Fig. 1. (a) Illustrations of three different compression simulations: pISE, UC-QIC, and NEMD-QIC. (b) and (c) Thermodynamic paths undergone by Cu₅₀Zr₅₀ metallic glass subjected to pISE, UC-QIC, and NEMD-QIC at an equivalent strain rate of ∼7.5×108/s.

Fig. 2. Illustration of the reference isentrope, instant isentrope, and quasi-isentrope without phase transition. States A and C have the same density, and states B, D, and E have the same density.

Fig. 3. Illustration of the framework integrating thermodynamics, mechanics, and microstructural analysis.

Fig. 4. (a) Shear modulus mapped into pressure and density. (b) Grüneisen parameter vs density at 300, 400, 500, and 600 K. (c) Vibrational entropy mapped into pressure and density. (d) Isochoric heat capacity vs nonaffine atomic strain at different densities and temperatures.

Fig. 5. (a) Isentrope recovered from P–ρ–T path. (b) Excess temperature increase ΔT_excess vs density. (c) Strength vs density during pISE, UC-QIC, and NEMD-QIC with an equivalent strain rate of ∼7.5×108/s. The light red box represents the elastic deformation regime for NEMD-QIC and roughly for UC-QIC. Gray dots indicate where ΔT_excess starts to increase.

Fig. 6. (a) Excess temperature ΔT_excess at ρ/ρ₀ = 1.29 and scaled strength vs rise time t_rise. A separation between rate-insensitive and rate-sensitive regime is revealed. (b) Taylor–Quiney factor β_int vs plastic strain during NEMD-QIC. (c) Replica distance and fraction of extra cage breakage at yielding vs t_rise. The separation between rate-insensitive and rate-sensitive regime is the same as in (a). (d) A cross-section colored by dX,Yi in a yield replica of the case t_rise = 67 ps.

Fig. 7. (a) Total entropy increase ΔS_tot, configurational entropy increase ΔS_conf, and vibrational entropy increase ΔS_vib vs relative density during NEMD-QIC. (b) Entropy increases at ρ/ρ₀ = 1.29 vs rise time. The entropy increases for shock compression (t_rise = 0 ps) are also shown.

Fig. 8. (a) Local fivefold symmetry (LFFS) vs relative density under compression. (b) Shannon entropy defined on a copper-centered polyhedron distribution vs relative density. Only compressions in the rate-insensitive regime are shown.

Fig. 9. (a) Configurational entropy vs Shannon entropy defined on copper-centered polyhedron distribution. (b) Taylor–Quinney factor (TQF) vs configurational entropy change during plastic regime for NEMD-QIC. Only compressions in the rate-insensitive regime are shown. Black solid lines are fitting curves.

Fig. 10. (a) Configurational energy vs plastic strain. (b) Effective temperature and temperature vs plastic strain. Only compressions in the rate-insensitive regime are shown.

Table 1. Fitting parameters of thermophysical properties of the model Cu₅₀Zr₅₀. The equation is $H(\rho ,T,{\gamma }_a)=h_0[1+h_1\rho /{\rho }_0+h_2{(\rho /{\rho }_0)}^2](1+h_{3}T+h_4{T}^2)(1+h_5{\gamma }_a+h_6{\gamma }_a^2)$, in which ρ is in units of kg/m³, T is in units of K, and γ_a is dimensionless. C_V is the isochoric heat capacity, γ_g the Grüneisen parameter, B the bulk modulus, G the shear modulus, S_vib the vibrational entropy, and F the Helmholtz free energy containing only a vibrational contribution.