Thermophysical Properties of the Fe48Cr15Mo14C15B6Y2 Alloy in Liquid State

In this work, the physical properties of Fe48Cr15Mo14C15B6Y2 alloy in liquid state at high temperature are studied. It was observed that the basic physical characteristics of the alloy, such as viscosity, electrical resistivity, and density, decrease with an increase of the temperature to 1700 °C. An abnormal increasing rate of viscosity for Fe48Cr15Mo14C15B6Y2 alloy in the temperature range from 1360 to 1550 °C was noted. The measurement of the electrical resistivity and density did not reveal any anomalies in the same temperature range.


Introduction
Interest in the synthesis and mechanical characterisation of Fe-based bulk amorphous alloys, also known as bulk metallic glasses (BMGs), started in the early 1990s [1][2][3]. These amorphous alloys have been shown to have a number of enhanced properties, such as superior mechanical properties and improved corrosion resistance [4][5][6][7]. Moreover, the soft magnetic properties of the Fe-based alloy led to widespread use of the material. An Fe-based amorphous alloy that has attracted a lot of attention in recent years is Fe 48 Cr 15 Mo 14 C 15 B 6 Y 2 , with several recent studies delving into its structural, mechanical, physical, and chemical properties [8][9][10]. However, the glass-forming ability (GFA) of Febased BMGs is the main limiting factor that has prevented their extensive use as structural and functional materials.
In the last few decades, studies on Fe-based BMGs have shifted their focus to their liquid state (metallic melts) [11,12]. This is due to the fact that the dynamical processes in liquids and how liquids transform to glasses are major outstanding questions in condensed matter science [13][14][15]. Furthermore, such information is essential to understand the glassforming ability of the alloys containing rare elements and boron [16][17][18][19][20][21]. One typical example is the strong influence that was observed between the casting temperature and the critical sample diameter for the Fe 50 Cr 15 Mo 14 C 15 B 6 alloy [22].
One of the temperature-sensitive parameters of such alloys in the liquid state is viscosity. This characteristic is crucial in revealing changes in the liquid state as the sample is heated or cooled. Nowadays, various techniques are applied to measure the viscosity of metallic melts, including parallel plate rheometry [23], tensile creep [24] penetration viscometer [25,26], and the bending beam test [27][28][29].
There are other methods for studying the behaviour of metallic melts, which include the measurement of density, surface tension, and electrical resistivity in the liquid state across a range of temperatures. These methods can give a general idea on how physical properties of the metallic melt change at various temperatures. However, it should be noted that the physical properties of metallic melts at high temperatures are significantly dependent on the chemical composition [22].
In recent years, various mathematical models have been developed to predict the properties of melts. However, it has been shown that such models are not able to predict the properties of multicomponent melts with sufficient accuracy [14]. Thus, the development of new alloys and technologies is still dependent on experimental data. In light of this, the main aim of the present research is to study the physical properties of the Fe 48 Cr 15 Mo 14 C 15 B 6 Y 2 alloy (viscosity, density, and electrical resistance) in a liquid state and to reveal changes in the physical properties during heating and cooling cycles.
As a result we obtained bulk amorphous-crystalline alloy and provided its viscosity measurements in the liquid state.

Alloy Preparation and Structure Characterisation of Samples
Fe 48 Cr 15 Mo 14 C 15 B 6 Y 2 (atomic % compositions) alloy ingots were prepared by inductionmelting a mixture of the commercial-grade pure elements in the required proportions (minimum purity of each element used was 99.9 wt.%). Then, the ingots were melted again and injected through a nozzle into a copper mould to produce ingots with a diameter of 12 mm. It should be noted that to achieve volumetric amorphization of the Fe 48 Cr 15 Mo 14 C 15 B 6 Y 2 alloy, the ingot diameter should not exceed 7 mm [30]. However, in this study, the ingot diameter for this alloy was deliberately set to 12 mm, in order to obtain ingots having both crystalline and amorphous phases.
The samples were first polished and then etched by using 25% aqueous HNO 3 . The resulting microstructure was then observed by a scanning transmission electron microscope (Tescan Vega 3 LMH, TESCAN ORSAY HOLDING, Brno, Czech Republic) operated at 20 kV, in back-scattered electron mode.
A Vicker's test (Thixomet SmartDrive MHT, Thixomet, St.-Petersburg, Russia) was also performed using a pyramidal diamond indenter to measure the microhardness of individual phases in the structure of the alloy under a load of 0.98 N. An X-ray diffraction (XRD) analysis was also carried out by using Bruker AXS GmbH (Bruker, Billerica, MA, USA) apparatus with a monochromatic CuKa radiation at the accelerating voltage of 30 kV and a current of 30 mA. The percentage of the amorphous phase was determined by using a kinematic standard method [31] and annealed nickel powder as a standard.

