*3.1. Microstructure*

Figure 1 illustrates the X-ray diffraction pattern of as-cast and treated Zr54.46Al9.9Ni4.95Cu29.7Pd0.99 BMGs. It shows a broad halo with the absence of detectable crystalline peaks, indicating that the treated sample remained a glassy structure.

*Materials* **2022**, *15*, x FOR PEER REVIEW 3 of 10

**Figure 1.** XRD patterns of Zr54.46Al9.9Ni4.95Cu29.7Pd0.99 BMGs before and after HRRF treatment [17]. **Figure 1.** XRD patterns of Zr54.46Al9.9Ni4.95Cu29.7Pd0.99 BMGs before and after HRRF treatment [17]. To gain a better understanding of heterogeneous microstructure for as-cast and

To gain a better understanding of heterogeneous microstructure for as-cast and treated Zr54.46Al9.9Ni4.95Cu29.7Pd0.99 BMGs, the distributions of the nano-hardness were obtained, as shown in Figure 2. It is obvious that the as-cast material is rather homogeneous, with hardness values ranging between 7.45 and 7.91 GPa. Conversely, the treated sample is much more heterogeneous and displays a wider range of hardness values (7.288–8.164 GPa). To a certain extent, these distributions demonstrated that the microstructure of treated BMGs can be considered more inhomogeneous than the as-cast samples. Confirmed by early studies, this phenomenon can be interpreted by Lennard-Jones-like potential function. During HRRF-treating, the drastic amount of mechanical work by the preset load was intruded into BMGs driving short-range atomic rearrangement [17,20], leading to hard regions with a higher atomic packing density, and soft regions with a To gain a better understanding of heterogeneous microstructure for as-cast and treated Zr54.46Al9.9Ni4.95Cu29.7Pd0.99 BMGs, the distributions of the nano-hardness were obtained, as shown in Figure 2. It is obvious that the as-cast material is rather homogeneous, with hardness values ranging between 7.45 and 7.91 GPa. Conversely, the treated sample is much more heterogeneous and displays a wider range of hardness values (7.288–8.164 GPa). To a certain extent, these distributions demonstrated that the microstructure of treated BMGs can be considered more inhomogeneous than the as-cast samples. Confirmed by early studies, this phenomenon can be interpreted by Lennard-Jones-like potential function. During HRRF-treating, the drastic amount of mechanical work by the preset load was intruded into BMGs driving short-range atomic rearrangement [17,20], leading to hard regions with a higher atomic packing density, and soft regions with a lower atomic packing density. treated Zr54.46Al9.9Ni4.95Cu29.7Pd0.99 BMGs, the distributions of the nano-hardness were obtained, as shown in Figure 2. It is obvious that the as-cast material is rather homogeneous, with hardness values ranging between 7.45 and 7.91 GPa. Conversely, the treated sample is much more heterogeneous and displays a wider range of hardness values (7.288–8.164 GPa). To a certain extent, these distributions demonstrated that the microstructure of treated BMGs can be considered more inhomogeneous than the as-cast samples. Confirmed by early studies, this phenomenon can be interpreted by Lennard-Jones-like potential function. During HRRF-treating, the drastic amount of mechanical work by the preset load was intruded into BMGs driving short-range atomic rearrangement [17,20], leading to hard regions with a higher atomic packing density, and soft regions with a lower atomic packing density.

(**a**) (**b**) **Figure 2.** (**a**,**b**) are the distribution of the nano-indentation hardness of the as-cast [17] and treated **Figure 2.** (**a**,**b**) are the distribution of the nano-indentation hardness of the as-cast [17] and treated sample, respectively. **Figure 2.** (**a**,**b**) are the distribution of the nano-indentation hardness of the as-cast [17] and treated sample, respectively.

### sample, respectively. *3.2. Roughness of Pre-Tested Surfaces 3.2. Roughness of Pre-Tested Surfaces*

*3.2. Roughness of Pre-Tested Surfaces*  It is well known that roughness of pre-tested surfaces would interfere with the friction properties of BMGs [21]. Therefore, before friction tests, AFM are used to evaluate pre-tested surfaces. The micro-morphology, pre-tested surfaces of the as-cast and treated BMGs are analyzed by AFM as illustrated in Figure 3a,b. It is intuitively found that all It is well known that roughness of pre-tested surfaces would interfere with the friction properties of BMGs [21]. Therefore, before friction tests, AFM are used to evaluate pre-tested surfaces. The micro-morphology, pre-tested surfaces of the as-cast and treated BMGs are analyzed by AFM as illustrated in Figure 3a,b. It is intuitively found that all It is well known that roughness of pre-tested surfaces would interfere with the friction properties of BMGs [21]. Therefore, before friction tests, AFM are used to evaluate pretested surfaces. The micro-morphology, pre-tested surfaces of the as-cast and treated BMGs are analyzed by AFM as illustrated in Figure 3a,b. It is intuitively found that all pre-tested surfaces exhibit the interlaced topography of "peak" and "valley" on the pre-tested surfaces. Significantly, the relative width of the pre-tested surface roughness for the as-cast glass

