Analysis of the fragility of the ${\mathrm{Zr}}_{77}{\mathrm{Ni}}_{23}$ metallic glass based on low-temperature heat capacity measurements

We present an experimental investigation of the fragility and heat capacity of metallic glass ${\mathrm{Zr}}_{77}{\mathrm{Ni}}_{23}$. The ribbon sample was produced by melt-spinning technique. Glass transition temperature $T_g$ obtained by differential scanning calorimetry with various heating rates was used to estimate fragility parameter $m$. Heat capacity measurements were performed in a wide temperature interval, ranging from 1.8 K up to room temperature, for as-cast and relaxed samples in different magnetic field strengths. Our results do not show any excess of vibrational density of states over the Debye contribution corresponding to the boson peak. Relaxation of the sample causes a slight decrease of Debye contribution consistent with the measured increase of Young modulus. The fact that no boson peak is observed in heat capacity, together with the obtained intermediate fragility of $m=53$, positions ${\mathrm{Zr}}_{77}{\mathrm{Ni}}_{23}$ well outside established correlations between fragility, boson peak strength, and glass forming ability.

Figure 1

XRD pattern of ${\mathrm{Zr}}_{77}{\mathrm{Ni}}_{23}$ MG.

Figure 2

$C_p$ of ${\mathrm{Zr}}_{77}{\mathrm{Ni}}_{23}$ as-cast and relaxed samples in zero magnetic field. The inset shows a $C_p/T^3$ vs $T$ plot for as-cast and relaxed samples.

Figure 3

Heat capacity in the range of SC transition for various magnetic fields. The inset shows the dependence of $T_c$ on the magnetic field.

Figure 4

The ${\displaystyle C_p/T}$ vs $T^2$ plots for (a) as-cast and relaxed samples without magnetic field, (b) as-cast sample with field $H=0$ T and $H=4$ T, and (c) relaxed sample with $H=0$ T and $H=4$ T.

Figure 5

Heat capacity parameters $\beta$ and $\gamma$ for as-cast and relaxed ${\mathrm{Zr}}_{77}{\mathrm{Ni}}_{23}$ samples for different magnetic fields.

Figure 6

$\mathrm{\Delta }{C}_p$ after subtracting the electronic contribution shows pure $T^3$ behavior, but it demonstrates also the effect of relaxing. The inset shows a ${\displaystyle (C_p-\gamma T)/\beta T^3}$ vs ${\displaystyle T/{\theta }_D}$ plot.

Figure 7

Inverse heating rate against $1/T$ normalized to $T_g^{\prime }$ at heating rate 1 K/min. All results except for ${\mathrm{Zr}}_{77}{\mathrm{Ni}}_{23}$ are taken from [43]. Error bar is indicated for measurement at heating rate of 30 K/min. Inset figure shows DSC measurement for heating rate 3 K/min.