Differential Scanning Calorimetry Measurement
The thermal properties of the glassy alloys were measured using a Setaram Labsys differential scanning calorimetry (DSC, Setaram Instrumentation, Caluire-et-Cuire, France) instrument under an Ar gas flow, in Al 2 O 3 crucible. The measured temperatures ranged from room temperature to 1340 • C with the heating rate 20 • C/min. The mass of the samples was 10-15 mg.

Measurement of the Kinematic Viscosity in Liquid State
Kinematic viscosity was determined by analysing the damped torsional oscillations of a BeO vessel (with an inner diameter of 13 mm) containing the metallic melt during heating (up to 1700 • C) and during subsequent slow cooling. A working chamber was initially evacuated to 0.001 Pa, and then it was filled with helium to a pressure of 10 5 Pa. The measurements were performed during isothermal holding (for at least 5 min), with stepwise changes in temperature of 20-30 • C. The temperature was maintained constant with an accuracy of 1 • C using a precision controller (optical pyrometer). During the measurements, oscillation parameters were monitored optically, using an automatic system for photo-recording of oscillation. The measurement and experimental data treatment procedures are described in [32,33].

Measurement of the Electrical Resistivity in Liquid State
The electrical resistivity of the Fe 48 Cr 15 Mo 14 C 15 B 6 Y 2 liquid was measured by using the rotating magnetic field method during heating (up to 1700 • C) and subsequent cooling. A working chamber was initially evacuated to 0.001 Pa, and then filled with helium to a pressure of 10 5 Pa. Samples were held in this chamber for 5-8 min at the melting temperature and then heated up to 1700 • C in increments of 30-50 • C. Isothermal holding at the measurement points was not less than 10 min. The electrical resistivity was measured as described in [34,35].

Measurement of the Density in Liquid State
The density of liquid metal upon heating up to 1760 • C and subsequent cooling was measured by the sessile drop method [20]. The test sample of a given mass and volume (about 1 cm 3 ) was placed on a horizontal fireproof backplate made of BeO. This was placed into an electric furnace. The working chamber was preliminarily evacuated to a pressure of 0.001 Pa. Helium was then pumped in until a pressure of 10 5 Pa was achieved. This was done to protect the sample from air contamination and boiling up. The sample was heated until an ellipse-shaped melt drop resting on the horizontal backplate was formed. Isothermal holding at the measurement points was at least 5 min. The samples were held in the chamber with an inert atmosphere at the melting temperature for 5-8 min and were then heated up to 1800 • C in increments of 30-50 • C. The drop profile was recorded with a digital camera. The drop images were then transferred to a computer. In order to calculate the drop's volume, the geometric parameters of the drop's contours were measured from the images obtained during the experiment using the SIAMS 700 image analysis software package (SIAMS, Yekaterinburg, Russia). The density rate for the liquid state was obtained as described in [36]. The absolute error of determining the melt density did not exceed 5%. A detailed methodology of the procedure performed is described in [37].

Results and Discussion
In this study, Fe 48 Cr 15 Mo 14 C 15 B 6 Y 2 alloys with both amorphous and crystalline phases were produced. This was confirmed through XRD analysis, SEM imaging, and microhardness measurements. Figure 1 shows the SEM images of the middle area of an alloy ingot. Several inclusions, in the form of dendric structures (see Figure 1a) and microinclusions (see Figure 1b,c), were found within the sample. Through microhardness analysis, it was determined that the dendritic matrices were composed of ledeburite, having microhardness values of about 1300-1350 HV, whilst the microinclusios were composed of complex carbides containing Mo and Fe, having a microhardness of about 2100-2300 HV. The microhardness result, obtained by investigation of ledeburite and carbides, is very close to literature data for cast iron, steel alloy, and amorphous films [38][39][40].
X-Ray analysis of the as-cast sample confirmed that the sample is in an amorphouscrystalline state (see the amorphous halo in Figure 1d). The determined crystalline volume fraction reached approximately 30%. Moreover, analysis shows that carbides structures are similar to the Fe 3 Mo 3 C composition. It should also be noted that the occurrence of an amorphous and crystalline phase in the as-cast sample was confirmed by the SEM analysis ( Figure 2). It is observed that the sample has amorphous and crystalline phases (Figure 2a,b). Crystallite particles are distribute in an amorphous matrix and have a dendrite structure (Figure 2c,d). It should also be noted that the obtained SEM results are in a good agreement with the XRD analysis (Figure 1d).
A number of physical properties of the alloy in the liquid state were also successfully measured and are discussed below. During the measurements of these physical properties, no evidence of melt evaporation or loss in mass of the sample was observed.