is 16.6 nm, much the same as that (18.4 nm) in the treated sample. Moreover, by further calculation, the average surface roughness of pre-tested surfaces (Ra) of the cast sample is 4.61 nm, and the Ra of the treated sample is 4.9 nm. Such tiny changes suggest that the pre-tested surfaces exhibit ideal flatness, minimizing the effect of the surface roughness of pre-tested surfaces on the accuracy of friction experiments. as-cast glass is 16.6 nm, much the same as that (18.4 nm) in the treated sample. Moreover, by further calculation, the average surface roughness of pre-tested surfaces (Ra) of the cast sample is 4.61 nm, and the Ra of the treated sample is 4.9 nm. Such tiny changes suggest that the pre-tested surfaces exhibit ideal flatness, minimizing the effect of the surface roughness of pre-tested surfaces on the accuracy of friction experiments.

*Materials* **2022**, *15*, x FOR PEER REVIEW 4 of 10

pre-tested surfaces exhibit the interlaced topography of "peak" and "valley" on the pretested surfaces. Significantly, the relative width of the pre-tested surface roughness for the

**Figure 3.** (**a**,**b**) are are high-magnification detailed micrographs of pre-tested surfaces for as-cast and treated samples to be tested under AFM. **Figure 3.** (**a**,**b**) are are high-magnification detailed micrographs of pre-tested surfaces for as-cast and treated samples to be tested under AFM.

### *3.3. Wear Performance 3.3. Wear Performance*

As shown in Figure 2, HRRF could enhance the microstructural heterogeneity of Zr54.46Al9.9Ni4.95Cu29.7Pd0.99 BMGs. Thus, the as-cast and treated BMGs were tested for wear resistance. Figure 4 presents the friction coefficient curves of as-cast and treated BMGs under dry conditions and 3.5% NaCl solution. As described in Figure 4a, it was found that both curves have a steady-state stage after an initial rapid increasing period under the dry-friction condition; the coefficients of friction in the steady-state stage are ~0.578 and ~0.676 for as-cast and treated samples respectively. The treated BMG displays a higher friction coefficient (COF). This phenomenon may be interpreted as that during the friction experiment, the harder regions in treated BMGs can remain intact for a long time, which increases the relative movement resistance between the material and the friction pair [22]. However, in Figure 4b, under the 3.5% NaCl solution condition, the friction coefficients of all samples exhibit around 0.29, and much lower, values than those in the air condition, owing to the lubrication effect of the NaCl solution. As shown in Figure 2, HRRF could enhance the microstructural heterogeneity of Zr54.46Al9.9Ni4.95Cu29.7Pd0.99 BMGs. Thus, the as-cast and treated BMGs were tested for wear resistance. Figure 4 presents the friction coefficient curves of as-cast and treated BMGs under dry conditions and 3.5% NaCl solution. As described in Figure 4a, it was found that both curves have a steady-state stage after an initial rapid increasing period under the dry-friction condition; the coefficients of friction in the steady-state stage are ~0.578 and ~0.676 for as-cast and treated samples respectively. The treated BMG displays a higher friction coefficient (COF). This phenomenon may be interpreted as that during the friction experiment, the harder regions in treated BMGs can remain intact for a long time, which increases the relative movement resistance between the material and the friction pair [22]. However, in Figure 4b, under the 3.5% NaCl solution condition, the friction coefficients of all samples exhibit around 0.29, and much lower, values than those in the air condition, owing to the lubrication effect of the NaCl solution. *Materials* **2022**, *15*, x FOR PEER REVIEW 5 of 10

**Figure 4.** The friction coefficient of as-cast and treated Zr54.46Al9.9Ni4.95Cu29.7Pd0.99 BMGs as a function of sliding time under (**a**) dry and (**b**) 3.5% NaCl solution. **Figure 4.** The friction coefficient of as-cast and treated Zr54.46Al9.9Ni4.95Cu29.7Pd0.99 BMGs as a function of sliding time under (**a**) dry and (**b**) 3.5% NaCl solution.

To be specific, under the dry conditions, due to the direct contact friction between the friction pair of metals, the friction-induced heat easily appears on the tested surfaces The wear rate is always regarded as an important parameter to evaluate the wear resistance of the materials [6,23]. Thus, the wear rate of as-cast and treated BMGs under dry and

of samples [24]. With the heat accumulated constantly, the worn surface not only becomes soft, but also reacts with air to form oxide layers. The hardness of the oxide layers is higher

due to the high hardness and limited ductility of the oxide layer, it is easy to induce the separation of the oxide layer from the BMGs matrix under the action of sliding shear force, which generates oxide layers which can wear out more easily than BMGs [25,26]. BMGs treated by HRRF would show a supernal microstructural inhomogeneity, which indicates that the hard regions become stronger and the soft regions become more fragile compared with the as-cast BMGs. Hence, for treated BMGs, because of the higher-density arrangement of atoms, it is difficult to oxidize the hard regions. As a result, the number of oxide layers generated on the worn surface of the treated sample may be less than those of the as-cast sample. Eventually, the treated BMGs exhibit higher wear resistance. Similarly, under 3.5% NaCl solution, although there are only a few oxide layers on the worn surfaces of BMGs for a good cooling effect of the solution, due to harder regions, the treated BMGs still show a better wear resistance ability. All results evidence that enhancing the microstructural inhomogeneity of BMGs can improve the wear resistance effectively in these

**Figure 5.** The wear rate of as−cast and treated Zr54.46Al9.9Ni4.95Cu29.7Pd0.99 BMGs in different environ-

two external environments.

ments.

