of specimen U-30 was examined. Figure 7a shows a scanning TEM (STEM) image with visible spherical precipitates, consistent with the dark contrast features observed in Figure 5b. Elemental mapping (Figure 7b–f) reveals that these precipitates are Cu-rich, indicative of local chemical inhomogeneity. The elemental line scanning image of Figure 7g shows the distribution of Zr, Cu, Al, Fe and Nb elements along the blue arrow in Figure 7a, which indicates that the concentration of the Cu element reaches a peak in the Cu-rich area and then becomes relatively stable in the Cu-poor area.

Viewed through the lens of thermodynamics, the emergence of distinct phases can be attributed to the endothermic nature of mixing for Cu-Fe (13 kJ/mol) and Cu-Nb (3 kJ/mol) pairs [30], which promotes segregation. Conversely, the Fe-Nb pair, displaying an exothermic mixing response with an enthalpy of −16 kJ/mol, favors the creation of regions rich in Fe and Nb. Although Fe and Nb are present in low concentrations and not clearly resolved in EDS, Cu is abundant and has a diffusion coefficient significantly higher than that of Zr [30], making it more prone to mobilization during deformation. The suppression of Cu mobility by larger Zr and Al atoms likely results in the formation of Cu-rich zones [31]. Interestingly, phase separation is absent in the untreated U-0 sample, suggesting that ultrasonic excitation plays a critical role in triggering this phenomenon. This aligns with prior findings regarding phase separation in amorphous alloys triggered by stress or heat, as documented in earlier works [32]. Ultrasonic stress and prolonged localized heating may synergistically promote the development of heterogeneous microstructures in U-30.

These Cu-rich zones act as soft zones [33], reaching critical shear stress earlier than the surrounding harder glass matrix. According to the concept of shear transformation zones (STZs), soft regions initiate deformation, while adjacent hard regions constrain it, resulting in STZ branching and the multiplication of shear bands. This behavior contributes to increased crack resistance and improved plasticity, as observed in the U-30 sample. Thus, deformation-induced phase separation provides an additional explanation for the enhanced mechanical performance (especially fracture toughness) and plastic platform observed in U-30.

In summary, ultrasonic treatment can induce structural changes in bulk metallic glass. The internal microstructure variation mechanism of bulk metallic glasses under ultrasonic treatment is shown in Figure 8. The amorphous structure is in an unstable form. Ultrasonic treatment induces multiscale structural modifications in bulk metallic glasses through thermomechanical coupling effects, and the amorphous structure undergoes phase separation or crystallization. In the first place, under the action of ultrasonic treatment, the atoms in the bulk metallic glass are rearranged. High-frequency ultrasonic vibrations generate localized stress concentrations, activating cooperative shear transformations that facilitate atomic reconfiguration through free volume redistribution. This process triggers structural relaxation with a reduced activation energy barrier under the continuous action of ultrasonic treatment, and atoms in certain regions of bulk metallic glasses may crystallize, forming nanocrystals. The ultrasonic-induced stress gradient drives preferential nucleation of nanocrystalline phases at stress-concentrated regions. These nanocrystals form semi-coherent interfaces with the amorphous matrix. Simultaneously, phase separation occurs in some regions of the bulk metallic glass under the continuous action of ultrasonic treatment. There are two different kinds of phase separation. One form of phase separation is amorphous–amorphous phase separation. The compositional fluctuations promote the formation of copper-rich amorphous regions in zirconium-based amorphous materials. Another form of phase separation is crystal–amorphous phase separation. Localized crystallization occurs in phase-separated regions through stress-assisted diffusion. This improvement in both strength and ductility is attributed to the formation of nanoscale crystalline phases and ultrasound-induced phase separation within the amorphous matrix, which collectively promote shear band multiplication and inhibit crack propagation.

5. Conclusions

In this research, how ultrasonic excitation influences the mechanical characteristics and the internal microstructure of Zr_58.75Cu_21.15Fe_4.7Al_9.4Nb₆ alloy in its bulk metallic glass (BMG) form was systematically investigated. The principal results have been methodically examined and are concisely presented below:

(1)

Enhanced Fracture Toughness: The fracture toughness of Zr_58.75Cu_21.15Fe_4.7Al_9.4Nb₆ BMG was significantly improved by approximately 28% through ultrasound excitation, as demonstrated by three-point bending tests.

(2)

Ultrasound-Induced Nanocrystallization: Ultrasound treatment led to partial nanocrystallization of the BMG, which contributes to increased strength and toughness by inhibiting shear band propagation and promoting structural relaxation.

(3)

Ultrasound-Induced Phase Separation: The ultrasound excitation also induced chemical inhomogeneity in the form of Cu-rich phase-separated zones. These soft zones facilitate the initiation of shear transformation zones (STZs), while the surrounding hard matrix impedes their propagation, resulting in the multiplication of shear bands and enhanced resistance to crack growth.

Overall, ultrasonic treatment provides a promising strategy for tailoring the mechanical performance of bulk metallic glasses by promoting beneficial microstructural modifications such as partial nanocrystallization and phase separation.