Kinematic Viscosity
From the literature, it is known that for homogeneous melts, the viscous flow decreases with increasing temperature [41]. However, micro-inhomogeneities transferred from solid state may be present in metallic melts. These micro-inhomogeneities are specific for each melt and can persist for a long time [41]. The presence of micro-inhomogeneities can be evaluated from the heating and cooling curves for some structure-sensitive property of the melt (most frequently from its viscosity). By heating above the temperature of homogeneity, the melt irreversibly converts to a true solution state, which produces significant changes in the metal solidification conditions [41].
In this study, we measured the temperature-dependent kinematic viscosity of the Fe 48 Cr 15 Mo 14 C 15 B 6 Y 2 alloy, whose cast structure consisted of two phases: a crystalline phase and an amorphous one as discussed above. An increase in viscosity for the Fe 48 Cr 15 Mo 14 C 15 B 6 Y 2 alloy was observed in the temperature range of 1360 to 1550 • C (Figure 3a). This unusual increase in viscosity may be attributed to the fragmentation of the dendritic microstructure of the alloy and the formation of fragments in the volume of the BeO crucible. Another factor that may be responsible for such an increase in viscosity is the transformation of the amorphous phase to a crystalline one. When the temperature exceeded 1550 • C we observed a sudden increase in the viscosity (Figure 3a), which is characteristic of the transition from a solid to a liquid state. The viscosity was then observed to decrease with increasing temperature within the temperature range of 1570 to 1650 • C (Figure 3a). After homogenization at 1650 • C, the viscosity of the metallic melt increased from 1510 to 1450 • C (Figure 3a). When the temperature of the sample was below its melting point of 1310 • C, the melt began to transition from the liquid state to a crystalline phase, avoiding amorphization.
The temperature-dependent kinematic viscosity of the Fe 48 Cr 15 Mo 14 C 15 B 6 Y 2 sample after a repeated heating-cooling cycle is shown in Figure 3b. In this case, the temperature of homogenization was achieved during the first heating cycle. This was followed by slow cooling to the temperature of crystallization. Thus, in this instance, one expects no significant differences between subsequent heating and cooling cycles. Nonetheless, such measurements can provide additional information on the state of the liquid phase at high temperatures. As shown in Figure 3b, the liquid state in this heating cycle was achieved at 1380 • C. Increasing the temperature to 1570 • C led to a decrease in the melt viscosity. Heating above this temperature resulted in a sudden increase in the viscous flow of the melt up to a temperature of 1590 • C, after which the viscosity started to decrease again (Figure 3b). Above 1300 • C, the alloy is in a liquid state. Equilibrium melt obeys the Arrhenius equation, i.e., the viscosity decreases with increasing temperature. At the stage of the first heating to 1560 • C, the melt is in a nonequilibrium state. Low viscosity corresponds to the flow of large clusters. An almost linear slight increase in viscosity at the stage of the first heating to 1560 • C corresponds to a slow detachment of atoms from clusters, i.e., reducing their average size. A sharp increase in viscosity corresponds to the decomposition of clusters and the transition of the melt to an equilibrium state. This state confirms the typical dependence of viscosity on temperature. Figure 4 is showing the DSC curves of the as-cast sample. It is clearly seen that the crystallization and phase transformation processes started at about 600 • C and the melting process has started at above 1100 • C. Unfortunately, the characteristic property of DSC equipment did not allow us to heat up the sample above the 1400 • C, therefore it was impossible to obtain the melting of the whole volume of it. However, several phases were partly melted (approximately, with two different composition). On the cooling DSC curves (Figure 4), the crystallization of two phases (two crystallization peaks) is also clearly seen.  The microstructure of the Fe 48 Cr 15 Mo 14 C 15 B 6 Y 2 alloy, which was obtained after the second heating/cooling cycle, is shown in Figure 5a. This micrograph clearly shows that the alloy melt has recrystallized into a fine dispersed phase. Microhardness measurements indicated the presence of ledeburite (1490-1550 HV) and separated inclusions of complex carbides (M 23 C 6 and M 6 C) (2200-2500 HV). Furthermore, X-ray diffraction of the sample revealed the presence of two phases, i.e., Cr 21.34 Fe 1.66 C 6 and Fe 3 Mo 3 C (Figure 5b). Thus, we can conclude that the cast structure of alloy after recrystallization from the liquid state corresponded to the structure of the high-alloy chilled cast iron.