3.5% NaCl solution are illustrated in Figure 5. The wear rates of as-cast BMG and treated BMG in dry are 22.1 <sup>×</sup> <sup>10</sup>−<sup>6</sup> mm<sup>3</sup> ·N−1m−<sup>1</sup> and 14.7 <sup>×</sup> <sup>10</sup>−<sup>6</sup> mm<sup>3</sup> ·N−1m−<sup>1</sup> , respectively. In addition, the wear rates of BMGs sliding in 3.5% NaCl solution decreased in the following order: 4.5 <sup>×</sup> <sup>10</sup>−<sup>6</sup> mm<sup>3</sup> ·N−1m−<sup>1</sup> for the as-cast BMG, and 3 <sup>×</sup> <sup>10</sup>−<sup>6</sup> mm<sup>3</sup> ·N−1m−<sup>1</sup> for the treated BMG. Evidently, the values of wear rates of as-cast samples show remarkably higher values than those of the treated samples under different conditions. as-cast sample. Eventually, the treated BMGs exhibit higher wear resistance. Similarly, under 3.5% NaCl solution, although there are only a few oxide layers on the worn surfaces of BMGs for a good cooling effect of the solution, due to harder regions, the treated BMGs still show a better wear resistance ability. All results evidence that enhancing the microstructural inhomogeneity of BMGs can improve the wear resistance effectively in these two external environments.

**Figure 4.** The friction coefficient of as-cast and treated Zr54.46Al9.9Ni4.95Cu29.7Pd0.99 BMGs as a function

To be specific, under the dry conditions, due to the direct contact friction between the friction pair of metals, the friction-induced heat easily appears on the tested surfaces of samples [24]. With the heat accumulated constantly, the worn surface not only becomes soft, but also reacts with air to form oxide layers. The hardness of the oxide layers is higher than that of the BMGs, thereby improving the wear resistance of the material. However, due to the high hardness and limited ductility of the oxide layer, it is easy to induce the separation of the oxide layer from the BMGs matrix under the action of sliding shear force, which generates oxide layers which can wear out more easily than BMGs [25,26]. BMGs treated by HRRF would show a supernal microstructural inhomogeneity, which indicates that the hard regions become stronger and the soft regions become more fragile compared with the as-cast BMGs. Hence, for treated BMGs, because of the higher-density arrangement of atoms, it is difficult to oxidize the hard regions. As a result, the number of oxide layers generated on the worn surface of the treated sample may be less than those of the

*Materials* **2022**, *15*, x FOR PEER REVIEW 5 of 10

of sliding time under (**a**) dry and (**b**) 3.5% NaCl solution.

**Figure 5.** The wear rate of as−cast and treated Zr54.46Al9.9Ni4.95Cu29.7Pd0.99 BMGs in different environments. **Figure 5.** The wear rate of as-cast and treated Zr54.46Al9.9Ni4.95Cu29.7Pd0.99 BMGs in different environments.

To be specific, under the dry conditions, due to the direct contact friction between the friction pair of metals, the friction-induced heat easily appears on the tested surfaces of samples [24]. With the heat accumulated constantly, the worn surface not only becomes soft, but also reacts with air to form oxide layers. The hardness of the oxide layers is higher than that of the BMGs, thereby improving the wear resistance of the material. However, due to the high hardness and limited ductility of the oxide layer, it is easy to induce the separation of the oxide layer from the BMGs matrix under the action of sliding shear force, which generates oxide layers which can wear out more easily than BMGs [25,26]. BMGs treated by HRRF would show a supernal microstructural inhomogeneity, which indicates that the hard regions become stronger and the soft regions become more fragile compared with the as-cast BMGs. Hence, for treated BMGs, because of the higher-density arrangement of atoms, it is difficult to oxidize the hard regions. As a result, the number of oxide layers generated on the worn surface of the treated sample may be less than those of the as-cast sample. Eventually, the treated BMGs exhibit higher wear resistance. Similarly, under 3.5% NaCl solution, although there are only a few oxide layers on the worn surfaces of BMGs for a good cooling effect of the solution, due to harder regions, the treated BMGs still show a better wear resistance ability. All results evidence that enhancing the microstructural inhomogeneity of BMGs can improve the wear resistance effectively in these two external environments.
