High-pressure hydrogen by machine learning and quantum Monte Carlo

Thermodynamic Modeling of Fluid Polyamorphism in Hydrogen at Extreme Conditions

“…Experiments and simulations have discovered that at extremely high pressures, highly-dense fluid (dimeric) hydrogen dissociates into atomistic fluid hydrogen 1,[8][9][10][36][37][38][39][40][41][42][43][44][45][46][47][48] . Using the generalized law of corresponding states, by reducing the temperature, pressure, and entropy by their critical values, we combine the available experimental data with the results of computations 1,39,40,[43][44][45][46][47][48][49] to predict the equation of state of hydrogen near the fluid-fluid phase transition (FFPT). We show predictions for the phase coexistence and the reaction equilibrium of the two alternative states of fluid hydrogen.…”

“…While the predictions of Cheng et al for the FFPT are not in agreement with the results of all other simulations and experimental studies 48 , their study provides a reasonable idea for how the non-ideality parameter, ω, might depend on pressure and temperature. Based on the suggested trend, we optimized ω T and ω P to agree with the behavior of hydrogen from the available computational data 1,[43][44][45][46]49 , and consequently, adopted these parameters as ω T = 2.062 and ω P = −0.175. The asymmetric Gibbs energy of mixing is illustrated in Fig.…”

“…If the Gibbs energy of mixing, G mix , would be symmetric with respect to x, then G BA = −µ mix = 0, could describe the conditions for both reaction equilibrium and fluid-fluid phase equilibrium 4 . However, since the monatomic and diatomic mixing is asymmetric, the condition for the balance of phase 39 (green), Perlioni et al 43 (dark green), Mozzola et al 44 (lime green), and Tirelli et al 49 (yellow green). The colored stars correspond to the reported (or adopted in this work) critical point for each data set.…”

“…(10) Using the most recent QMC simulations presented in Tirelli et al 49 , ρ A is estimated by Eq. ( 10) with coefficients: ρ 0 = 0.2, ρ 1 = 0.45, ρ 2 = 0.45, ρ 3 = 0.17, and ρ 4 = 0.1.…”

“…The computational results are indicated by the triangles: blue tints correspond to the Density Functional Theory (DFT) simulations of Bonev et al 75 (dark blue), Morales et al 1 (blue), Hinz et al 46 (sky blue), and Karasiev et al 48 (light blue). Meanwhile, green tints correspond to the Quantum Monte Carlo (QMC) simulations of Morales et al 1 (dark sea green), Lorenzen et al39 (green), Perlioni et al43 (dark green), Mozzola et al44 (lime green), and Tirelli et al49 (yellow green). The colored stars correspond to the reported (or adopted in this work) critical point for each data set.…”