Electrical Resistivity
The electrical resistivity of the Fe 48 Cr 15 Mo 14 C 15 B 6 Y 2 alloy produced in this study was measured within the temperature range from 1200 to 1700 • C. The electrical resistivity was observed to decrease with increasing temperature, as shown in Figure 6. No obvious deviations were observed on the heating/cooling polytherms at 1580 and 1600 • C, the temperature ranges where an increase/decrease in kinematic viscosity was observed for heating and cooling, respectively. This indicates that electrical resistivity methods lack sensitivity when observing the changes occurring in the liquid state at these temperatures. The decrease in electrical resistivity with increasing temperature may be explained through the microheterogeneous structure of metallic melts [32][33][34][35][36][37][38][39][40][41][42][43][44][45][46][47]. Melting of a multiphase cast iron ingot does not lead to the immediate formation of a homogeneous solution of the alloying elements in iron, with a microheterogeneous state persisting within a certain temperature range. The temperature corresponding to the irreversible transition of the melt into a homogeneous state may be determined from the high-temperature coinciding portion of the heating and cooling polytherms. This was found to be close to 1700 • C ( Figure 6) for the Fe 48 Cr 15 Mo 14 C 15 B 6 Y 2 alloy produced in this study when using electrical resistivity measurements.  Figure 7 shows how the density and shape of a drop of the Fe 48 Cr 15 Mo 14 C 15 B 6 Y 2 alloy produced in this study change when the temperature is increased up to 1800 • C. During the first heating cycle, the density of the sample changed slightly with heating (7.03-7.07 g/cm 3 ). However, during the first cooling cycle, larger changes in the density were recorded (Figure 7a). This may be explained by the fact that when the temperature of homogenization is achieved, the alloy drop is cleaned from any film that may be present on its surface.

Density
The heredity of the cast state for the examined sample was destroyed at 1800 • C and the behaviour of the liquid state upon cooling from 1800 to 1260 • C corresponded to the formation of another type of alloy, such as a high-alloy chilled cast iron. This means that the nucleation and growth of the crystalline phase, mainly the carbide phases, upon crystallization are accompanied with a decrease in volume of the sessile drop in the liquid state. As a result of this transition, the density of the metallic melt is increased (Figure 7a). The density of the sessile drop at 1260 • C was 7.43-7.48 g/cm 3 . The repeated measurement of density for this sessile drop showed that the heating and cooling curves were coincident in the wide range of temperature from 1310 to 1700 • C (Figure 7b). This confirms that the transitions in the liquid phase do not have a significant effect on the density after the process of recrystallization. However, some oscillations in density were observed in the heating curve from 1260 to 1310 • C (Figure 7b). These oscillations were due to the destruction of the film on the surface of the sessile drop (Figure 7b). As a general observation, one may conclude that the density of the Fe 48 Cr 15 Mo 14 C 15 B 6 Y 2 alloy decreases with increasing temperature.

Conclusions
The thermophysical properties (viscosity, electrical resistance, and density) of a Fe 48 Cr 15 Mo 14 C 15 B 6 Y 2 bulk amorphous-crystalline alloy at high temperatures were studied. Viscosity measurements gave the most complete picture of the processes taking place in the liquid state of the alloy studied. In fact, only viscosity measurements showed a possible change in the structure of the liquid within the temperature range of 1580 and 1600 • C, whereby the viscous flow of metallic melts was extremely increased. Furthermore, it was confirmed that the density and electrical resistivity of Fe 48 Cr 15 Mo 14 C 15 B 6 Y 2 alloy in the liquid state tend to decrease with increasing temperature.

Data Availability Statement:
The data presented in this study are available on request from the corresponding author.

Conflicts of Interest:
The authors declare no conflict of interest.